Woolhouse - Metaphysics and Philosophy of Science 1988

METAPHYSICS AND PHILOSOPHY OF SCIENCE IN THE
SEVENTEENTH AND EIGHTEENTH CENTURIES
THE UNIVERSITY OF WESTERN ONTARIO

SERIES IN PHILOSOPHY OF SCIENCE
A SERIES OF BOOKS
IN PHILOSOPHY OF SCIENCE, METHODOLOGY,
EPISTEMOLOGY, LOGIC, HISTORY OF SCIENCE,
AND RELATED FIELDS
Managing Editor
ROBERT E. BUTTS
Dept. of Philosophy, University of Western Ontario, Canada

Editorial Board
JEFFREY BUB, University of Western Ontario
L. JONATHAN COHEN, Queen's College, Oxford
WILLIAM DEMOPOULOS, University of Western Ontario
WILLIAM HARPER, University of Western Ontario
JAAKKO HINTIKKA, Florida State University, Tallahassee
CLIFFORD A. HOOKER, University of Newcastle
HENRY E. KYBURG, JR., University of Rochester
AUSONIO MARRAS, University of Western Ontario
rORGEN MITTELSTRASS, Universitiit Konstanz
JOHN M. NICHOLAS, University of Western Ontario
BAS C. VAN FRAASSEN,
Princeton University
VOLUME 43
METAPHYSICS AND
PHILOSOPHY OF
SCIENCE IN
THE SEVENTEENTH
AND EIGHTEENTH
CENTURIES
Essays in honour of Gerd Buchdahl
R. S. WOOLHOUSE
Reader in Philosophy, University of York
KLUWER ACADEMIC PUBLISHERS

DORDRECHT / BOSTON / LONDON
Library of Congress Cataloging-in-Publication Data
MetaphysIcs and phIlosophy of scIence In the seventeenth and

elghteenth centurles : essays for Gerd Buchdahl / edIted by A.S.
Woolhouse.
p.
cm. -- (The Unlverslty of Western Ontarl0 serIes ln
phllosophy of scIence: v. 43)
ISBN-13: 978-94-010-7846-7
1. Metaphyslcs--Hlstory--17th century. 2. Metaphyslcs-Hlstory--18th century. 3. Sclence--Phllosophy--Hlstory--17th

century. 4. SClence--Phllosophy--Hlstory--18th century.
5. Buchdahl. Gerd.
I. Buchdahl. Gerd. II. Woolhouse. R. S.
III. Serles.
BD111.M56 1988
110' .9'032--dc19
88-8826
CIP
Published by Kluwer Academic Publishers,

P.O. Box 17, 3300 AA Dordrecht, The Netherlands.
Kluwer Academic Publishers incorporates
the publishing programmes of
D. Reidel, Martinus Nijhoff, Dr W. Junk and MTP Press.
Sold and distributed in the U.S.A. and Canada
by Kluwer Academic Publishers,
101 Philip Drive, Norwell, MA 02061, U.S.A.
In all other countries, sold and distributed
by Kluwer Academic Publishers Group,
P.O. Box 322, 3300 AH Dordrecht, The Netherlands.
All Rights Reserved

1988 by Kluwer Academic Publishers
Softcover reprint of the hardcover 1st edition 1988
No part of the material protected by this copyright notice may be reproduced or
utilized in any form or by any means, electronic or mechanical
including photocopying, recording or by any information storage and
retrieval system, without written permission from the copyright owner
IPhotograph by RSWI
GERD BUCHDAHL
(July 1987)
CONTENTS
PREFACE
Gerd Buchdahl: Biographical and Bibliographical

ROBERT E. BUTTS / Gerd Buchdahl: A Tribute
JORGEN MITTELSTRASS / Nature and Science in the Renaissance
RICHARD S. WESTFALL / Galileo and the Jesuits
WILLIAM R. SHEA / Descartes and the Rosicrucian Enlightenment
STEPHEN W. GAUKROGER / Descartes' Conception of
Inference
GEORGE MACDONALD ROSS / The Demarcation between
Metaphysics and Other Disciplines in the Thought of
Leibniz
R0GER s. WOOLHOUSE / Leibniz and Occasionalism
MARY B. HESSE / Vico's Heroic Metaphor
PETER M. HARMAN / Dynamics and Intelligibility: Bernoulli
and MacLaurin
CATHERINE WILSON / Sensible and Intelligible Worlds in
Leibniz and Kant
OLIVER N. H. LEAMAN / Transcendental Reasoning and the
Indeterminacy of the Human Point of View
MICHAEL K. POWER / Buchdahl and Rorty on Kant and the
History of Philosophy
DWIGHT C. BARNABY / The Early Reception of Kant's
Metaphysical Foundations of Natural Science
JOHN G. McEVOY / The Enlightenment and the Chemical
Revolution
NICHOLAS JARDINE / The Significance of Schelling's "Epoch
of a Wholly New Natural History": An Essay on the
Realization of Questions
vii
ix
1
9
17
45
73
101
133
165
185
213
227
245
265
281
307
327
Vlll
CONTENTS
NOTES ON THE CONTRIBUTORS

INDEX
351
355
PREFACE
The essays in this collection have been written for Gerd Buchdahl, by
colleagues, students and friends, and are self-standing pieces of original
research which have as their main concern the metaphysics and
philosophy of science of the seventeenth and eighteenth centuries. They
focus on issues about the development of philosophical and scientific
thought which are raised by or in the work of such as Bernoulli,
Descartes, Galileo, Kant, Leibniz, Maclaurin, Priestly, Schelling, Vico.
Apart from the initial bio-bibliographical piece and those by Robert
Butts and Michael Power, they do not discuss Buchdahl or his ideas in
any systematic, lengthy, or detailed way. But they are collected under a
title which alludes to the book, Metaphysics and the Philosophy of
Science: The Classical Origins, Descartes to Kant (1969), which is
central in the corpus of his work, and deal with the period and some of
the topics with which that book deals.
In getting these essays together I have benefited not only from the
obvious help of the contributors themselves and the enthusiasm with
which they greeted the project, but also, in particular, from the
encouragement (which always seemed to come when it was most
needed) of Robert Butts and Mary Hesse. I am grateful, too, to Gerd
Buchdahl for supplying his own bibliography. Gratitude, and more, is
due him for the very creation of all the items listed there, of course, as
also for the generous enthusiasm, both intellectual and personal, which
so noticeably mark him out. He is the sine qua non of this collection in
many more ways than one. I hope it will serve as a token of the regard
and esteem due him.
R. S. WOOLHOUSE
ix
GERD BUCHDAHL: BIOGRAPHICAL

AND BIBLIOGRAPHICAL
1. BIOGRAPHICAL
Gerd Buchdahl was born on 12 August 1914 in Mainz, Germany of

liberal Jewish parents. His earlies~ memory, from 1917, is of lying in a
cot in the basement listening to the explosions of English and French
bombs, his mother trembling by his side. His father Max, who ran a
retail bedding and furniture business, came from Brilon in Westphalia;
his mother Emmy (nee Bendix) came from Hamlyn. He matriculated
from the Realgymnasium Mainz in March 1933, two months after
Hitler came to power. Later that year the offer of a job in Berlin, which
would have given him business experience prior to running the family
firm, was withdrawn because of his race. Consequently, and as it had
always been intended he should spend some time abroad, he came to
England, where he quickly decided he wanted to stay. He worked for a
few months in a downquilt factory before enrolling (May 1934) in a
Diploma Course in Structural Engineering and Reinforced Concrete at
what is now Brixton Polytechnic. He completed this in 1936, and acted
for a time as a "half commission man" with the London stockbroking
firm, Cassell and Co. In 1938 the receipt of an alien's work permit
allowed him to begin his career as a civil engineer with H. J. Paton and,
later, Mouchel and Partners for whom he designed reinforced concrete
structures. Though in that same year he had qualified for naturalisation
he was still nominally German at the outbreak of the war and so was
interned in Liverpool in June 1940 and then transported to Australia
on the S.S. Dunera. Late in 1941 he was released (as what eventually
turned out to be Australian immigrant no. 8 since the outbreak of the
war) to work again as a civil engineer, which he did till 1947.
Gerd Buchdahl introduced himself to philosophy at the age of fifteen
when, "with bated breath", he read Kant's arguments about space and
time in his father's so far unopened Insel Verlag edition of the Critique.
Later, in London, a girlfriend's cousin gave him A. N. Whitehead's
Adventure of Ideas to read. Out of two thousand or so internees on the
Dunera, he was the only one to have brought a philosophy book with
1
R. S. Woolhouse (ed.), Metaphysics and Philosophy of Science in the Seventeenth and
Eighteenth Centuries, 1-7.
1988 by Kluwer Academic Publishers.
GERD BUCHDAHL
him - C. E. M. Joad's Guide to Philosophy. As one of the "table

fathers" on his deck, he gathered around him a large number of
deck-mates, aged fifteen to seventy, among other things using Joad's
long quotes from Plato and Aristotle as the basis for reading and
discussion sessions.
In March 1942, having recently read J. Laird's Gifford Lectures
Theism and Cosmology he decided his real interests lay in philosophy.
So, persuading him of a fascination with refutations of proofs of the
existence of God, he talked Professor Boyce Gibson into allowing him
late into the philosophy course at Melbourne University. For four years
he worked all but full-time in the engineering office, and read philosophy throughout the evenings with the guidance of a very few seminars,
lectures, and weekend discussions with teachers such as Paul Edwards,
Dan Taylor, G. A. Paul, and A. C. Jackson. These last two, having been
graduate students of Wittgenstein, made a great impression in terms of
the ideals and standards they brought to bear. In his second year Kurt
Baier (then a third year student) asked him to conduct a Kant seminar,
because he seemed "to be the only one who knew what was going on
there". He eventually graduated with a B.A. (First Class) in 1946, and,
there being no Melbourne Ph.D., completed a 200 page M.A. thesis on
"The Relation between Concept and Object" in 1953.
In June 1947, now with qualifications in philosophy as well as in
engineering, he started work in the University'S recently begun Arts
department of "General Science", and was soon put in charge of it.
Particularly after being introduced to Mach's Science of Mechanics he
saw the desirability of teaching the subject historically. But, even
approached in this way, science did not go well with Arts students and
he began to develop his present conception of History and Philosophy
of Science. His ideas were stimulated by J. B. Conant's Understanding
Science (1946), the Harvard Case Histories in Experimental Science
(1950-4), and H. Butterfield's Origins of Modern Science (1949). By
the time he left Melbourne in 1957 the department had moved under
him from "General Science", through "History and Methods of Science",
to "History and Philosophy of Science", and had grown to have a
building, secretarial staff, a sizeable library, and half a dozen teaching
staff.
The year 1953-4 involved an exchange, as Lecturer in Philosophy
of Science, with Stephen Toulmin in Oxford, during which he gave
a lecture in Cambridge. This eventually led to his being offered a
BIOGRAPHICAL AND BIBLIOGRAPHICAL
Lectureship in Philosophy of Science at Cambridge when N. R. Hanson

resigned in 1957.
Since 1952 the teaching of History and Philosophy of Science at
Cambridge had been administered though not academically led, by a
Committee which, through the medium of Norman Hanson and Rupert
Hall, contributed half a Tripos Part 1 subject. In about 1960, two years
after his arrival in Cambridge, Buchdahl became Secretary of the
Committee and, over the years, came to have the responsibilities of an
academic head of department. Under his guidance books, secretaries,
teaching rooms, and more teaching staff were acquired. In 1965 the
Certificate in History and Philosophy of Science became a fully-fledged
Part 1. In 1971 a postgraduate Diploma was introduced, being replaced
in 1977 by an M. Phil. These years saw also a growing number of Ph.D.
students. In 1972 HPS at Cambridge became an official "Department"
run by a Syndicate, and with Buchdahl as its Head (a post he resigned
in 1974). By this time it had grown to have two Readers, two Lecturers,
two Assistant Lecturers, and two Assistant Directors of Research.
Buchdahl's official retirement from the Department was in 1981,
although, as Emeritus Reader in History and Philosophy of Science, he
has continued to teach and work there. During his career at Cambridge
he has been a visiting professor at the Universities of Stanford (1965),
of Western Ontario (1966-7), of Texas at Austin (1969, 1971), and of
California at San Diego (1982).
In about 1968 Gerd Buchdahl conceived the idea of a journal,
Studies in History and Philosophy of Science, which would encapsulate
his view of that subject as it had been developing over the years. As he
has described it, in its pages "authors would be not only permitted but
explicitly encouraged to discuss philosophical issues by reference to
their historical context, and historical issues in terms of the philosophical framework in which they had occurred" ('Philosophy of science: Its
historical roots', Epistemologia 10 (1987), 40).
2. SELECTIVE BIBLIOGRAPHY
'Logic and history: An assessment of R. G. Collingwood's Idea of History', Australian

Journal of Philosophy 23 (1948), 94-113.
'History and methods of science', University of Melbourne Gazette 6 (1950), 71-2.
'Induction and scientific method', Mind 60 (1951),16-34.
'Science and logic: Some thoughts on Newton's second law of motion in classical
mechanics', British Journalfor the Philosophy of Science 2 (1951), 217-35.
GERD BUCHDAHL
'Inductive process and inductive inference', Australian Journal of Philosophy 34 (1956),

164-81.
'Science and metaphysics', in The Nature of Metaphysics, ed. D. F. Pears (London:
Macmillan, 1957),61-82.
Review of Readings in the Philosophy of Science, eds. H. Feigl and M. Brodbeck, Mind
66 (1957), 411-14.
Review of International Encylopedia of Unified Science, eds. O. Neurath, R. Carnap,
and C. Morris, Australian Journal of Philosophy 35 (1957), 60-7.
'Has Collingwood been unfortunate in his critics?', Australian Journal of Philosophy 36
(1958),95-108.
Review of The Crime of Galileo, Giorgio de Santillana, The Cambridge Review 80
(1958),47.
'Sources of scepticism in atomi(' theory', British Journal for the Philosophy of Science
10 (1959),120-34.
'The making of modern science', The Listener 61 (1959),927-9.
Review of History and Philosophy of Science. An Introduction, L. W. H. Hull, The
Cambridge Review 81 (1959),63-5.
Review of Patterns of Discovery, N. R. Hanson, Nature 184 (1959), 572-3.
'The natural philosophy', in The Making of Modern Science, ed. A. R. Hall (Leicester:
Leicester Univ. Press, 1960),9-16.
'Convention, falsification and induction', Proceedings of the Aristotelian Society Supplementary Volume 34 (1960), 113-30.
'Philosophy and science', Cambridge Opinion 19 (1960), 11-16.
'Interpreting science to non-scientists', The Listener 63 (1960), 1007-10.
'Newton on the nature of light: The history of a controversy' (review of The Correspondence of Isaac Newton, Vol. 1: 1661-1675, ed. H. W. Turnbull), The Cambridge
Review 81 (1960),398-401.
Review of The Reach of science. Henryk Mehlberg, Mind 69 (1960), 101-4.
Review of An Introduction to the Logic of the Sciences, R. Harre, Philosophical Books
1 (1960),5-7.
Review of Etudes Coperniciennes, ed. S. Wedkiewicz, Archives Internationales d'Histoire
des Sciences 45 (1960), 421-23.
The Image of Newton and Locke in the Age of Reason (London: Sheed and Ward,
1961).
'The problem of negation', Philosophy and Phenomenological Research 22 (1961),
163-78.
'The philosophical basis of physics', Contemporary Physics 3 (1962), 182-94.
'History and philosophy of science at Cambridge', History of Science 1 (1962), 1-8.
'The study of science as vocation and as criticism', The Cambridge Review 83 (1962),
512-17.
'Models in science' (review of Forces and Fields, Mary B. Hesse), The Cambridge
Review 83 (1962), 41-2.
'Inward nature versus objectivity' (review of The Edge of Objectivity. An Essay in the
History of Scientific Ideas, C. C. Gillispie), History of Science 1 (1962),90-5.
Review of Einstein und die Sovietphilosophie. Bk. 1: Die Grundlagen, Die spezielle
Relativitiitstheorie, S. Muller-Markus, Soviet Studies 13 (1962), 443-6.
Induction and Necessity in the Philosophy ofAristotle (London: Aquin Press, 1963).
'Descartes' anticipation of a logic of scientific discovery', in Scientific Change, ed. A. C.
Crombie (London: Heinemann, 1963), 399-417.
'The relevance of Descartes' philosophy for modern philosophy of science', British
Journal for the History of Science I (1963), 229-49.
'Minimum principles in science and philosophy during the seventeenth and eighteenth
centuries', Proceedings of the Tenth International Congress of the History of Science
(Ithaca, 1962) (Paris: Hermann, 1964) I, 299-302.
'Theory construction: The work of Norman Robert Campbell', Isis 55 (1964), 151-62.
'Causality, causal laws and scientific theory in the philosophy of Kant', British Journal
for the Philosophy of Science 16 (1965),187-208.
'A revolution in historiography of science' (review of The Structure of Scientific Revolutions, Thomas S. Kuhn, and Towards an Historiography of Science, Joseph Agassi),
History of Science 4 (1965), 55-69.
Review of The Displacement of Concepts, Donald A. Schon, The Philosophical
Quarterly 16 (1966), 86-7.
'Semantic sources of the concept of law', Synthese 17 (1967), 54-74; also in In
Memory of Norwood Russell Hanson, eds. R. S. Cohen and M. W. Wartofsky
(Dordrecht: Reidel, 1967) (= Boston Studies in the Philosophy of Science, vol. 3),
272-92.
'The relation between "understanding" and "reason" in the architectonic of Kant's
philosophy', Proceedings of the Aristotelian Society 67 (1967), 209-26.
'N. R. Campbell', in Dictionary of Philosophy, ed. Paul Edwards (New York: Macmillan,
1967) 2,13-15.
Metaphysics and The Philosophy of Science. The Classical Origins, Descartes to Kant
(Oxford: Blackwell, 1969).
'The Kantian dynamic of reason with special reference to the place of causality in
Kant's system', in Kant Studies Today, ed. Lewis White Beck (La Salle IL: Open
Court, 1969),341-74.
'Gravity and intelligibility: Newton to Kant', in The Methodological Heritage of Newton,
eds. Robert E. Butts and John W. Davis (Toronto: Univ. of Toronto Press, 1970),
74-102.
'History of science and criteria of choice', in Historical and Philosophical Perspectives
of Science, ed. Roger H. Steuwer (Minneapolis: Univ. of Minnesota Press, 1970) (=
Minnesota Studies in the Philosophy of Science, vol. 5), 204-29, 239-45.
'George Berkeley', in Dictionary of Scientific Biography, ed. C. C. Gillispie (New York:
Scribner, 1970)2, 16-18.
'History and methodology' (review of Fact and Theory. An Aspect of Philosophy of
Science, W. M. O'Neil) History of Science 9 (1970),93-101.
'Inductivist versus deductivist approaches in the philosophy of science', Monist 55
(1971),343-67.
'The conception of lawlikeness in Kant's philosophy of nature', in Kant's Theory of
Knowledge, ed. Lewis White Beck (Dordrecht: Reidel, 1974), 128-50; also in
Synthese 23 (1971), 24-46; and Proceedings of the Third International Kant
Congress (University of Rochester, 1970) (Dordrecht: Reidel, 1972), 149-71; and
as 'Der Begriff der Gesetzmiissigkeit in Kants Philosophie der Naturwissenschaft', in
GERD BUCHDAHL
Zur Kantforschung der Gegenwart, eds. Peter Heintel and Ludwig Nagl (Darmstadt:
Wissenschaftliche Buchgesellschaft, 1981), 90-121.
'Is science cumulative?', New Edinburgh Review, nr. 13 (1971), 4-11.
'Hegel's philosophy of nature', British Journal for the Philosophy of Science 23 (1972),
257-66.
'Methodological aspects of Kepler's theory of refraction', Studies in the History and
Philosophy of Science 3 (1972), 256-98; also in Internationales Kepler-Symposium
(Weil der Stadt, 1971), eds. F. Krafft, K. Meyer, and B. Sticker (Hildesheim:
Gerstenberg, 1973), 131-54.
'Hegel's philosophy of nature' (reviews of G. W. F. Hegel (1842), Philosophy of Nature,
tr. and ed. M. J. Petry, and G. W. F. Hegel (1847), Philosophy of Nature, tr. A. V.
Miller), British Joumalfor the Philosophy of Science 23 (1972), 257-66.
Review of Heidegger, Kant and Time, Charles M. Sherover, Isis 63 (1972), 569-70.
'Explanation and gravity', in Changing Perspectives in the History of Science, eds.
Mikulas Teich and Robert Young (London: Heinemann, 1973), 167-203.
'Leading principles and induction: The methodology of Matthias Schleiden', in Foundations of Scientific Method: The Nineteenth Century, eds. Ronald N. Giere and
Richard S. Westfall (Bloomington: Indiana Univ. Press, 1973),23-52.
'Hegel's conception of "Begriffsbestimmung" and the philosophy of science', in Proceedings of the Fourth International Congress of Logic, Methodology and Philosophy
of Science (Bucharest, 1971), eds. P. Suppes, L. Henkin, A. Joja, and GR. C. Moisil
(Amsterdam: North-Holland, 1973),943-55.
'Hegel's philosophy of nature and the structure of science', Ratio 15 (1973) 1-27; also
in Hegel, ed. Michael Inwood (Oxford: Oxford Univ. Press, 1985).
'Transcendental reduction: A concept for the interpretation of Kant's critical method',
Kant-Studien (Special Issue) 65 (1974), 28-44.
'Christian Wolff', in Dictionary of Scientific Biography, ed. C. C. Gillispie (New York:
Scribner, 1976) 14, 482-4.
'Philosophische Grundlagen einer historischen Bewertung der Wissenschaft', in Wiener
Jarbuch for Philosophie, ed. Erich Heintel (Wien: Braumuller, 1979) 12, 16-42.
'The interaction between science, philosophy and theology in the thought of Leibniz',
Studia Leibnitiana, Sonderheft 9 (1979), 74-83.
'Neo-transcendental approaches towards scientific theory appraisal', in Science, Belief
and Behaviour, ed. D. H. Mellor (Cambridge: Cambridge Univ. Press, 1980), 1-22.
'The dynamical version of Kant's transcendental method', in Acts of the Fifth International Kant-Congress (Mainz, 1981), ed. Gerhard Funke (Bonn: Bouvier Verlag
Herbert Grundmann, 1981) 1.1, 394-406.
'Reduction-realization: A key to the structure of Kant's thought', in Essays on Kant's
Critique of Pure Reason, eds. J. N. Mohanty and Robert W. Shahan (Norman: Univ.
of Oklahoma Press, 1982), 39-98.
'Response to David Bloor's "Durkheim and Mauss revisited: Classification and the
sociology of knowledge', Studies in the History and Philosophy of Science 13 (1982),
299-304.
'Styles of scientific thinking', in Proceedings of the International Conference on Using
History of Physics in Innovatory Physics Education (Pavia, 1983), eds. F. Bevilacqua
and P. J. Kennedy (Pavia: Centro Studie per la Didattica, Univ. di Pavia, and The
International Commission on Physics Education, 1983), 106-27.
'Conceptual analysis and scientific theory in Hegel's philosophy of nature (with special
reference to Hegel's optics)', in Hegel and the Sciences, eds. Robert S. Cohen and
Marx W. Wartofsky (Reidel: Dordrecht, 1984) (= Boston Studies in the Philosophy
of Science, vol. 64),13-36.
'Zum Verhiiltnis von allgemeiner Metaphysik der Natur und besonderer Metaphysischer
Naturwissenschaft bei Kant', in Probleme der "Kritik der reinen Vernunft", ed.
Burkhard Tuschling (Berlin: de Gruyter, 1984),97-142.
Transzendentale Beweisfiihrungen in Kanis Philosophie der Wissenschaft', in Bedingungen der Moglichkeit: 'Transcendental Arguments' und transzendentales Denken,
eds. Eva Schaper and Wilhelm Vossenkuhl (Stuttgart: Klett-Cotta, 1984), 104-14.
'Kant's "special metaphysics" and The Metaphysical Foundations of Natural Science', in
Kant's Philosophy of Physical Sciences, ed. Robert E. Butts (Dordrecht: Reidel,
1986),127-62.
'Metaphysical and internal realism: The relations between ontology and methodology in
Kant's philosophy of science', in Proceedings of the Seventh International Congress
of Logic, Methodology and Philosophy of Science (Salzburg, 1983), eds. Ruth
Barcan Marcus et al. (Amsterdam: North-Holland, 1986),623-41.
'Stadien der begrifflichen Entwicklung von Atomtheorien', in Begriffswandel und
Erkenntnisfortschritt in den Erfahrungswissenschaften, eds. Friedrich Rapp and
Hans-Werner Schutt (Berlin: TUB-Dokumentation Kongresse und Tagungen,
1987),101-30.
'Philosophy of science: Its historical roots', in Les relations mutuelles entre la philosophie des sciences et l'histoire des sciences, in Archives de I'institut international des
sciences theoriques (Bruxelles: Office International de Librairie, 1987), 39-56; also
in Epistemologia 10 (1987), 39-56.
'Inductivist versus deductivist approaches in the philosophy of science as illustrated by
some controversies between Whewell and Mill', in William Whewell: A Composite
Portrait, eds. Menachem Fisch and Simon Schaffer (Oxford: Oxford Univ. Press,
1988) (forthcoming).
'Realism and realization in a Kantian light', in Reading Kant: Critique and Transcendental Arguments, eds. Eva Schaper and Wilhelm Vossenkuhl (Oxford: Blackwell,
1988) (forthcoming).
'Oas Problem des Wissenschaftesrealismus in Kantischer Sicht', in Tradition und
Innovation, ed. Wolfgang Kluxen (Hamburg: Meiner, 1988), 110-34.
Kant and the Dynamics of Reason (Oxford: Blackwell, 1988) (forthcoming).
'Reductive realism and the problem of affection in Kant', in An Intimate Relation:
Studies in the History and Philosophy of Science, eds. 1. Brown and 1. Mittelstrass
(Dordrecht: Reidel, 1989) (forthcoming).
ROBERT E. BUTTS
GERD BUCHDAHL: A TRIBUTE
I cannot remember the exact date, but it was one day in autumn 1962
that I went to meet Gerd Buchdahl in his office in Free School Lane,
Cambridge. Those who have come to know me in more recent times
will think I am joking when I say that I entered the offices of the
Department of History and Philosophy of Science in an uneasy and
hesitating state of mind. After teaching for eight years in isolated liberal
arts colleges in the United States I was in awe of Cambridge and its
academic "stars", which led me to be circumspect. Gerd seemed sensitive to my caution and I soon felt completely at ease. I had sent him
reprints of two of my papers on Kant's theory of hypothesizing. He
remarked that he liked my articles, but that he "saw these matters in a
quite different way". (Of course he did! There was a major blunder in
those papers, one that his own "looseness of fit" interpretation of Kant
would expose, and one that I hope to have decisively eliminated in later
work.) He talked about that "way", one that was subsequently to
energize his important and path-breaking work on Kant. We talked
about how my work on the Whewell papers was progressing and about
academia on both sides of the Atlantic.
At that first meeting Gerd presented features of himself that have
been reconfirmed at each subsequent meeting. One is struck by his
immense learning and his deep scholarship - his ability to direct us to
apt pieces of text provides a model for exact work in the history and
philosophy of science, One is struck by his intellectual energy and by
the lively way in which he presents his ideas. The strategy is Humean:
by sheer force of presentation he seeks to convert an initially dull idea
into that which has the persuasive power of a lively impression. One is
struck by the boldness of his interpretive schemes, schemes that so
often go against historical and philosophical readings thought to be
sacrosanct. Gerd projects a kind of methodology for "living" history
and philosophy. It is not just a matter of appraising arguments and
getting the textual facts straight. In person (if not in his writings) Gerd
behaves as if fruitful ideas are to be communicated in a way analogous
to the way in which diseases spread. He seeks to infect us with his ideas
through contagion.
9
10
ROBERT E. BUTTS
Those persons who have been his students and colleagues in

Cambridge will know in detail how these features of Gerd's personality
and his talents took them to deeper levels of understanding. They will
know of his powerful influence in the creation of a new ''field'' of study:
history and philosophy of science (HPS); and of his role in launching
Studies in History and Philosophy of Science. Others will know him
from his presentations at scholarly meetings, and from his writings;
chief among them his extraordinarily influential Metaphysics and the
Philosophy of Science, a work that seeks with a high level of success to
bring an understanding of the history of philosophy into contact with
current work in philosophy of science. Many will not know about
another of Gerd's academic achievements, one that in the Victorian
period would have been called, no doubt, his "colonial initiative". I refer
to his contributions to the early development of the graduate programme in philosophy of science in the University of Western Ontario.
The graduate programme in philosophy of science at Western was
launched in 1965. No longer in awe of the personalities at Cambridge,
but in a mood of full appreciation of their talents and accomplishments,
I decided to ask Gerd Buchdahl to serve as our first visiting professor.
The point was to give our early students (and many of our colleagues)
direct exposure to important methods of working in history and
philosophy of science. Gerd's contributions to the early development of
our programme cannot be overestimated. What he chiefly accomplished
was the creation of the kind of ambience required in order to convert
local understanding of philosophy as an amiable, non-aggressive and
largely contentless exchange of opinions between gentlemen and gentlewomen into a recognition of philosophy as the hard professional work
of clarifying and hopefully of solving fundamental conceptual problems.
Gerd served as visiting professor at Western during the academic
year 1966-7. Many of our students (and staff) were overwhelmed by
his Humean presence: lively impressions broke out all over the place.
He also provided the occasion for the first of many successful colloquia
on topics in history and philosophy of science, the colloquium on
implications of Newtonian methodology. The approach to philosophy
of science his work represented took hold at Western, and is now an
entrenched feature of our programme, one in which we seek to do
justice both to more formal work in philosophy of science and to the
history of both science and philosophy.
For me personally, that academic year was to provide decisive
11
direction in several areas. I learned something from Gerd about

building academic programmes. I also learned an immense amount
about Kant. Gerd was writing the section on Kant in his book,
Metaphysics and the Philosophy of Science; I was directing a seminar
on Kant's second and third Critiques. Gerd joined the seminar (I had by
now lost all of the awe and felt like a colleague at last), at which point it
became our seminar. For me, and I think for our best students, it was
an experience not to be missed. Gerd claims that it was also a turning
point for him. In 1983 I chaired the session of the Salzburg Congress of
Logic, Methodology and Philosophy of Science in which Gerd gave his
invited paper on Kant. We offered a fair amount of criticism of his
"looseness of fit" interpretation of Kant (more of which below). After
the session, he said to me "It is all your fault", alluding to a time when I
asked him to take charge of the seminar during my absence. I am happy
to take the blame for thus providing a context for the development of
some of the most exciting ideas about Kant to come forward in this
century. But I am afraid that my only direct contribution to the book he
was then writing was the loan of my Insel Verlag edition of Kritik der
reinen Vemunft!
Our parallel careers as students of Kant have had non-Euclidian
connections at various points: the graduate seminar, the production of
the colloquium volume, The Methodological Heritage of Newton, the
symposium on Kant's philosophy of science at the Third International
Kant Congress in 1970, the Salzburg congress in 1983, cooperation in
the preparation of the volume, Kant's Philosophy of Physical Science,
published in 1986. For me it would have been especially rewarding if
the lines had intersected more frequently.
I hope that these personal recollections are also read as an act of
appreciation of Gerd's impressive work as teacher, colleague, author
and friend. Because I know his work on Kant best, I will close this essay
by trying to give some impression of the importance of his work on this
philosopher. Gerd has given us interesting and fruitful ideas about the
work of a number of modem philosophers, and Metaphysics and the
Philosophy of Science introduces a methodology for studying the
history of philosophy that is novel and stimulating. I hope that limiting
my remarks to his work on Kant will not subtract from the acknowledged importance of his work on other historical figures and on
methodology.
Compare work done on Kant by English and North American
12
ROBERT E. BUTTS
scholars prior to 1969 (the date of publication of Metaphysics and the

Philosophy of Science) and in the period beginning in 1978 with
publication of Gordon Brittan's important Kant's Theory of Science.
A useful way in which to make the comparison is to begin with
Scott-Taggart's "Recent Work on the Philosophy of Kant" in Kant
Studies Today (edited by Lewis White Beck), also published in 1969.
Scott-Taggart's selected bibliography lists four important book-length
studies of Kant's theory of science: Gottfried Martin, Immanuel Kant,
Ontologie und Wissenscltaftstheorie (1951; translated into English as
Kant's Metaphysics and Theory of Science, 1955); Jules Vuillemin,
Physique et mhaphysique Kantiennes (1955); Peter Plaass, Kants
Theorie der Naturwissenschaft (1965); and Lother Schafer, Kants
Metaphysik der Nature (1966). These works (particularly the ones in
German) in some sense continue the tradition, established by the
Marburg Neo-Kantians, of viewing the Critique of Pure Reason as at
least in the main a theory of science. Scott-Taggart's bibliography
reveals that no major studies of Kant's philosophy of science had appeared in English in the post-Second World War years up to 1969. To
be sure, major works in English on Kant were available. Discussions of
Kantian epistemology were dominated by attention to Peter Strawson's
Bounds of Sense, Jonathan Bennett's Kant's Analytic and Graham
Bird's Kant's Theory of Knowledge. We had only a handful of shorter
studies in English of aspects of Kantian philosophy of science to appeal
to (papers by Buchdahl, Butts, McRae, Parsons, Hintikka). I do not
suggest that the scholarly impact of the shorter papers was slight.
Buchdahl's papers worked out details of things to come; those of
Parsons and Hintikka (particularly those dealing with Kant's views on
mathematics) came to be quite influential.
In the main, Kant's crowning achievements in epistemology were
studied in Anglo-America on the tacit assumption that whatever quaint
thoughts about science Kant may have had, attention to such ideas will
only serve to distract us from paying attention to the central issues Kant
raises. Such ideas have nothing to do with the lasting insights of the
epistemology of Kant, insights that can be understood in complete
isolation from whatever dated and subsequently discredited ideas about
science Kant may have expressed. A battery of issues in epistemology
had been bequeathed to Kant in the work of Descartes, Leibniz, Locke,
Berkeley and Hume. What we need to study is Kant's contribution to
these central philosophical issues, ones that the history of modem
13
philosophy displays as having a life of their own that is quite independent of Kant's contemporary scientific situation. Well, not only a life of
their own, but indeed an immortality, since it is part of the philosophical folk wisdom that each new generation needs to find its own way of
expressing and addressing the problems of theory of knowledge. After
all, the science of Kant's day has been superseded or discarded. Not so
the philosophy of his day (his philosophy) - we may even be able to
show that Kant's solutions to the key problems of theory of knowledge
are immortal. We cannot do this, of course, if we see these proposed
solutions as closely tied to scientific ideas now thought to be false or in
need of serious qualification.
Buchdahl's chapter on Kant in Metaphysics and the Philosophy of
Science changed these attitudes towards Kant's views on science in
decisive ways. After Buchdahl's contributions had been absorbed by a
group of younger scholars who were more than well versed both in the
ways of science and the ways of philosophy, it became more difficult to
regard the Kantian epistemology as importantly separate from his
philosophy of science. That is what Buchdahl proved to us: Kant had a
philosophy of science. Brittan's 1978 book followed the lead; his
subsequent work also acknowledges the influence of Buchdahl. More
recently, the extremely important work on Kant's philosophy of science
offered by Michael Friedman and Philip Kitcher has acknowledged its
debt to Buchdahl.
What is it about Buchdahl's reading of Kant that comes to have such
an important place in recent studies of Kant's philosophy of science?
First and foremost, Buchdahl revived the emphasis of the Marburg
School on Kant's own explicit insistence that his major work in
foundations of science - the Metaphysical Foundations of Natural
Science - provides the speCUlative science that instances the table of
the categories of the first Critique. If the link between these two works
is understood as Kant intended, then the Critique itself must be read as
providing a theory of scjence, and the major emphasis of that work is
now seen as a philosophical attempt to find the right assessment of
Newton's achievement against the background of Kant's continuing
acceptance of certain features of Leibniz's metaphysics. If we read the
great Critique in this way every important ingredient of that work
requires a new understanding.
Most importantly, the relationships Kant thought to have established
between reason and understanding, and between conceptual formalism
14
ROBERT E. BUTTS
and experience, take on a significance hardly even noticed in earlier

Anglo-American studies of Kant. Many readers, convinced that the
sections on teleology in the Critique of Judgement are incoherent, and
that the Appendix to the Transcendental Dialectic of the first Critique
bestows regulative status only on the ideas of god, freedom and
immortality, misunderstood the message Kant endeavored to convey
about regulation as system, about the crucial role played by reason's
urge to systematize in the development of scientific methodology. If we
fail to see the crucial importance for the development of science of the
ideal of the maximum, of system, we are left with only an odd
philosophy of religion (and the first stirrings of a theory of morality) as
the legacy of Kant's discussion of reason in the first Critique. We then
have difficulty understanding the antinomies, and are perhaps likely to
think that the third and fourth are either incoherent or inconsistent.
However, if the conditions on intelligibility introduced in the discussion
of the third antinomy, and the regulative use of the idea of a necessary
being as described in the third are understood in the context of reason
as system, and scientific theorising as projecting an order of nature,
much of the distress is alleviated, and we get a new perspective on
Kant's strategy.
Buchdahl's illuminating discussions of the role of reason in Kant's
philosophy of science show pretty clearly that Kant's theory of method
contrasts sharply with the empiricist one, moving the stress points of
method away from concerns about local confirmation or verification of
empirical hypotheses and on to efforts to show that theoretical unification of separate laws provides evidential strength to hypotheses whose
content is not obviously directly empirical. Buchdahl's contribution to
our understanding of Kant's own struggle to deal adequately with
"experience" is equally important.
The application of the categories to sensible intuitions "constructs"
objects of experience. "Experience" is ambiguous in usages in the
Kantian literature. It can mean "experience as objective", in which case
that which is experienced is measurable, substantive, related causally to
other experiences, and, in general, is that which is subject to proper
scientific treatment. There is also "experience" as the sense-content of
that which is intuited; experience as appearance. How is the lawful
structure of objects of experience in the objective sense related to
largely subjective experience as that which appears? One of the most
15
interesting and contentious theses of Buchdahl's reading of Kant is what

he calls the "looseness of fit" between the formalism of the categorial
structure and the world of empirical happenings viewed as appearances.
One might have thought that the a priori status of the categories the set of the categories is privileged and unique - requires that the
"fit" emerging from the application of each category to an appearance
would likewise be privileged and unique. For human knowers there can
be no possible object of experience that is not determined by the
application of the categories. This view may be taken to entail the
Leibnizian conceit that, given a complete knowledge of all possible
applications of the categories, we would be able to deduce the empirical
(appearance) statements that turn out to be "truths" about our world.
The entailment seems all the stronger when we reflect upon Kant's
claims about the a priori nature of the pure laws of natural science.
Each category is the ingredient concept in a principle of the understanding. For example, the category of cause and effect is conceptualized in the synthetic a priori causal principle of the Second
Analogy. This principle is instanced by the (apparently equally synthetic
a priori) law of pure physics, Kant's second law of mechanics, what
we know as the law of inertia. But if the laws of mechanics are true
a priori, what role does experience in any Kantian sense of that term
play in the confirmation or verification of scientific laws?
Buchdahl shows conclusively that there is no deductive road from
the system of categories to empirical laws. He also shows that the
connection between the system of categories and experience must be
loose in another respect if we are to rescue moral imperatives and
judgements in cultural matters from the apparent threat of determinism
raised by the necessity of second analogy causality. Thus is reason's
reign over matters both natural and cultural assured. One may question
(as I have) whether Kant's system involves such a large measure of
looseness of fit as Buchdahl's reading requires. Ironically, Buchdahl
shares one impulse with those earlier students of Kant who wished to
show that the central ideas of his epistemology survive changes in
substantive science. For if the fit is loose between categorial structure
and laws of experience, then the formalism can be retained in spite of
changing knowledge of experience. Sometimes Buchdahl writes as if the
system of the categories can survive all changes in empirical knowledge.
Kant would applaud this reading; surely he wanted an eternal a priori
16
ROBERT E. BUTTS
guarantee for the privileged and unique status of the system of categories. Some of us are no longer sure that Kant, and Buchdahl, can
have it that way.
I have here provided only the barest outline of major features of
Buchdahl's rich and energetic interpretation of Kant's philosophy and
metaphysics of science. I think he would welcome the suggestion that
those who would pay him tribute should not dwell on my words, but
should return to his texts themselves. Such a tribute will be lasting and
widespread: no-one who would understand the Kantian philosophy in
its fullest extent can ignore the writings of Gerd Buchdahl. One can
make a bolder claim: no-one who would understand the history of
classical modern philosophy and its connection with science can ignore
the writings of Gerd Buchdahl. This is more than reason enough for his
friends, students and colleagues to offer him the present volume of
essays.
JURGEN MITTELSTRASS
NATURE AND SCIENCE IN THE RENAISSANCE 1
Abstract. The essay traces the development of the concept of nature in the Renaissance
against the background of a philosophical tradition that has its origins in classical times
and with special emphasis on views of nature in natural philosophy. The point of
departure is the distinction between natura naturans and natura naturata which within
the context of Western philosophy involves elements of both Aristotelian and Platonic
theories of nature. Additional matters of concern are the relation between art and
nature (particularly in Cusanus), the concept of machina mundi in the astronomical
tradition and the so-called mechanization of the world-picture. The final decline of an
Aristotelian concept of nature occurs within the context of Boyle's mechanized concept
of nature. The individual stages of this development are extensively documented.
When we think of nature we usually think of either the green world

outside our windows, a world in most cases not green any longer, or the
world of sub-atomic particles and rays, the world science investigates.
Indeed, it is sometimes difficult to believe that the nature of everyday
and common-sense experience is the very same as the nature of
scientific research. One reason for this difficulty is that the nature we
investigate in laboratories is in most cases "manufactured" (with a
considerable amount of hardware or machinery) before or even while
we examine it. The nature that science investigates has become more
and more an artifact. One of the consequences of this development is
that the "world-picture" science paints is no longer the same as the
"world-picture" of everyday life. Indeed, how can one really live in the
"world-picture" of relativity theory? On the other hand, the "worldpicture" of everyday life loses a substantial part of its rationality
because rationality has been largely identified with scientific rationality.
In other words, we have, for better or worse, become accustomed to a
world about which we know more and more in a scientific way, a world
which, for that very reason, has lost its original familiarity and its
everyday or common-sense rationality.
This was not always the case. As a matter of fact, even well into
modem times the scientific "world-picture" and the common-sense or
everyday "world-picture" were generally viewed as coextensive with
each other. Accordingly, the concept of nature was one and the same
17
R. S. Woo/house (ed.), Metaphysics and Philosophy of Science in the Seventeenth and
18
JURGEN MITTELSTRASS
both within and outside science. A particular example is found in the

relation between nature and science in Renaissance thought and in its
Aristotelian pre-history. Here, generally speaking, nature was conceived
of as an independent agent, in Aristotelian terms: as an ensemble of
independent agents. Science in this context then was to codify everyday
factual knowledge, knowledge which was also accessible to commonsense. The science or philosophy of nature, oriented towards the idea of
nature, turns out to be a part of the philosophy of human practice,
dealing with the physical "conditions" of that practice. Since the phrase
"acting nature" was originally Aristotelian as well as Platonic, let me
first make some remarks about the ancient and medieval history of the
concept of nature.
1. THE ACTING NATURE
The statement that nature is what we have not made, which can already
be found in Aristotle, is compatible with the idea that nature becomes
intelligible only because it can itself be conceived of as a system of
producing processes. (Since the Aristotelian term 7rot1]at~ means a
producing or manufacturing activity, I shall call this system, throughout
my essay, a poietic system and the producing or manufacturing activity
a poietic activity.) What comes into being and how it comes into being
is explained by the way in which it is brought about or brings itself
about. 2 Nature reveals itself to be an interacting ensemble of natural
agents which, according to Aristotle, have one principle of motion and
rest in themselves (Phys. Bl. 192b13-14). Man and nature have the
same structure, that is, they have a poietic structure. The same is true
with respect to natural and artificial processes.
No one has expounded this idea more clearly than Aristotle himself.
In his Physics, which is nothing other than the theory of his "poietic"
concept of nature, he says:
if a house were a thing created by nature, it would have been created in a way similar to
that in which it is created by art. So if things by nature were to be created not only by
nature but also by art, they would have been created just as they are by nature disposed
to be created. Also in nature they would have been created according to the order of
means and aims. In general, in some cases art completes what nature cannot carry out
to an end, in others, it imitates nature (Phys. B8.199a 12-17).
In other words: within the concept of poiesis, nature and human
NATURE AND SCIENCE IN THE RENAISSANCE
19
activity form a structural unity. In accordance with this conception the

Aristotelian concept of nature first of all does not refer to a "system of
nature" but to a particular physical thing; 3 physics in its Aristotelian
sense is not theory of nature but theory of the natural thing, defined as
having one principle of motion in itself.
Now, according to the medieval distinction between an acting or
creating nature (natura naturans) and a created nature (natura naturata),
nature in its Aristotelian sense is always acting (or creating) nature
(natura naturans). The distinction between natura naturans and natura
naturata is essentially an Aristotelian one (see [75]). Through this
distinction, the opposition between the first cause (prima causa) and
the first effect (primum causatum), which again refers to Aristotelian
concepts, is transferred to God's relation to the created world.4 This
terminological distinction between natura naturans and natura naturata
is sometimes linked to the distinction between a boundless (uncreated)
nature (natura infinita) and a bounded (created) nature (natura finita) 5
as well as to the distinction between a universal nature ( natura
universalis) and a particular nature (natura particularis ).6 The expression "natura naturata" is obviously derived from the Latin commentaries of Averroes on the Aristotelian Physics and De caelo,7 whereas
the expression "natura naturans" can probably be attributed to Michael
Scot ([94] 105). J. S. Eriugena already distinguishes in the same sense
between a creating, uncreated nature (natura creans non creata), a
created, creating nature and a created, non-creating nature.8 Even in
Aristotle - in the light of this conceptual history, but without its later
theological framework - natura naturans becomes the paradigm for
any producing activity. Poiesis appears as an imitation of nature in its
poietic character, the object produced as an imitation of "natural"
processes. The further development of the concept of nature is marked
by this Aristotelian concept of poietic nature, which, of course, has also
played an essential role in the history of poetics and aesthetics. The
adoption of Platonic elements also gave it its particular historical
profile.
Unlike Aristotle, Plato explains the structure of the physical world
not as the cooperation of natural agents but as the work of a demiurge.
Poiesis in its Aristotelian sense seems to be re!ltricted to its "technical"
aspect - no longer being the self-organization of natural things but the
activity of a powerful craftsman who, in accordance with ideas which
function as paradigms, creates the world (including the world soul)
20
JURGEN MITTELSTRASS
(Tim. 28bff). In Plato, too, describing the genesis of existing things,

serves as an explanation of what exists. Unlike Aristotelian physics,
however, the concept of created things (rtxVfJ l)v), not the concept of
natural things (cpvaet l)v), plays the fundamental role. "Poietic", according to Plato, is not nature itself, but the act which created it, even if this
is clearly designated as fictive. 9
There are two aspects that are decisive for the history of the concept
of nature in medieval and early modern times. First, the concept of
an intelligible world (x6aflo~ v017r6~),10 which has its "place" in a
hypostatic mind identified with the demiurge, was adopted from Plato's
thought by the Neo-Platonists. Second, in Augustine's (and Philo's of
Alexandria)ll Neo-Platonism the Platonic ideas become the ideas (or
thoughts) of God who creates the world according to themY The
"order of nature" (ordo naturae) owes itself to God's continuing
creative power (potentia jabricatoria)P No wonder Augustine, without
realizing the difference between Plato and Plotinus, believes himself to
be in perfect agreement with Plato. An intelligible world (mundus
intelligibilis) that became part of God's mind (ratio Dei) and real in the
course of creation, guarantees both the intelligibility and the dignity of
nature.
In the course of history, however, science, particularly physics and
physical astronomy, remained essentially Aristotelian, whereas the
terminology in natural philosophy increasingly conformed to Platonic
distinctions with Augustine as model. The intelligible or archetypal
world 14 became the infinite nature (natura infinita) of God and the
visible world (mundus sensibilis) became finite nature (natura finita).15
The gap between the Platonic concept of (the system of) nature and the
Aristotelian concept of the natural thing is bridged by the distinction
between the concepts of universal nature (natura universalis) and
particular nature (natura particularis).16 The concept of universal
nature also allows again for the Platonic idea of the world soul. This
soul sometimes appears as a fabricator (fabricator mundi) serving
God's creational purposes,17 sometimes (in a more Aristotelian sense)
as poietic nature, for example in Arabic philosophy of nature 18 and in
Philoponus: "Nature is a kind of life or force that is diffused through
bodies, that is formative of them, and that governs them; it is the
principle of motion and rest in things, and in such things alone, in
which it inheres primarily and not incidentally" ([82] 67b; ct.
[98] 286ff). According to this extended Aristotelian definition, it is the
world soul which constitutes the poietic element in nature.
21
2. NATURE AND ART
Before we turn to a history which, in the Renaissance, eventually leads

to a mechanistic interpretation of nature, to a mechanistic "worldpicture", let us once again call back to mind the mimetic aspect which is
linked to the Aristotelian concept of a poietic nature. I may recapitulate
what has already been pointed out: acting (or creating) nature (natura
naturans) is the paradigm of all poiesis; poiesis consists in imitating
(the poietic) nature. Linking the Timaeus-paradigm (i.e., Plato's account
of the genesis of the world) to the report on creation by describing the
natural thing (cpvaEl ov) in the Aristotelian sense as a created thing
(TtXVTJ ov), strengthens the poietic analogy: nature itself is nature
brought about. Imitating nature, accordingly, is imitating a producing
(or creating) activity which constitutes nature itself. It was the Cardinal
Nicholas of Cusa who articulated these relations.
Following 19 Aristotle's theorem, often quoted during the Middle
Ages,20 that art imitates nature?l Cusanus points out that art not only
presupposes nature but also expands it. 22 This again is in accordance
with Aristotle's definition, already quoted, that in some cases art
completes what nature leaves incomplete, while in others it imitates
nature (Phys. B8.199a 12-17). Cusanus, however, goes beyond this by
dissolving the conceptual opposition between nature and art. He says
"that nothing can exist which is not both nature and art; all things rather
share both nature and art .... That all perceptible things are either
natural things or artificial things is obvious" (and this, again, is literally
quoted from Aristotle (Met. Z7.1032a12-13)). "Yet it is impossible",
Cusanus continues, "that natural things are entirely separated from art
and, vice versa, that artificial things are entirely devoid of nature".23
Cusanus uses language as an example, because it results from art but
rests upon nature, as well as (logical) inference which, according to
Cusanus, belongs to the nature of man, but relies upon art. 24
This fulfills all conceptual requirements for applying the concept of
imitation, which in Aristotle is defined by the concept of poietic nature,
to poiesis itself. According to Cusanus, divine art is the paradigm of
human art,25 knowledge of the world without knowledge of (the
creating) God is impossible. Hence imitating means imitating God's
infinite art (ars infinita),26 which can be studied in nature. 27 "Our
mind", Cusanus writes, "understands God analogous to the way we
understand a builder exercising his capacity as an artisan." 28 This finally
establishes the structural identity of nature and art, already indicated in
22
rORGEN MITTELSTRASS
the Aristotelian concept of poietic nature and used in substituting the

concept of the created thing (r:tXVTJ l>v) for the concept of the natural
thing (cpVOel l>v) in the Platonic tradition. Cusanus, too, identifies in this
context Aristotelian nature with the Platonic world souI,29 And just as
the idea of nature in natural philosophy is now basically determined by
architectonic metaphors, so thought itself is seen as a poietic capacity
(virtus fingendi). The prime example of this for Cusanus is the art of the
potter, the sculptor, the blacksmith and the weaver. 30 Plato's demiurge
becomes the paradigm of a constructing mind and of the Renaissance
builder: "the visible globe", Cusanus writes in De ludo globi, "is the
image of the invisible globe which exists in the mind of the artisan." 31
3. RENAISSANCE IDEAS OF NATURE
Cusanus is not the only philosopher who, on the edge of a new age,
turned towards the new vision of nature. The nature philosophers of the
Italian Renaissance followed in his footsteps. They fall into two main
groups:
First, there were those like Cardano, Scaliger, and Patrizzi who were impressed by the
flourishing mathematical sciences, and themselves worked intensively in them. For them
nature was ultimately a vision of mathematics. They were attracted by the mathematical
Platonism of Cusanus, and by the more Pythagorean elements of the Platonic tradition
revived by the humanists. Secondly, there were those who, like Fracastoro, Telesio, and
Campanella, took seriously the Aristotelian insistence on observation and experience,
so seriously that they turned it against Aristotle's own interpretation of experience
([86J 198).
On the whole, however, Aristotelianism, with the exception of Paduan

Aristotelianism, now dissolves into conceptual dreams trying to renew,
as in Cardano, Telesio, and Giordano Bruno, speCUlative cosmologies
or, like Agrippa of Nettesheim, the van Helmonts, and Jakob Bohme,
mystic and theosophic viewpoints. Here one finds all the common
features of Renaissance conception of nature such as the panpsychistic
interpretation of nature, the parallelism between man and cosmos,32 or
the mysterious principles of nature as in the case of Archeus, a "natural
force" which guides the growth and evolution of all living beings.33
Natural philosophy falls back upon pre-Socratic conceptions of nature
as a substance, instances of which are things.
The "novelty" sometimes attributed to these conceptions in the
23
history of science and philosophy is mainly a result of the departure

from the scholastic substance-attribute schema. We can see this most
readily in theories of space in physics.34 For example, according to
Patrizzi, who uses a conceptual mixture of pre-Socratic and neoPlatonic terminology, all things originate from one principle, the One,
which combines the properties of unity and plurality in all thingS.35
Cardano combines the Aristotelian concept of first matter (materia
prima) - the counterpart of which, second matter (materia secunda),
has, since Avicenna, denoted the object of the sciences - with
hylozoistic elements typical of pre-Socratic thought ([8]). Telesio's
empiricism, which caused Francis Bacon to give him the title "the first
of the moderns",36 is again indebted to pre-Socratic and Galenian
orientations, and as Patrizzi already notes 37 is without any methodological basis (De rerum natura . .. , (Rome, 1565; enlarged 1586.
What is evident in all these conceptions is that physics, without its
former Aristotelian systematic, has now lost its bearings.
The emphasis put on the poietic structure of the order of nature (still
assumed to be divine) now becomes even more common. Let me give
three examples of this. In his influential, annotated Timaeus-translation,38 still used by Galileo 39 in his day, Marsilio Ficino calls universal
nature (natura universalis) in scholastic tradition the divine instrument
(instrumentum divinitatis),40 in other places the instrument of providence (instrumentum providentiae).41 "All parts of the universe", he
writes in De amore, "are the works of one artist and the limbs of a
single structure".42 Nature, as "God's creation" ([34]1, 119), "is guided
by God" ([34]1, 916); the "order of things", which according to Ficino
lies open to the human mind, points to the "organizer of all things"
([34]1, 709), as if physics in his time consisted entirely of a commentary on Augustine. Nature itself does not move anything in vain ([34]1,
916), because the "highest mind" does nothing in vain ([34]1, 753). Its
poietic structure manifests itself in that it acts in things like an artificer,
and in the identity of nature and art that was again being emphasised:
What is human art? A certain nature which treats matter from outside. What is nature?
An art which forms matter from inside. . .. What is a work of art? The spirit of the
artist in separated matter [Le., separated from himl. What is a work of nature? The
spirit of nature in connected matter [Le., connected with itl ([3511, 146).
Giovanni Pico della Mirandola makes even more extensive use of

poietic metaphors. Once again, God is the "highest architect"43 who
24
JURGEN MITTELSTRASS
creates the "cosmic system" in accordance with an archetypal pattern

(mundo intelligibile).44 The Timaeus is expressly mentioned in this
context. Natural causalities are "instruments" of the divine architect.
Referring to the "standpoint of the peripatetic philosophers" (!), Pico
explains how each work of nature is a product of the intelligence:45 The
most excellent part of the world machine (machina mundi) is the
celestial machine (machina caelestis), the planetary world, animated
and described according to the attributes of the Platonic world sou1. 46
Finally, at the end of the creation, the divine architect determined man
to be the spectator of the universe (contemplator mundi) "who is to
examine thoughtfully the reason in this sublime work, to love its beauty,
and to admire its greatness".47 Placed in the centre of the universe 48 in
order to see it better, man would be able to admire in the being of
things the creative capacity of God, to esteem in the truth of things the
wisdom of the architect, and to recognize from the unity of their being
the unity of their creator.49
Even one hundred years later, Tommaso Campanella in his Epilogue
on the Essence of the Universe almost literally repeats this architectonic
definition of world and man: "Man is the keystone of the whole
universe, its admirer when he wants to know God, since this is the
reason why he was created. The universe is the statue, the image, the
living temple of God".50 As far as science is concerned, it is also selfsufficient and independent: "For God has clearly constructed the world
and created things, and bestowed powers of self-preservation and
mutual change through time; those powers remain as Nature, until the
whole machine of things achieves its great end" (18) 16). Here, the
process of profaning the world is clearly perceptible: God is no longer
part of the world, he is its designer. The world can become the object
of rational (though still pious) science.
However, back to Pico. Pico recognized man as an architect, as a
being lacking an archetype and without counterpart in the intelligible
world. Unlike its spiritualization in Augustine, however, in Renaissance
thought this world becomes more worldly. According to the divine
architect man himself must determine the ''form'' in which he wants to
live.51 The poietic clasp which in Cusanus joined the object and the
manner of knowing, closes once again. And yet, if one examines Pico's
cosmology in all its detail, the universe he describes is still that of the
Middle Ages.
Apart from Cusanus' speculative cosmology and Copernican astron-
25
omy, this situation does not change until Giordano Bruno. In his
metaphysics of the infinite (influenced again by Cusanus) Bruno leaves
the medieval universe - and, unfortunately, often also the limits of
intelligibility. But here again his adherence to the architectonic metaphor is noteworthy. He calls the universal intellect, which, according to
him, determines all natural causalities, the organ of the world soul 52
and identifies it with a Platonic architect. 53 It is "the intellect of the
world which produces all things" 54 or the "inner craftsman",
because it forms matter and shapes it from within, just as it draws out and unfolds the
trunk from within the seed or root, pushes out the branches from within the trunk,
forms the twigs from within the branches, and the buds from within these, forms,
shapes and interweaves the leaves, flowers and fruits from within as if from an inner
life. And at certain times, as from within, the sap is led back from the leaves and fruits
to the twigs, from the twigs to the branches, from the branches to the trunk, and from
the trunk to the root. 5S
Even within this speculative enlargement of the language of natural

philosophy, Bruno, too, keeps to the idea of the poietic structure of the
order of nature: the universe (in Fieino and Pico) as the creation of a
divine architect, as created nature (natura naturata) in the form of a
complex piece of art, but also (in Bruno) as acting (or creating) nature
(natura naturans) guided by the effects of an internal and forming
intelligence. 56
4. MACHINA MUNDI
The expression machina mundi, or the world machine, reflects the idea
that nature itself has a poietic structure. We can already find it in
Lucretius,57 but it also occurs in Caleidius's58 Timaeus-commentary,
which was still used, as already mentioned, by Fieino. Like the expression systema mundi (system of the world) used, for example, in
Galileo's Dialogue on the Two Chief World Systems (1632) and in
Newton's De mundi systemate (1728),59 it denotes a particular part of
the world, the celestial world. As Edward Phillips states in his New
World of Words (1706), this expression
is taken for the general Fabrick, Constitution and Harmony of the Universe, or any
orderly Representation of it according to some noted Hypothesis, in which the
Heavenly Bodies are so dispos'd among themselves, as their Situation, Order, Motions
and Properties may in such an Author's Opinion best answer Appearances and
Philosophical Demonstrations.60
26
JURGEN MITTELSTRASS
The expression systema mundi (system of the world) means here the
world order as represented in an astronomical model, while the expression machina mundi (world machine) refers to the idea that this order
is a "mechanical" one. The Latin word machina renders the Greek
word f,lTJXavij (or f,laxava) which originally meant artifice or craft, but
then came to mean also the skilful use of tools and, eventually, tool
itself ([53]). Unlike l.i(>Yavov, which means "atomic" tool, f,lTJXavij, as a
rule, is used to refer to compound tools (examples are war machines or
theatre machines).61 The expression machina mundi thus emphasises
the "technological" or, again, the poietical character of the notion of
nature, without attributing to it the idea of a spiritless (or dead)
mechanism.
I have already mentioned the use of the expression machina mundi
in Pico; it is also used in Cusanus (see note 27) and Grosseteste.62 It is
safe to assume that this terminology above all was introduced into
medieval and Renaissance thought by Sacrobosco in his Sphaera
(around 1220).63 This textbook of astronomy was republished and
commented upon even in the seventeenth century.64 According to
Ch.-F. Abra de Raconis (author of a system of philosophy published in
several editions)65 it also ranks in the seventeenth century as an
important textbook of astronomy.66 Galileo, for instance, used it in his
Paduan lectures.67 In a commentary on the peripatetic book on
Mechanical Problems (Quaestiones mechanicae), published in 1599,
the world is moreover described as the greatest, most productive, most
stable and best designed of all machines. The commentary also states
that this machine, as the composition of all bodies, is God's instrument.68 Referring explicitly to Genesis and Timaeus, J. Hevelius, the
founder of lunar topography, still calls God the artificer of the machina
mundi or the universa machina mundana.69
Of course, Copernicus also uses the expression machina mundi. As
he says in the preface of his book On the Revolutions of the Heavenly
Spheres (1543), it ''was created for us by the best and most systematic
craftsman of all" ([23]5). According to Copernicus, astronomy, "divine
rather than human",7 is the search for the order this craftsman
created.71 At the same time he reminds the reader that natural
philosophers call the celestial part of the universe a "living God".72 On
the other hand, he expressly dissociates himself from the "quarrel of the
natural philosophers" (by referring to the question whether the universe
is finite or infinite in space).73 Notions about the machina mundi, which
27
are not presented in the form of textbooks of astronomy (or a textbook

of mechanics), begin to vanish from what is now called a "scientific"
examination of nature.
Within this development nature now becomes an artifact even from a
methodological point of view. Kepler, for example, substitutes (in
dynamic matters) the concept of force for the concept of SOUU4 He also
eliminates the idea of the mach ina mundi as a visible God: "My aim",
he writes in 1605 to his friend Herwart von Hohenburg,
is to show that the celestial machine is not a kind of divine living being but a kind of
clockwork ... , because the whole variety of motions depends upon a single, corporeal
magnetic force in the same way all motions of a clock depend on a very simple weight. I
also show that it is possible to determine this physical cause in a numerical and
geometrical way.7S
Here Kepler, by picking up a comparison already found in Nicholas of

Oresme,76 takes an important step towards a "mechanistic" worldpicture. He deliberately substitutes the concept of celestial physics
for the concepts of celestial theology or celestial metaphysics,77 but
maintains the Platonic view that the physical laws can be "deduced
from the idea of the creation"J8 For Kepler himself such a deduction is
achieved by linking (in his Mysterium cosmographicum (1596 planetary theory with the geometry of regular polyhedra, i.e., the so-called
Platonic solids, and by harmonics (in his Harmonice mundi (1691.
"We perceive", he writes in the dedication letter of his Mysterium
cosmographicum,
how God, like one of our own architects, approached the task of construction of the
universe with order and pattern, and laid out the individual parts accordingly, as if it
were not art which imitated Nature, but God himself who had attended to the mode of
building of the Man who was to be. 79
This is the same metaphor we found in Cusanus, Pico, and Campanella,

but now there is a remarkable variation. In creating the world, it is as if
God the craftsman had imitated "Man who was to be". It is in this
striking formulation that, in the manner of an enduring Greek imagination, the human architect begins to supersede the divine architect.
5. THE RISE OF MECHANICS
By subsuming the Keplerian laws of planetary motion under laws
28
JURGEN MITTELSTRASS
effective for terrestrial bodies (particularly in the framework of ballistics), Newton transforms Kepler's efforts toward causal hypot,heses
into a uniform mechanical explanation. Now there is a dynamic and not
just a kinematic explanation of the structure and mechanism of the
machina mundi. In its orientation towards "celestial issues", the scientific revolution in the sixteenth and seventeenth centuries turns out to
be a gift of heaven. Nevertheless, it takes place on earth (in the
framework of a theory of falling bodies).
This becomes clearer if one realizes that up to this point mechanics
has not been part of a theory of nature. According to the meaning of
f.1:YJXaV'fJ, mechanics in its Greek sense is the theory of (the mechanics
of) compound tools. As a "mechanical art" (f.1:YJxavtxf! 1:EXVIJ) mechanics
is not a theory of (the mechanics of) natural bodies but a theory of
artifacts designed to achieve what nature in fact is unable to achieve
(e.g., lifting heavy bodies). To put it into an Aristotelian terminology:
unlike physics (in its Aristotelian sense) mechanics in this tradition
deals not with natural motions but with motions contrary to nature. Its
application (e.g., in irrigating plants) does not explain nature but the
work of man. This idea is already succinctly expressed in the introductory remarks to the Quaestiones mechanicae (probably the peripatetic
extension of a work by Aristotle, ct. [54] 13ff):
Our wonder is excited, firstly, by phenomena which occur in accordance with nature
(xaTd qiVUtv) but of which we do not know the cause (aiTtOv), and secondly by those
which are produced by art (dtd TEXVTJV) despite nature (.na(>d qJVotv) for the benefit of
mankind. Nature often operates contrary to human expediency; for she always follows
the same course without deviation, whereas human expediency is always changing.
When, therefore, we have to do something contrary to nature (.naed qiVatv), the
difficulty in doing it causes us perplexity and art (TtXVTJ) has to be called to our aid. The
kind of art which helps us in such perplexities we call mechanical skill (ftTJXavfJ).80
And this understanding of mechanics did not change as long as

Aristotelian physics held the field in natural philosophy. Even in 1577
Giubaldo del Monte, a patron of Galileo, declares in his textbook of
mechanics that to this art belongs what, for example, carpenters,
builders and those carrying heavy loads achieve against the laws of
nature.8 ! Such a view was not abandoned until Galileo, who understood
that simple machines like lever, wheel and pulley do not "outwit" nature
(which is the original meaning of the Greek term flYJXaV'fJ) but represent
the skilful application of its laws. 82 Mechanics as the theory of the
motion of bodies under the influence of physical forces becomes a
29
natural science for the first time. Galileo speaks legitimately here of an
entirely new science.83 The programme to explain all natural processes
in a mechanistic way,84 determining now the further course of science,
marks the final turning-point between the Aristotelian and the modem
science of nature - not only in the light of a new theory but also with
respect to the prevailing concept of science. In retrospect, the progress
in physical theory was not on the side of "physics" in its Aristotelian
sense but on the side of peripatetic "mechanics" which, according to the
former concept of science, was not regarded as a science at all. From
the viewpoint of an eighteenth-century physicist this looks as follows:
To the art of mechanics is owing all sorts of instruments to work with, all engines of
war, ships, bridges, mills, curious roofs and arches, stately theatres, columns, pendent
galleries, and all other grand works in building. Also clocks, watches, jacks, chariots,
carts and carriages, and even the wheel barrow. Architecture, navigation, husbandry,
and military affairs, owe their invention and use to this art .... Without mechanics, a
general cannot go to war, nor besiege a town, or fortify a place. And the meanest
artificer must work mechanically, or not work at all. So that all persons whatever are
indebted to this art, from the king down to the cobbler. H5
Surprisingly, the "discovery" of mechanics as a natural science does

not, for the time being, lead to any essential change in the views of
nature in natural philosophy. On the contrary, the theological and
metaphysical implications of the concept of machina mundi, which
Kepler objected to, remain unchanged. In Galileo there is even a kind
of a rehabilitation of the concept of acting (or creating) nature (natura
naturans). "Nature", he says in the Dialogue, ''first made things in her
own way, and then made human reason skilful enough to be able to
understand [her]".86 "Nature can not be exceeded or deceived by art",
he writes elsewhere, insinuating the former idea of mechanics ([38] 8,
572). Its characterization as a kind of acting subject also remains
Aristotelian: "it does nothing in vain",87 it also always ''uses the most
simple and easiest tools".88 In speaking about laws of nature, both the
new effort toward mechanics and the old idea of a divine machina
mundi are brought to bear. Descartes' formulation of the law of gravity
(in two propositions) is indicative of that: the laws of nature are created
by God as law-maker; his perseverance (immutabilitas) guarantees their
validity.89
The language of physics, in other words, has remained poietic in its
original sense even after the Galilean revolution. Nature is conceived of
as the work of divine architecture; its effect itself is the building. The
30
JORGEN MITTELSTRASS
intelligibility of nature, which, in its former sense, is understood both as

created nature (natura naturata) and as creating (or acting) nature
(natura naturans), is guaranteed by the poietic practice of man; nature
is, in Kepler's definition, a part of this practice. This finds its methodological expression in the way the "new science" of mechanics treats
experience, in its scientific sense, along with the distinction between
conceptual and experimental procedures and, linked to that, under the
conditions of a technical practice, as a construction. 90 This step, in turn,
makes any statement about the "essence" of nature (supposed to
precede physical theories) dispensable. The phrase referring to the
"mathematization" (or, to be more precise, "geometrization") of nature
([471 20) accounts only insufficiently for this fact.
Galileo's famous version of this formula runs as follows:
Philosophy is written in that vast book which stands forever open before your eyes, I
mean the universe; but it cannot be read until we have learned the language and
become familiar with the characters in which it is written. It is written in mathematical
language, and the letters are triangles, circles and other geometrical figures, without
which means it is not humanly possible to cemprehend a single word.o l
What is meant here is that the propositions of mechanics are based

upon the measurement of geometrical quantities. This is the methodologically sound meaning of the phrase. In this interpretation it may
even already be found in Robert Grosseteste where he deals with
matherr.;uical principles of optics, the one universal science of nature he
knows: Every natural phenomenon must be explained by geometrical
lines, figures and angles. Nature always acts in the mathematically
shortest and best possible way.92 Only Descartes identifies a method
which marks out geometrical quantities with a statement about the
structure of its object (nature now apparently sufficiently determined by
the geometrical property of extension). By confusing a poietic method
("mathematization") with a natural state (nature as mathematical structure) the insight into the nature of a constructing scientific mind is lost
once again.
It is also weakened by the rise of an empiricist ideology. According
to this ideology "observation and experiment" alone are said to
guarantee that physics does not fall back into philosophical visions. 93 A
nice passage from Diderot's Indiscreet Treasures (1747) may serve here
as an illustration: Mangogul tells his dream which led him to a building
without foundations. Among crooked, naked creatures he meets an old
31
man who blows soap-bubbles into a crowd. Mangogul is in the land of

hypotheses and has met a natural philosopher. From a distance a child
approaches which gradually takes on the form of a giant. Its name is
experiment. It sets about destroying the old building - with Plato, who
had served Mangogul as a leader, escaping in time. 94
6. THE MECHANIZATION OF THE WORLD-PICTURE
Plato's flight from modern physics, which has taken the place of the old
building, is limited, however, to its own empiricist understanding of
itself. With regard to former and still prevailing views about nature this
does not (yet) happen. These views remain "pious" and, therefore, if
one takes into account the history of Platonism and its influence on the
concepts of machina mundi and systema mundi, Platonic. Even for
Newton the aim of physics consists in talking in its own way about God,
i.e., a divine mechanic. 95 An additional entry on "nature" in the second
volume of John Harris' influential Lexicon Technicum (1710), which
opens with a detailed statement about Newtonian mechanics, gives
expression to this fact. Harris calls nature "this vast Machine of the
Universe, the wise Production of Almighty God, consisting of a great
number of lesser Machines, every one of which is adjusted by the same
Wisdom in Number, Weight and Measure".96 The programme of
rational mechanics to base physics upon length, time and mass (or
force), is linked here without gaps to the historical view of the world as
machina mundi. We can find much the same written at the same time
in Vico, who interprets the now achieved "mechanization of the worldpicture" 97 as the culmination of the architectonic views of former
philosophy of nature: "Like us", he writes in his On the Study Methods
a/Our Time (1709),
the Ancients utilized geometry and mechanics as instruments of research in physics, but
not as a constant practice. We apply them consistently, and in better form .... Modem
scientists, seeking guidance in their exploration of the dark pathways of nature, have
introduced the geometrical method into physics. Holding to this method as to Ariadne's
thread, they can reach the end of their appointed journey. Do not consider them as
groping practioners of physics: they are to be viewed, instead, as the grand architects of
this limitless fabric of the world: able to give a detailed account of the ensemble of
principles according to which God has built this admirable structure of the cosmos
([95J9f).
We find it hard today to understand how such a view of a divine
32
JURGEN MITTELSTRASS
world order agrees with a mechanistic physics which tries to reduce the
world to a set of universal causal relationships. Apparently, the fascination resulting from the propagated insight into the principles of
construction and effects of God's machina mundi was greater than the
idea that man now had to accommodate himself in an automaton. The
mechanics of gravitational motions (Newton) now not only becomes
the paradigm of physical explanations within inorganic nature, it also
continues on paradigmatically into the organic, psychic and social
cosmos, particularly in the English and French Enlightenment (see
[69) 54ff). It is only a small step from Oresme's graphic rendering of
the intensity distributions of a qUality98 up to the mechanization of
qualities in Boyle. And it was Boyle, too, who, in his Free Inquiry into
the Vulgarly Receiv'd Notion of Nature (1686), taught the prevailing
philosophy of nature. As Boyle writes, it has been guided by the
following ideas:
Nature is a most wise being, that does nothing in vain; does not miss of her ends; does
always that, which (of the things she can do) is best to be done; and this she does by the
most direct or compendious ways, neither employing any things superfluous, nor being
wanting in things necessary; she teaches and inclines every one of her works to preserve
itself: and, as in the microcosm, (man) it is she, that is the curer of diseases; so in the
macrocosm (the world) for the conservation of the universe, she abhors a vacuum
([1415,219).
What Boyle describes here is, indeed, the Aristotelian concept of

nature enlarged by what natural philosophy in its history has made out
of the Aristotelian ideas. For Boyle himself, as well as modern science
in the wake of his views, this all has a merely metaphorical character:
Sometimes, when it is said that nature does this or that, it is less proper to say, that it is
done by nature, than that it is done according to nature: so that nature is not to be
looked on as a distinct or separate agent, but as a rule, or rather a system of rules,
according to which those agents and the bodies they work on, are, by the great Author
of things, determined to act and suffer ([14J 5, 219).
To get rid of these metaphors, Boyle recommends abandoning the

distinction between creating (or acting) and created nature as a
whole and replacing it with the concept of a cosmic mechanism
(mechanismus cosmicus).99 Like Newton in a methodological context,IOO
he is following here to a certain extent Bacon, who, in his criticism of
the "anticipations of nature" (anticipationes naturae), 101 already opposed
hypostatic conceptions of nature, referring to his methodology of
33
inductive arguments. According to this view, the "forms of nature" are

hidden (Bacon quotes Salomon: "it is the glory of God to hide things,
and the glory of the king to investigate them") 102 and the philosophical
mind in research into nature is blind.103 Therefore, according to Bacon,
natural history (within the meaning of the terminological distinction
between Principia-literature and Historia-literature in scientific research
(see [50], [74], [72]), not natural philosophy is the only way to the
"forms of nature".104 This, however, means that, with respect to the
Baconian tradition within modem physics, speaking about laws of
nature also becomes problematic. Perhaps this is the reason why
Newton consciously talks about laws of motion (leges motus), not (like
Descartes) about laws of nature. lOS On the other hand, the idea of a
cosmic mechanism, which Ooyle holds against the idea of an acting
nature, is also true of the Newtonian view.
As we know, even this is not the last chapter in the story of nature.
For reasons which are linked to the development of electrodynamics
and thermodynamics, the mechanistic world-picture turns out to be too
narrow, like a "mechanical myth", to use Ernst Mach's expression
([61]443). But this is another story. It tells not only of the disappearance of the pious machina-mundi metaphor from the introductory
chapters of scientific textbooks but also, for example, of Hegel's lament
over the loss of the Aristotelian concept of nature. 106 The development
of science does not admit any longer the view that God, world (or
nature) and man are in the same poietic framework (as described
impressively by Kepler). Furthermore, with the dissolution of the
"mechanical myth", it is no longer possible to view such a framework
intuitively. Nature gradually disappears not only as acting nature but
also as intuitive nature from (scientific) world-pictures. The Aristotelian
and Renaissance history of nature finally comes to an end.
NOTES
I This essay relies to some extent on material included in my [701. It was first presented at the Israel Colloquium for the History, Philosophy and Sociology of Science
(4th Annual Series 1984-5) on Tuesday 21 May 1985 at Tel-Aviv University. An
earlier version was read at a meeting of the Renaissance Society at Pittsburgh University in October 1984. I want to thank Stephen Gillies (Constance) who assisted me
in preparing the English text.
2 Met. K9.1065b15-16, cf. 1065b21ff, 1066a27ff, also Phys. fl.201a11-12. Particularly striking is the "poietic" terminology in the scientific work: nature brings about (~
34
JURGEN MITTELSTRASS
q;Vat~ d1JluOV(tyei); cf. De part. an. A5.645a9, B9.654b32, De gen. an. A23.731a24,
B6.743b23, f4.755aI9-20, De an. incessu 12.711aI8. Moreover, nature acts according to aims, it never brings about anything in vain: De an. incessu 12.711aI8 (If q;Vat~
oVdtv d1JIlWV(tyei lla7:1Jv), cf. De cael. A4.271a33 (6 dE f}eo~ xat lf q;Vat~ OMEV
lla7:1Jv nowVatv), B8.289b26, 290a31, 11.291bI3-14, De an. f9.432b21, De an.
incessu 2.704bI5, Polito A8.1256b20. Teleologically, it always strives after the best and
the perfect: De gen. et corr. BlO.336b28, Phys. 97.260b22-3, De cael. BI4.297aI6.
Cf. (101) 422ff; (92) 84ff.
3 For q;Vat~ in the sense of nature as a whole see De cael. A1.268alff., Met.
A6.987b2, f3.1005a32-3.
4 See Summa Theologia I-II quo 85, art. b; (2) IV, 21; Bonaventura, In Sent. Petri
Lombardi III, dist. 8, dub. 2 ([12)4, 183). Master Eckhart distinguishes between God's
ungenatUrter natUre and genaturter natUre ([32) 537,29-32).
, See 1. S. Eriugena, De divisione naturae libri quinque II 1, in (64)122, 523D-6C.
6 See the work of Thomas Aquinas cited above; see also below p. 20.
7 See (7) 4, 52-7 (for Phys. B1.193bI2-18), [7)5, 2 (for De cael. A1.268aI9).
H De divisione naturae libri quinque I 1 ([64)122, 441B-2B). For the history of the
distinction between natura naturans and natura naturata see also [43) 6, 504-9. See
also [96) XV 4, p. 1372 ("Natura primo dicitur dupliciter. Uno modo natura naturans,
idest ipsa summa lex naturae, quae Deus est .... Aliter vero dicitur natura naturata, et
haec multipliciter").
9 Timaeus, 29d, cf. Aristotle, De cael. B 14.297b 14-17.
IU See Plotinus, Enneades, VI 9.5,14 ([85)6.2,178).
II De opijicio mundi 6, p. 5M ([81) I, 5f (the intelligible world being the logos of God
creating the world.
12 See De diversis quaestionibus LXXXIll fiber unus, quo 46 (De ideis 2) ([6) 10, 124).
1.1 De civitate Dei XII 26 ([5)1, 553).
14 A translation of "na(!ddetYlla" (Tim. 38c) with "archetypus" already appears in the
Timaeus-translation by Ca1cidius ([16)30).
15 See J. S. Eriugena, De divisione naturae II 1 ([64)122, 523D-6C), also I 1
([64) 122,441B-2B).
16 Lexically documented in (40) 740f (entry on natura), also in (62) 880. For the
history of these concepts see [13) 6, col. 509-17.
17 See Apuleius, De Platone et eius dogmate I (1) 3, 92).
18 Documentary evidence in [29)5 (Die Naturanschauung und Naturphilosophie der
Araber im zehnten lahrhundert. Aus den Schriften der lautern BrUder) 142.
19 See Compendium IX, n. 27 ([24)2, 637), De docta ignorantia II 1, n. 94 ([24)1,
38), De coniecturis II 12, n. 131 ([24)1, 169), De ludo globi I 7 ([24) 2, 577).
20 See Thomas Aquinas, Summa contra gentiles II 75 (ars imitatur naturam, [3] 13,
475), III 10 (ars enim in sua operatione imitatur naturam, [3)14, 26); In libros
posteriorum analyticorum Aristotelis expositio I, lect. 1, n. 5 (ars imitatur naturam,
[3)1, 138); In octo libros physicorum Aristotelis II, lect. 13, n. 4 (ars imitatur naturam,
[3)2,93).
21 Phys. B2.194a21-2; cf. Phys. B8.199aI5-17; Meteor. ~3.381b6.
22 Compendium IX, n. 27 ([24)2,637).
2.1 De coniecturis II 12, n. 131 ([24) 1, 169).
24 De coniecturis II 12, n. 131 ([24)1, 169).
25 Idiota de sapientia 123 ([24) 1,224), De ludo globi I 45 ([24)2, 59lf).
35
Idiota de mente 10, n. 127 ([24]2, 263), 2, n. 61 ([24] 1,240).

Nature in this context is preferably called machina mundi, see De docta ignorantia
II 11, n. 156 (machina mundana, [24[ 1,61), II 12, n. 162 (machina mundi, [24]1,63),
II 13, n. 175 (machina mundi, [24]1, 67), II 13, n. 179 (machina mundi, [24]1, 69);
De venatione sapientiae 32, n. 95 ([24]2, 563).
28 Idiota de mente 13, n. 146 ([24J 1,271).
29 Idiota de mente 13, n. 145 ([24J 1,271).
3() De ludo globi I 44 ([24J 2, 591).
31 De ludo globi I 44 ([24] 2, 591 ).
32 For the history of this distinction see [93].
33 For example, in the work of J. B. van Helrnont and F. M. van Helrnont (see [100]
and [73J. F. M. van Helrnont combined the Archeus-conception (Seder olam, sive Ordo
seculorum. Historica enarratio doctrinae, s.p. 1693) with atomistic conceptions of
physical and mental "Minima" ("monades physicae", A Cabbalistical Dialogue ([Ist ed.
1677J, London, 1682, which (as well as possibly his father's conceptions) likely
influenced G. W. Leibniz in his theory of monads, at least terminologically (see G. W.
Leibniz, Nouveaux essais sur l'entendement humain (1704), in [58] 6.6,72).
34 Cf. [49] 88 ("The works of Telesio, Patritius, and Campanella show that Italian
natural philosophy must be credited with having emancipated the concept of space
from the scholastic substance-accident scheme. In the physics of the early seventeenth
century space becomes the necessary substratum of all physical processes. It is this
emancipated concept, divested of all inherent differentiations or forces").
35 Nova de universis philosophia ... (Ferrara, 1591) pp. 9ff (Panarchia).
36 On Principles and Origins, in [9J 5, 495 (= De principiis atque originibus (1623),
[9]3,114).
37 Documentary evidence in [36] 2, 37 5ff.
38 Compendium in Timaeum ([33J 672ff).
39 See [38J 1,35,57,72 (all passages are from the so-called Juvenilia).
4() Compendium in Timaeum ([33]672) (= [34]2,1438). Cf. [17]482 (Signa a natura
26
27
quae est instrumentum Dei multiplicia sunt).

Argumentum Marsilii Ficini in dialogum primum de Legibus, in [33] 743 (= [34]2,
1489); (In Plotinum) In librum secundum de providentia comment V (= [34]2,1705).
41
[34J 2, 1330. Cf. [55J IlIff ([56J 93ff).

Oratio de hominis dignitate, in [84J 1, 104 (= [83J 1,314).
44 Commento alia canzona d'amore I, 4-6 ([84]1, 465ff ([83]1, 898ff.
45 Heptaplus I 1 ([84]1, 204 ([83J 1, 11 .
46 Heptaplus II 6 ([84J 1,240 ([83]1, 21.
47 Oratio de hominis dignitate ([84]1, 104 ([83]1, 314. The same conception is in
Cusanus, De docta ignorantia II 13, n. 179 ([24]1,69).
48 Oratio de hominis dignitate ([84]1,106 ([83]1, 314.
49 De ente et uno VIII ([84]1, 432 ([83J 1,253.
5() 1181331 (Epilogo del senso dell'Universo).
51 Oratio de hominis dignitate ([84]1, 106 ([83]1, 314.
52 De la causa, principio, et uno II 39 ([15]1, 231).
53 De la causa, principio, et uno II 39 ([15J 1,231).
54 De la causa, principio, et uno II 41 ([15]1,232).
55 De la causa, principio, et uno II 40f ([15]1, 23lf). See the corresponding extension
of the Aristotelian concept of nature in Philoponus, above p. 20.
42
43
36
JURGEN MITTELSTRASS
De la causa, principio, et uno IV 92 ([15]1, 262f).

De rerum natura V 96. The expression is also well documented outside the tradition
of astronomy textbooks. See Gregory of Nyssa, De hominis opificio III ([65J 44, 134c);
Martianus Capella, De nuptiis Philologiae et Mercurii libri VIII, II, n. 202, IX, n. 921
([19J 76.17, 490.4); Hugh of St. Victor, De arca Noe morali libri IV, IV 7 ([64J 176,
672D); Alanus ab Insulis, Sermo de sphaera intelligibili ([59J 302, 305). See also [60J 3,
1335f (reference to Manil. II, 807, Lucan 1,79, Stat. Silv. II, 1.211).
5K [16]184.19,301.19. "Machina mundi" serves here to explain the Platonic concept
of the world body (1'0 1'oiJ x6o/iov OW/ia, Tim. 32c1). In his translation Calcidius
renders the Platonic expression as "praeclara ista machina visibilis" ([16]25.7).
59 The title refers to the second part (published posthumously) of the first version of
the Philosophiae naturalis principia mathematica (London, 1687; 2nd ed. 1713; 3rd
ed. 1726). This second part, which was included in several pieces in book 3 of the
Principia, was originally titled De motu corporum, liber secundus. Cf. [22]1 09ff, 327ff.
60 A New World of Words. Or Universal English Dictionary (London, 6th ed. 1706)
(entry on "System of the World"). See also B. Faber, Thesaurus Eruditionis Scholasticae
([1 st ed. 1572J, Leipzig, 1726) p. 2424 (Sic apud astronomos Systema mundi dicitur universi constitutio, forma, ordo etc.). For the theological usage (besides summa, corpus,
syntagma, compendium etc.) see [91].
61 See Vitruvius Pollio, De architectura XI, 3 ([97] 460f).
62 De sphaera ([41]11). The concept of the world machine can be found no less than
three times in the very first thirteen lines, which begin with the statement: "lntentio
nostra in hoc tractatu est describere figuram machinae mundanae et centrum ..."
(]41] 11, 1-2).
63 Tractatus de sphaera I (Universalis autem mundi machina in duo dividitur), V in
fine (Aut deus nature patitur, aut machina mundi dissolventur) ([89]78, 117); on the
question of dating see ]89J 8ff. (Sacrobosco's De sphaera being earlier than Grosseteste's De sphaera). The distinction between mundus archetypus and mundus sensibilis
can also be found in Sacrobosco (]89J 80; see note 14 above).
64 Sphaera 10annis de Sacrobosco emendata (Cologne, 1606; Lyon, 1606; Paris, 1608;
Cologne, 1610; Venice, 1620; Leiden, 1656); Libellus de Sphaera (Wittenberg, 1531;
1629 (includes about eleven other editions; Sphaera (Paris, 1545; Leiden, 1647
(includes about nine other editions. The most important commentary, also published
in the seventeenth century in several editions (Rome, 1606; Lyon, 1607; Cologne,
1608; Lyon, 1618) was written by Christoph Clavius: In Sphaeram loannis de Sacro
Bosco commentarius (Rome, 1581) pp. 28f, pp. 446f (= Venice, 1591) pp. 28f, pp.
462f, cf. Praefatio (p. 5). For the history of the commentaries in the fifteenth, sixteenth
and seventeenth centuries see [94] 40ff.
65 Totius philosophiae, hoc est, logicae, moralis, physicae et metaphysicae brevis et
accurata ... tractatio, 2 vols. (Paris 1617, 2nd ed. 1622, 4th ed. 1627-8, 6th ed.
1637).
67 According to H. M. Nobis, who refers to the university calendar for the academic
years 1593-4 and 1599-1600 ([77J 45). In his Discorsi, Galileo himself draws
attention to his study of the Sphaera (]38]8, 101).
68 H. Monantholius, Aristotelis Mechanica, Graeca, emendata, Latina facta, et Commentariis illustrata (Paris, 1599) Epistola dedicatoria aUIr. This description is preceded
by the remark that according to Plato God always does geometry, but that Plato would
56
57
37
have better served God's majesty if he had said that God always does mechanics
too. According to Monantholius God is the wisest, the first and the most powerful
#TJxavt,,6~ and #TJxavolrOL6~ (ibid. aIVr). Cf. A. Piccolomini, Della sfera del mondo.
De Ie stelle fisse (3rd ed. Venice, 1552 (1559, 2r (gran Machina), 8r (Questa
Machina, che noi chiamiamo sfere del Mondo . .. ), 9v (gran machina del Mondo).
69 Machina coelestis, 2 vols. (Danzig, 1673-9) 1, 2f, 78. So also Christoph Clavius, In
Sphaeram Ioannis (see note 64 above), p. 28 (God as opifex mundi, again with
reference to Genesis and Timaeus). Even in general textbooks of philosophy the
concept of machina mundi gained wider acceptance. See P. Mako, Compendiaria
metaphysicae institutio (Vienna, 1761), 123; Compendiaria logicae institutio (Vienna,
1760), 94; J. B. Horvath, Institutiones logicae, et metaphysicae (Augsburg, 5th ed.
1781),2.
70 De revolutionibus I [ProoemiumJ ([23J 9).
71 De revolutionibus I [ProoemiumJ ([23]8).
72 De revolutionibus I [ProoemiumJ ([23J 8). The source is Plato, Tim.30d3.
73 De revolutionibus I 8 ([23]19).
74 Prodromus dissertationum cosmographicarum, continens Mysterium cosmographicum [l st ed. 1621J XX (Notae)([51J 8, 113).
75 Letter of 10 February 1605 to Herwart von Hohenburg ([51J 15, 146). Well known
are Descartes' analogous remarks on the background of the metaphysical dualism of res
cogitans and res extensa; see Principia philosophiae IV 203 ("... there are absolutely
no judgments [or rules] in Mechanics which do not also pertain to Physics, of which
Mechanics is a part or type: and it is as natural for a clock, composed of wheels of a
certain kind, to indicate the hours, as for a tree, grown from a certain kind of seed,
to produce the corresponding fruit" ([25J 8.1, 326; [26J 285f); compare also the
characterization of the monads as natural machines in Leibniz (De ipsa natura . .. ,
[57J 4, 505) ("every machina naturalis, conceived of as a part of natura universalis
which is again characterized as artificium Dei, is composed of infinitely many [simple]
instruments").
76 Tractatus de eommensurabilitate vel incommensurabilitate motuum eeli III
([79]294,117-20).
77 Letter of 4 October 1607 to Johann Georg Brengger ([51]16,54).
78 Prodromus dissertationum cosmographicarum, continens Mysterium cosmographicum [1st ed. 1596] II ([51J 1,26)(= [1621J II, [51]8,47).
79 [51J 1,6 [1596]; [51]8, 17 [1st ed. 1620] (English in [52J 53ff). On the geometrical
language of Keplerian cosmology in this context, see Harmonice mundi V 3 ([51]6,
299) (reference to a Platonic God doing geometry); IV 1 ([51J 6, 223) (geometry
precedes the origin of things); letter of October 1605 to Christoph Heydon ([51]15,
235); and letter of 12 May 1608 to Joachim Tanckius ([51J 16, 161) (limits of geometry
as limits of God's creation). On this topic, see [67J.
80 Mechanica 1.847al1-19 (English by E. S. Forster in [4J 6).
81 Mechanicorum tiber (Pesaro, 1577) (Praefatio). Monantholius already emphasises
the theoretical character of mechanics (Aristotelis Mechanica (see note 68 above),
Praefatio oIVr), though for him, too, physics and mechanics are two different things. In
this he follows A. Piccoiomini, whom he quotes (loc. cit., p. 9). Cf. A. Piccoiomini, In
Mechanicas Quaestiones Aristotelis paraphrasis (Rome, 1547) I1Ir.
82 Le mecaniche, [38J 2, 155ff. This work first appeared in a French translation by M.
38
JURGEN MITTELSTRASS
Mersenne (Les mechaniques de Galile'e mathematicien et ingenieur du Due de Florence

(Paris, 1634. Later it was published posthumously in an Italian version (Della Scienza
Mecanica . .. (Ravenna, 1649. In a lecture on mechanics in 1581-2 which was based
upon the Quaestiones mechanicae and has survived in manuscript form (Bibliotheca
Ambrosiana, Codex G 69 inf.), G. Moleti, Galileo's predecessor in office and his
patron, had already defended the thesis that mechanics does not outwit nature, as
taught by the Aristotelian tradition, but rather imitates her (see [78]).
H3 Letter of 7 May 1610 to Belisario Vinta ([38) 1O,35lf).
H4 See Christiaan Huygens, Traite de la lumiere 1[1678) ([48) 19, 461); G. W. Leibniz,
Theoria motus concreti (1671) para. 58 ([57) 4,210f).
H5 W. Emerson, The Principles of Mechanics (3rd ed. London, 1773) p. V.
H6 Dialogo sopra i due massimi sistemi del mondo II ([38) 8, 289; English in [37) 265).
H7 Dialogo I ([38) 7, 85) 7 (this remark, however, is attributed to the "Aristotelian",
Simplicio). Also compare Newton's explanation of his first "regula philosophandi"
which repeats this phrase (Principia, 3rd ed. 1726, p. 387).
H8 Discorsi e dimostrazioni matematiche intomo a due nuove science III ([38) 8, 197);
see also Dialogo II ([38) 7,143).
89 Principia philosophiae II 37 ([25) 8.1,62).
90 On this development see [68).
91 II Saggiatore ([38) 6, 232). The phrase already appears in the statutes of the
Accademia dei Lincei (quoted [31) 81).
92 De lineis angulis et figuris seu de fractionibus et reflexionibus radiorum, in [41) 60f,
d. [39] 26lff.
93 See [76) 404; also R. Hooke in [99) 146ff. Hooke's work A General Scheme, or Idea
of the Present State of Natural Philosophy and How Its Defects May Be Remedied
([45) 1-70) has the characteristic subtitle By a Methodical Proceeding in the Making
Experiments and Collecting Observations. Whereby to Compile a Natural History, as the

Solid Basis for the Superstructure of True Philosophy.
94 Les bijoux indiscrets (1748) in [27) 4,255ff.
95 Principia, 2nd ed. 1713, p. 483; 3rd ed. 1726, p. 529.
96 Lexicon Technicum or an Universal English Dictionary of Arts and Sciences . .. , 2
vols (London, 1704-10) (entry on Nature in 2).
97
Cf. the relevant titles in [30) and [63).
Tractatus de configurationibus intensionum III 7 ([20) 358f).

[14]5, 219; d. the Latin edition Tractatus de ipsa natura, sive libera in receptam
notionem disquisitio ad amicum (Genf, 1688) p. 71. Boyle's own distinction between
natura universalis and natura particularis ([14]5, 177) no longer has anything to do
with the old distinction.
100 Newton had accepted the fact that H. Pemberton in the introduction to his book A
View of Sir Isaac Newton's Philosophy (London, 1728) presented Newton's physics as
the realization of Bacon's ideas on methodology (see [66) 321).
101 Novum organum I 26 ([9]1,161).
102 Novum organum 1129 ([9]1, 221); Salomon, Proverbs XXV, 2.
103 Novum organum I 41 ([9) 1, 163f). Cf. [90) 215ff.
104 R. Harre characterizes Bacon's empiricist programme in the following way: "The
perceptual world allows no clear distinction between Natures as to their Reality. What
are the most real Natures will emerge from the empirical investigations of science.
98
99
39
Those Natures which occur as the end products of tabular inductions are the Forms of
Natures, and the total set of most real Natures are, for the purposes of science and
technology, the fundamental properties of matter" ([42]113).
\05 Principia (1st ed. 1687) pp. 12f. I. B. Cohen has drawn attention to the fact that "in
a letter to Roger Cotes, Newton unconsciously wrote about his own 'laws of nature', but
then crossed out 'nature' to make it 'motion'" (]21] 102).
106 Vorlesungen uber die Geschichte der Philosophie ([44]18, 342ff).
REFERENCES
1. Apuleius, L., Opera, 3 vols, ed. P. Thomas (Stuttgart: Teubner, 1970).
2. Aquinas, Thomas, In librum Beati Dionysii de divinis nominibus expositio, ed. C.
Pera (Turin and Rome: Marietti, 1950).
3. Aquinas, Thomas, Opera omnia, incomplete. Ed. Leonine (Rome, 1882-).
4. Aristotle, The Works of Aristotle, 12 vols, ed. W. D. Ross (Oxford: Oxford Univ.
Press, 1910-52).
5. Augustine, A, De civitate Dei, 2 vols, ed. B. Dombart (Leipzig, 1877-1905).
6. Augustine, A, Oeuvres de Saint Augustin, Ire serie [Opusclesl, 12 vols, eds. B.
Roland-Gosselin et al. (Paris: Desclee, de Brouwer et Cie, 1936-50).
7. Averroes, Aristotelis opera cum Averrois commentariis, 10 vols (Venice, 156274; repro Frankfurt: Minerva, 1962).
8. A vicenna, De subtilitate rerum (Basel, 1554).
9. Bacon, Francis, The Works of Francis Bacon, 14 vols, eds. J. Spedding, R. L.
Ellis, and D.D.Heath (London, 1857-74).
10. Baron, W. (ed.), Beitriige zur Methodik der Wissenschaftsgeschichte (Wiesbaden:
Steiner, 1967).
11. Bolgar, R. R. (ed.), Classical Influences on Western Thought A.D. 1650-1870
(Cambridge: Cambridge Univ. Press, 1979).
12. Bonaventura, G. F., Opera omnia, 15 vols, ed. A C. Peltier (Paris, 1864-71).
13. Borsche, T. and B. Hoppe, 'Natura universalis/particularis', in 18816, cols.
509-17.
14. Boyle, Robert, The Works of the Honourable Robert Boyle, 6 vols, ed. Thomas
Birch (2nd ed., London, 1772).
15. Bruno, Giordano, Le opere Italiane di Giordano Bruno, 2 vols, ed. P. de
Lagarde (Gottingen, 1888).
16. Calcidius, Timaeus, a Calcidio translatus commentarioque instructus, ed. J. A
Waszink (London and Leiden: Warburg Institute and E. J. Brill, 1962).
17. Campanella, T., Philosophiae rationalis partes quinque (Pars secunda: Logicorum
libri tres, iuxta propria dogmata) (Paris, 1638).
18. Campanella, T., Del senso delle cose e della magia, ed. A Bruers (Bari: Laterza,
1925).
19. Capella, Martianus, De nuptiis Philologiae et Mercurii libri, ed. A Dick (Stuttgart: Teubner, 1925).
20. Clagett, M., The Science of Mechanics in the Middle Ages (Madison: Univ. of
Wisconsin Press, 1959).
21. Cohen, I. B., 'Dynamics: The key to the "new science" of the seventeenth
40
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
JURGEN MITTELSTRASS
century', Acta historiae rerum naturalium necnon technicarum, Special Issue 3
(Prague: CSAV, 1967).
Cohen, I. B., Introduction to Newton's "Principia" (Cambridge: Cambridge Univ.
Press, 1971).
Copernicus, N., De revolutionibus orbium caelestium libri sex, eds. F. Zeller and
C. Zeller (Munich: Oldenbourg, 1949) (Gesamtausgabe, vol. 2).
Cusanus, Nicolas, Nikolaus von Kues. Werke. [Neuausgabe des Strassburger
Drucks von 1488], 2 vols, ed. P. Wilpert (Berlin: de Gruyter, 1967).
Descartes, Rene, Oeuvres de Descartes, 12 vols, eds. C. Adam and P. Tannery
(Paris: Vrin, 1964-76).
Descartes, Rene, Principles of Philosophy, tr. with notes by V. R Miller and R
P. Miller (Dordrecht: Reidel, 1983).
Diderot, Denis, Oeuvres completes, 20 vols, eds. J. Assezat and M. Tourneux
(Paris, 1875-7).
Diemer, A. (ed.), System und Klassifikation in Wissenschaft und Dokumentation
(Meisenheim: Hain, 1968).
Dieterici, F., Die Philosophie bei den Arabern im X. lahrhundert n. Chr., 14 vols
(Leipzig, 1884-6).
Dijksterhuis, E. J., De mechanisering van het wereldbeeld (Amsterdam: Meulenhoff, 1950).
Drake, S., Galileo Studies. Personality, Tradition, and Revolution (Ann Arbor:
Michigan Univ. Press, 1970).
Eckhart, J., Meister Eckhart (in [80J 2).
Ficino, Marsilio, Omnia divini Platonis opera, tr. M. Ficini . .. (Basel, 1546).
Fieino, Marsilio, Opera omnia, 2 vols (Basel, 1576).
Fieino, Marsilio, Theologia Platonica, 3 vols, ed. R Marcel (Paris: Societe
d'edition "Les belles lettres", 1964-70).
Fiorontino, F., Bernardino Telesio, 2 vols (Florence, 1872-4) (Documenti
inediti et rari).
Galilei, Galileo, Dialogue Concerning the Two Chief World Systems - Ptolemaic
and Copernican, tr. S. Drake (2nd ed., Berkeley: Univ. of California Press,
1967).
Galilei, Galileo, Le opere di Galileo Galilei, 20 vols, ed. A. Favaro (Florence,
1890-1909) (Edizione Nazionale).
Gilson, E., History of Christian Philosophy in the Middle Ages (New York:
Random House, 1955).
Goclenius, R, Lexicon philosophicum (Frankfurt, 1613).
Grosseteste, Robert, Die philosophischen Werke des Robert Grosseteste, Bischofs
von Lincoln, ed. L. Baur (Miinster: Aschendorff, 1912) (= Beitriige zur
Geschichte der Philosophie des Mittelalters 10).
Harre, R, Matter and Method (London: Macmillan, 1964).
Hedwig, K., 'Natura naturans/naturata', in [88J 6, cols. 504-9.
Hegel, G. W. F., Siimtliche Werke, 20 vols, ed. H. Glockner (Stuttgart:
Frommann, 1927-30).
Hooke, Robert, The Posthumous Works of Robert Hooke (London, 1705).
41
46. Hiibner, K. and A. Menne (eds.), Natur und Geschichte (X. Deutscher Kongress
fiir Philosophie Kiel, 8-12 Oktober 1972) (Hamburg: Meiner, 1973).
47. Husserl, E., Die Krisis der europaischen Wissenschaften und die transzendentale
Phanomenologie, ed. W. Biemel (2nd ed., The Hague: Nijhoff, 1962) (=
Husserfiana, vol. 6).
48. Huygens, c., Oeuvres completes, 22 vols (The Hague: Nijhoff, 1888-1950).
49. Jammer, M., Concepts of Space. The History of Theories of Space in Physics
(Cambridge MA: Harvard Univ. Press, 1954).
50. Kambartel, F., 'Naturgeschichte', in ]88] 6, cols, 526-8.
51. Kepler, J., Gesammelte Werke, eds. W. v. Dyck, M. Caspar, and F. Hammer
(Munich: Beck 1937-).
52. Kepler, J., Mysterium cosmographicumIThe Secret of the Universe, tr. A. M.
Duncan, intro. and commentary E. J. Aiton, preface I. Bernard Cohen (New
York: Abaris Books, 1981).
53. Krafft, F., 'Die Anfiinge einer theoretischen Mechanik und die Wandlung ihrer
Stellung zur Wissenschaft von der Natur', in ]10], 12-33.
54. Krafft, F., Dynamische und statische Betrachtungsweise in der antiken Mechanik
(Wiesbaden: Steiner, 1970).
55. Kristeller, P.O., The Philosophy of Marsilio Ficino (Gloucester MA: Smith,
1964).
56. Kristeller, P.O., Die Philosophie des Marsifio Ficino (Frankfurt: Klostermann,
1972).
57. Leibniz, G. W., Die philosophischen Schriften, 7 vols, ed. C. I. Gerhardt (Berlin
and Leipzig, 1875-90).
58. Leibniz, G. W., Samtfiche Schriften und Briefe, incomplete. Ed. Preussische
Akademie der Wissenschaften (Darmstadt and Leipzig: Akademie Verlag,
1923-).
59. de Lille, Alain, Alain de Lille. Textes inedits, ed. M. Th. d'Alverny (Paris: Vrin,
1965).
60. Lucretius Carns, Titus, Titi Lucreti Cari De rerum natura fibri sex, 3 vols, ed.
and tr. C. Bailey (Oxford: Clarendon Press, 1947).
61. Mach, E., Die Mechanik lin ihrer Entwickelung] Historisch-kritisch dargestellt
(9th ed., Leipzig: Brockhaus, 1933).
62. Micraelius, J., Lexicon philosophicum terminorum philosophis usitatorum (2nd
ed., Stettin, 1662).
63. Maier, A., Die Mechanisierung des Weltbildes im 17. lahrhundert (Leipzig:
Meiner, 1938).
64. Migne, J. P. (ed.), Patrologiae cursus completus, Series 2 (Ecclesia latina), 221
vols (Paris, 1841-64).
65. Migne, J. P. (ed.), Patrologiae cursus completus, Series 1 (Ecclesia graeca), 167
vols (Paris, 1857-1912).
66. Mittelstrass, J., 'The Galilean revolution. The historical fate of a methodological
insight', Studies in History and Philosophy of Science 2 (1972), 297-328.
67. Mittelstrass, J., 'Methodological elements of Keplerian astronomy', Studies in
History and Philosophy of Science 3 (1972), 203-32.
42
JDRGEN MITTELSTRASS
68. Mittelstrass, J., 'Metaphysik der Natur in der Methodologie der Naturwissenschaften. Zur Rolle phanomenaier (Aristotelischer) und instrumentaler
(Galileischer) Erfahrungsbegriffe in der Physik', in [46]63-87.
69. Mittelstrass, 1., "'Phaenomena bene fundata": From "saving the appearances" to
the mechanization of the world-picture', in [11]39-59.
70. Mittelstrass, J., 'Das Wirken der Natur. Materialien zur Geschichte des
Naturbegriffs', in [87] 36-69.
71. Mittelstrass, J. (ed.), Enzyklopiidie Philosophie und Wissenschaftstheorie,
incomplete (Mannheim, Vienna, and Zurich: Bibliographisches Institut, 1980).
72. Mittelstrass, 1., 'Experimentalphilosophie', in [71]1, 622-4.
73. Mittelstrass, J., 'F. M. van Helmont', in [71]2, 69-72.
74. Mittelstrass, 1., 'Naturgeschichte', in [71]2,967-8.
75. Mittelstrass, J., 'Natura naturans', in [71]2,966-7.
76. Newton, I., Opticks, or, A Treatise of the Reflections, Refractions, 1nflections and
Colours of Light (4th ed., London, 1730), eds. I. B. Cohen and D. H. D. Roller
(New York: Dover, 1952).
77. Nobis, H. M., 'Fruhneuzeitiiche Verstandnisweisen der Natur und ihr Wandel bis
zum 18. Jahrhundert', Archiv [iir Begriffsgeschichte 11 (1967),37-58.
78. Nobis, H. M., 'Dber zwei Handschriften zur friihneuzeitiichen Mechanik in
italienischen Bibliotheken', Sudhoffs Archiv 53 (1969), 326-32.
79. Oresme, Nicholas of, Nicole Oresme and the Kinematics of Circular Motion, ed.
E. Grant (Madison: Univ. of Wisconsin Press, 1971).
80. Pfeiffer, F. (ed.), Deutsche Mystiker des vierzehnten fahrhunderts, 2 vols (Leipzig,
1845-57; repr. Aalen: Scientia, 1962).
81. Philo of Alexandria, Opera, 6 vols, eds. L. Cohn and P. Wendland (Berlin,
1896-1915).
82. Philoponus, J., Physicorum libri quatuor (Venice, 1558).
83. Pico della Mirandola, G., Opera omnia, 2 vols (Basel, 1557).
84. Pico della Mirandola, G., De hominis dignitate, Heptaplus, De ente et uno, e
scritti vari, ed. E. Garin (Florence: Vallecchi, 1942) (Edizione Nazionale dei
classici del pensiero 1taliano, vol. 1).
85. Plotinus, Enneades, 6 vols, ed. E. Brehier (Paris: Societe d'edition "Les belles
lettres", 1924-33).
86. Randall, J. H. Jr., The Career of Philosophy. From the Middle Ages to the
Enlightenment (New York: Columbia Univ. Press, 1962).
87. Rapp, F. (ed.), Naturverstiindnis und Naturbeherrschung. Philosophiegeschichtliche Entwicklung und gegenwiirtiger Kontext (Munich: Fink, 1981).
88. Ritter, J. and Grunder, K. (eds.), Historisches Worterbuch der Philosophie;
incomplete (Basel and Stuttgart: Schwabe, 1971-).
89. de Sacrobosco, J., The Sphere of Sacrobosco and Its Commentators, ed. L.
Thorndike (Chicago: Univ. of Chicago Press, 1949).
90. Schmidt-Biggemann, W., Topica Universalis. Eine Modellgeschichte humanistischer und barocker Wissenschaft (Hamburg: Meiner, 1983).
91. von der Stein, A., 'Der Systembegriff in seiner geschichtlichen Entwicklung', in
[28/1-13.
92. Theiler, W., Zur Geschichte der teleologischen Naturbetrachtung bis aUf Aristoteles (Zurich: Fiissli, 1925; 2nd ed., Berlin: de Gruyter, 1965).
43
93. Thiel, Ch., 'Makrokosmos', in 17112, 748-50.

94. Thorndike, L., Michael Scot (London: Nelson, 1965).
95. Vico, G., On the Study Methods of Our Time, tr. E. Gianturco (Indianopolis:
Bobbs-Merrill, 1965).
96. Vincent of Beauvais, Speculum doctrinale (Duaci, 1624; repro Graz: Akademische Druck- und Verlagsanstalt, 1965).
97. Vitruvius Pollio, De architectura, ed. C. Fensterbusch (Darmstadt: Wissenschaftliche Buchgesellschaft, 1964).
98. Wallace, W. A., Prelude to Galileo. Essays on Medieval and Sixteenth-Century
Sources of Galileo's Thought (Dordrecht: Reidel, 1981) (= Boston Studies in the
Philosophy of Science, vol. 62).
99. Weld, C. R., A History of the Royal Society with Memoirs of the Presidents.
Compiled from Authentic Documents, vol. 1 (London, 1848).
100. Wolters, G., 'J. B. van Helmont', in 17112, 69-72.
101. Zeller, E., Die Philosophie der Griechen in ihrer geschichtlichen Entwicklung,
vol. 2.2 (4th ed., Leipzig: Reisland, 1921).
RICHARD S. WESTFALL
GALILEO AND THE JESUITS
Abstract. More than once, after his trial, Galileo asserted that the Jesuits were the
source of all his troubles. He was not alone in that judgment; it was the accepted view
of the age. Galileo's relations with the Jesuits are then a matter of some importance.
Without denying the fundamental issue of authority in questions of natural philosophy
that separated them, this essay explores how much the system of patronage, which
supported Galileo in his life in science, can help to illuminate their relations.
The essay insists on the centrality of the period 1615-16, the Inquisition's first
inquiry into Galileo and also the watershed in his dealings with the Jesuits. Until that
time the Jesuits at the Collegio Romano had been Galileo's steadfast supporters, crucial
figures in his rise to prominence. In 1616, as a result of charges brought against
Galileo, the Church condemned Copernicanism and placed several books on the Index,
but neither the Holy Office nor the Index touched Galileo in public. The essay argues
that the system of patronage had made him immune. It argues as well that the Jesuits,
reflecting on his apparent immunity and on his increasingly abrasi\e manner, encouraged by the system of patronage, decided then that it would be necessary to defend
their position of authority in the Catholic world and to cut Ga1ileo down to size.
Without pretending that there was not an important divergence of outlook between
Galileo and the Jesuits, this analysis moves the origin of their conflict from the realm of
intellectual differences to another context. It asks how Ga1ileo, the successful client,
impinged on others, and it asks how those with established authority in intellectual life
viewed the arrival of a new man whose genius was the more compelling because it
enjoyed the support of the patron class.
On 7 August 1632, Filippo Magalotti wrote to his friend Mario

Guiducci in Florence. Magaloui had recently arrived in Rome with
eight copies of Galileo's Dialogue, among the first copies of the book to
reach the Papal city because of an interruption of communications
caused by the plague. He reported that Niccolo Riccardi, the Master of
the Sacred Palace, the ultimate censor of the church and thus the man
who had licensed the Dialogue for publication, had hunted him out and
asked him to hand them over. It was too late. They were presentation
copies, and as Riccardi well knew, since he was the recipient of one,
Magalotti had already distributed them. Why did he make such a
request? Riccardi had evaded the query with a ridiculous objection to
the frontispiece, but Magalotti indicated to Guiducci that the overt
objection was the Dialogue's conclusion, which did not assert the
45
46
RICHARD S. WESTFALL
supremacy of faith over scientific reason vigorously enough. "This is the

pretext", Magalotti continued, "but the reality must be that the Jesuit
Fathers are working behind the scene with all their strength to have the
book banned, for he [Riccardi) himself said as much to me in these
words: The Jesuits will pursue him with extreme bitterness" ([6) 14,
368-70). Magalotti's letter was among the earliest indications Galileo
received that trouble was brewing, trouble that would lead him during
the following year into the Inquisition's prison, impose on him a public
abjuration of Copernicanism, and condemn him to house arrest for the
remaining decade of his life. The trial of Galileo was one of the epochal
events of the seventeenth century, an enduring symbol of the tension
between two different systems of thought as modem science began to
displace Christianity from the focus of Western civilization. And in
Magalotti's letter, which is among the earliest words pointing directly
toward it that we have, the Jesuits stand at the centre of attention.
Galileo never ceased to believe that the Jesuits had been the leading
authors of his troubles. l He was not alone. It was the accepted wisdom
of the day, a judgment endlessly repeated. 2
Manifestly, Galileo's relations with the Jesuits are a historical question
of some importance. It should be clear, first of all, that a divergence of
belief about ultimate intellectual authority and the role of Christian
revelation in science did separate them. Consider the implications of a
letter that Galileo received from Guiducci late in May 1633. Late May
was the very climax of the trial. As Galileo was being tormented (not
physically, to be sure) by the Inquisition in Rome, his native Florence,
and the whole of northern Italy, were being tormented by the plague,
the same visitation of the plague familiar to many from the pages of
Manzoni's 1 promessi sposi. "In regard to health", Guiducci reported,
the situation is slowly improving, and there have not been as many deaths as we have
written about [earlier]. This morning the Holy Madonna of Impruneta was brought to
Florence, and she will stay here until Monday. The hope vested in this image, which has
always been so miraculous in succouring this city, is very great, and the people have
conceived the greatest hope that through the intercession of the Holy Virgin we will
remain free [of the plague]. May it please the Lord God that we have prepared
ourselves in such a way as not to be unworthy of so much assistance.]
Who can doubt that we stand

Western civilization passed (not
worldview, represented here by
another, represented by Galileo
before an historic threshold whereby

with a single step, of course) from one
the Inquisition and the Madonna, into
and modem science? Who can doubt,
47
for that matter, that the Jesuits perceived more clearly than Galileo
what the ultimate impact of modem science upon Christianity would
be? And if it is understanding we pursue, we must not permit ourselves
to wonder that the Jesuits were ready to defend that world in which
they lived and moved and had their being.
I am convinced that an issue of belief, associated with that transformation of worldview, lay at the very core of the trial of Galileo. Great
historical events have many dimensions, however. Galileo and the
Jesuits pursued their evolving relations within the context of a specific
society in which the system of patronage supported all of the higher
culture as it supported Galileo himself. Without in any way denying the
importance of the issue of belief, I seek here to explore how far the
perspective provided by the system of patronage can help to illuminate
Galileo's relations with the Jesuits. My phrase is "help to illuminate".
Galileo and the Jesuits lived in a complex society in which many factors
beyond the question of belief to which I have referred and the system
of patronage contributed to shape men's actions. The best one can do is
to isolate for examination individual features without forgetting that
they were single strands in an intricate web. My intention is to inquire
how far the system of patronage can help to illuminate a matter as
complex as epochal events are wont to be.
Galileo burst on the intellectual scene in 1609 when he was already
forty-five years old. To those who had known him, his sudden emergence was no surprise, for he had been a witty conversationalist and an
exhilarating teacher, one of those rare men capable of galvanizing his
companions with his brilliance. Nevertheless, beyond two narrow
circles in Venice-Padua and Florence, his very existence had scarcely
been known. Except for a small pamphlet describing the operation of a
calculating device he had improved, and a later defence of his priority
with the device, he had not published anything, and in that age as surely
as now, those who did not publish perished. Then, in 1609-10, came
the telescope and the observations of the heavens announced in the
Sidereus Nuncius. It is not possible, I am convinced, to overestimate the
impact of the telescopic discoveries. Galileo was revealing the existence
of a wholly new world, a new world not only unknown but not dreamed
of before, even by the ancients. Correspondents vied with each other in
plundering history and mythology for the image that could appropriately describe its discoverer - Columbus, Magellan, Atlas. Ontioco
Bentivogli, who admitted that he had not originally believed the claims,
48
RICHARD S. WESTFALL
found himself forced to confess that Galileo, "not only like Linceus, but
like another Prometheus, has truly mounted to the heavens and penetrated their most profound secrets; so that I thank God that by your
means He has been pleased to let me also participate in the knowledge
of things so rare and concealed for so many ages".4 After his visit to
Rome in 1611, Galileo received a letter from his friend Paolo Gualdo,
citing a letter Gualdo had received from Mark Welser in Augsburg.
Welser had described in detail a banquet given to honour Galileo in
Rome. As Gualdo commented, Galileo was now watched and reported
throughout the world. 5
The telescope and the Side reus Nuncius were not, of course, the end
of Galileo's achievement. He quickly consolidated their promise with
two more books, and later there would be much more. To us, looking
back from more than three and a half centuries, the discoveries with the
telescope seem almost the least of Galileo's accomplishments. This was
not the case in his own age. The great works we think of immediately
when his name is mentioned, the Dialogue and the Discourses, were the
products of the final years of a long life, and their impact was felt
primarily by later periods. In his own time, Galileo's badge of genius
was the telescope. Instantly it transformed him into the most celebrated
intellectual of Italy, and probably of Europe. And it transformed him as
well into one of the most desirable clients, a man likely to bring glory to
the patron who supported his activity.
It is essential to my theme that Galileo comprehended the system of
patronage and knew how to exploit it. He demonstrated his knowledge
of the system from the very beginning, with the Side reus Nuncius. That
treatise was literally born on the night, 11 January 1610, when Galileo
realized that the stars he had discovered near Jupiter were satellites
circling the planet. As we know from his correspondence, he had been
seeking to leave Padua, to escape from the burdens of teaching, to find
leisure to pursue his research and to write; and as we also know, he had
concluded that only the patronage of an absolute prince could supply
him with these benefits. Galileo named the satellites for the ruling
family of his native Florence and composed the Side reus Nuncius,
which he dedicated to the Grand Duke, to proclaim their existence to
the world. For his pains, he was appointed Professor of Mathematics at
the University of Pisa, without obligation to teach or to reside in Pisa,
and Mathematician and Philosopher to the Grand Duke, with an annual
stipend of one thousand scudi. To the court in Florence, Galileo
49
expressed the fear that the rest of the world might doubt the existence
of the Medicean stars if they had to accept them on his authority alone.
In order for the Grand Duke to gain the maximum esteem from the
new celestial bodies that bore his name, it was essential for others to see
the satellites for themselves. Hence he had prepared a number of
excellent telescopes for the other rulers of Europe, so that they might
know by their own eyes that the Medicean planets were real, though he
would send the telescopes only at the Grand Duke's orders, an inspired
manoeuvre whereby Galileo enlisted the Grand Duke as his public
relations agent. 6
With his following publications, Galileo took care to send copies to
the wealthy and powerful, especially to Cardinals of the Church. In the
twentieth century, scholars advertise their achievements by sending
presentation copies to academic colleagues. In the early seventeenth
century, under the system of patronage, Galileo sent presentation
copies to potential patrons. As far as possible, he timed his visits to
Rome to maximize their effect. Thus he paid his first visit (after the
advent of the telescope) only when he had a new discovery of great
importance, the phases of Venus. The following visit, of which we shall
hear, was not entirely spontaneous; but on the next one, in 1624, he
carried microscopes with him, one of which we know he presented to
Cardinal von Zollern for the Duke of Bavaria. "I saw a fly that Signor
Galileo himself arranged for me to see", Johannes Faber, a prominent
member of the Accademia dei Lincei reported to Federigo Cesi, the
Principe of the Accademia. "I was astonished, and said to Signor
Galileo that he was another Creator since he made things appear that
until now were not known to have been created." 7 On responses such
as that a career could rest secure.
During the critical years of Galileo's career, however, microscopes
and books that he published were secondary. The telescope remained
the foundation of his enormous reputation. Until the late 1630s, not
many years before his death, Galileo's telescopes remained the best that
were made, and he continued to receive requests for them until the end
of his life. Some of the instruments went to fellow investigators; for
example, Gassendi sought and obtained one, though he was careful to
indicate that Peiresc was involved in the request. 8 The overwhelming
majority of them, however, went to men of the patron class, like
Peiresc, rather than other scientists. Thus, toward the end of 1629,
Galileo received a request for a telescope for Phillip IV, the King of
50
RICHARD S. WESTFALL
Spain, with the information that the court insisted on paying, without
further ado, whatever price Galileo set. At that time Galileo was
completing the Dialogue. Nevertheless, he immediately undertook the
preparation of an instrument suitable for the King of Spain. As to
payment, he humbly begged to inform His Majesty that he could not
similarly serve him. He had never, Galileo stated, accepted payment for
a telescope, and he never intended to.9 From the evidence I have seen,
the assertion appears to have been true. Do not be misled to conclude
that the action was disinterested. The adventurer from Flanders who
came to Venice with the instrument that launched Galileo's career had
tried to sell it to the Senate. Galileo had presented his improved version
to the Doge and thereby made his fortune. He never forgot the lesson.
In such particulars we begin to perceive the material circumstances
within which Galileo was able to pursue a life of science that could not
easily have supported him otherwise in the early seventeenth century.
He could, it is true, have stayed on in Padua - though a university
chair, which demanded periodic reappointment, with decisions made by
men of precisely the same dominant class, differed less from the world
of patronage than one might at first think. Galileo's decision to leave
Padua, the decision of a man without personal means, was a decision to
throw himself wholly into the world of patronage. Activities such as the
presentation of books and of telescopes to men of wealth and power
were the social reality of the day, and by accepting their necessity and
conforming to the limits they set, Galileo was able to realize his
magnificent achievement.
We can contemplate in this respect a letter that Galileo received
from Benedetto Castelli in 1638. Castelli had been in Rome for more
than a decade in the service of the Barberini, the Papal family, and
through their influence he had been appointed to the chair in mathematics at the university in Rome. In 1638, when a death vacated the
chair in mathematics at the University of Pisa, where Castelli had once
taught with distinction, the Grand Duke attempted to lure him back.
Castelli was anxious that he not offend the Grand Duke, but he also
could not accept his proposal. In his letter of 30 July, he explained to
Gaiileo why he could not. He was, Castelli reminded Galileo, a member
of the Cassino Congregation of the Benedictine Order, subject to the
"protection" of Cardinal Francesco Barberini, the powerful nephew of
the Pope, and he was as well in the personal service of the Cardinal. "I
have no way to free myself from this place", Castelli continued,
51
"without the danger of ruining my affairs in a way that I could never

reverse". Six months earlier, Castelli's enemies had sent a monk to
Rome to seek his professorship, a man who had some qualifications
and warm recommendations to Cardinal Barberini. Only on the day of
the letter had Castelli received his reappointment; he had to walk with
caution and not give any offence.!O Castelli's letter is a perfect expression of the dependence of a client on the good will and pleasure of his
patron, even in the case of Galileo, whose achievements supplied him
with some leverage. The relation between patron and client was not
solely economic. It involved recognition, prestige, and standing in
society. But along with these benefits the relation was also economic;
and men such as Castelli and Galileo, who did not possess personal
wealth, were not at liberty to ignore the economic realities. Galileo was
attractive to the Grand Duke in direct proportion as he was attractive
to the class of patrons as a whole. The Grand Duke displayed his
magnificence by supporting a client with whom every patron would
have been delighted to see his name associated. Requests for telescopes
from Cardinals and from the King of Spain were of maximum utility to
Galileo. They were absolutely necessary. Without them he was nothing.
Galileo understood as much and acted accordingly.
Meanwhile, he was not the only hopeful man making his way in early
seventeenth-century society, and other potential clients were not necessarily as exhilarated at Galileo's success as he was. Galileo himself was
always convinced that their envy and malice, as he repeatedly expressed
it, was the source of all his troubles. There appears to have been a large
measure of truth in his assessment.
Soon after he arrived in Florence, Galileo became engaged in a
dispute at some social gathering. Apparently he was the aggressor.
Anxious as he was to display his brilliance, he was only too ready to
pounce on a statement by an Aristotelian philosopher that ice is
condensed water, and the discussion quickly expanded into a debate on
why bodies float. Galileo's opponents were convinced that he was
wrong and that they could use the incident to destroy his credibility
with the Grand Duke. The leading spokesman of the group was a
lecturer at the University of Pisa named Ludovico delle Colombe.
"Colombe" is the Italian for "dove"; hence one finds references in
Galileo's correspondence at this time to the League of Pigeons, that is,
Galileo's opponents. l !
Out of the dispute came Galileo's second major publication, the
52
RICHARD S. WESTFALL
Discourse on Bodies in Water, a treatise which brought the mathematical principles of Archimedean hydrostatics to bear on the categories of
Aristotelian natural philosophy. I wish to follow this controversy only in
so far as it concerns the question of patronage. Four replies to Galileo's
Discourse appeared. All of them were composed by academics at the
University of Pisa. All of them were dedicated to one or another
member of the Medici family. According to one assertion by Galileo,
their authors bound themselves together (hence the notion of a league),
before his own Discourse appeared, to oppose whatever he might say in
itP That is, by every indication, we have to do with a struggle within
the system of patronage. As Arturo d'Elci, the Overseer of the university, who wrote one of the attacks under the pseudonym of ''The
Unknown Academician", expressed it, Galileo was "greedy for glory",
and d'Elci clearly feared that Galileo's assertiveness was bound to
infringe on the welfare of established clients like himself.13 A further
indication that the controversy arose out the issue of patronage appears
in a letter from Castelli. The time of these events preceded the letter
from Castelli that I quoted above by twenty-five years; Castelli had just
been appointed to a professorship of mathematics at the University of
Pisa. In one of his earliest letters to Galileo from Pisa, he sent the
greetings of an anatomist, Ruschio, who suggested very strongly (as
Castelli put it, "in block letters") that Galileo was not envied
for the great and marvellous achievements of his intellect, matters that do not fall within
the knowledge or the consideration of these malicious men, but for those thousand
scudi [Galileo's stipend from the Grand Duke], which are probably better known by
them and more avidly desired than by yoU. 14
Let me pause briefly with the controversy over bodies in water

because it revealed two features of Galileo that are relevant both to
patronage as a whole and to his later relations with the Jesuits. The first
of these features is Galileo's ego. To describe it as king-sized is
seriously to understate it. There is a passage, for example, from the
Reply to his critics, which was published in 1615 without an author's
name but with Castelli's signature attached to the dedication. From the
surviving manuscript, we know that Galileo composed most of it,
though Castelli composed its opening section, which Galileo then
revised thoroughly. From the manuscript we can identify the author of
each individual phrase. Thus, at one point, Castelli wrote that Colombe,
in trying to destroy Galileo, has in fact bestowed the highest praise on
53
him, for Colombe celebrates the discoverers of new things and compares them to the Gods, and Castelli added that among the discoverers
of new things Galileo was, by common agreement, appropriately
included. Galileo amended the passage by the insertion of one adverb,
"meritamente", so that it now read: "and properly compares them to the
Gods ...".15
Add to his ego, perhaps as one facet of it, his inability to carry on a
controversy in a civil manner. Colombe's pamphlet against Galileo was
hardly a model of civility, and one can scarcely wonder that he chose to
reply in kind. The other three objections to Galileo's treatise, however,
remained properly on a plane of philosophic discourse and refrained
from any appeal to personal invective. One cannot, alas, say the same
about Galileo's reply. It was addressed throughout ad hominem. Its
basic instrument was ridicule. It is impossible, in my opinion, to read it
without concluding that Galileo's aim was less to establish the truth of
the issue under discussion than absolutely to humiliate his opponents.
"There is no point", he exclaimed, "in undertaking to refute someone
who is so ignorant that it would require a huge volume to refute his
stupidities (which number more than the lines of his essay) ...".16
Perhaps there was no point in the exercise, but he performed it
nevertheless. To pamphlets of 50 pages he replied with a volume of
350. A grenade had been exploded outside his door; for his response,
nothing less than a bomb could assuage his outrage.
These are not easy matters for an historian of science to discuss.
There is no doubt in my mind that Galileo had good cause for his ego.
He was one of the handful of rare geniuses that have graced mankind.
His four opponents in the controversy were not even close to the same
calibre. My goal at this point, however, is not to assess relative merits
but to comprehend Galileo's impact on his contemporaries. We have a
word for his conduct, and we should not fear to use it. Galileo was
insufferable. And, what is important to my argument, the characteristics
that made him insufferable were not solely personal. They were also
intimately related to patronage. Where certain institutional arrangements lead one to disguise and suppress his egotism and contempt for
others, patronage encouraged these characteristics in Galileo. It does
not seem much too strong to say that patronage demanded them. It is
relevant, surely, that humanistic controversies of that period, involving
men who worked within the same system, frequently displayed similar
features. And it is revealing that Galileo's young followers, identifying
54
RICHARD S. WESTFALL
with his status as the Grand Duke's philosopher, quickly learned to ape
his manners. Thus Niccolo Aggiunti lamented the "calumnies and
impostures of malicious men" that afflicted Galileo; and without reflecting apparently on what generated malice toward Galileo, went on
to decribe Chiaramonti's book on the stability of the earth, which in
fact he had not even seen yet, as "this excrement of a melancholy
humour".17 Galileo was attractive as a client exactly in so far as he was
perceived as a godlike discoverer of new things and a brilliant philosopher who easily outshone his opponents in philosophic discourse. The
praise heaped upon him by men of the highest rank, their gifts, his
special appointment by the Grand Duke, did nothing to diminish his
self-esteem. Egotism may have come easily to Galileo, but he received
every possible stimulus to behave as he did.
While the controversy on bodies in water was still in progress,
Galileo became involved in a second one, which did not address itself
to the Tuscan scene but nevertheless involved matters that bear on my
topic. The opponent in the new controversy, which concerned sunspots,
was a German who wrote under the pseudonym of Apelles. Their
differences involved the interpretation of the spots seen on or in front
of the sun, with manifold opportunities for sarcastic thrusts like those
mentioned above; and they included as well the question of priority in
the discovery, with all of the potential for bitterness that such questions
seem usually to entail. Galileo eventually learned that Apelles was
Christopher Scheiner, a Jesuit whom the order called to its Collegio
Romano in 1624. Although Scheiner does not seem initially to have
looked upon Galileo as an enemy, he was not one quietly to swallow
ridicule, and when he did learn to hate Galileo, he remained implacable
until his death.
Soon there was a third controversy as well. Like the first, this one
began in Tuscany, though it soon spread beyond. In December of 1613,
the court was in Pisa, and Castelli, then in the early months of his
appointment to the university, was present at a state dinner. It is surely
significant that the incident took place at the court, a firm indication
that the controversy began as another attempt to undermine Galileo
with his patron. The Grand Duchess Christina of Lorraine, the widow
of Ferdinand I and the mother of Cosimo, was a very pious woman too pious by twice for Galileo's good. During the dinner a cleric was
whispering in her ear. When Castelli was about to leave, he was called
back and forced into a discussion of the Copernican system and
55
Scripture, a discussion in which it was clear to all that Castelli stood as

the representative of Galileo. There had been suggestions before that
Galileo's enemies might employ the theological weapon against him.
Late in 1611, Cigoli reported hearing a rumour that his opponents,
looking for any device they could use against him, had tried to enlist a
preacher to attack Galileo from the pUlpit. 18 The following summer, as
he was writing the Letters on Sunspots, which did not mince words in
their assault on Aristotelian philosophy, Galileo raised a query with
Cardinal Conti, who at that time presided over the Index, as to whether
the Scriptures demanded Aristotelianism. 19 The query came six months
before the ecclesiastical censors forced Galileo to delete a discussion of
Scripture before they would licence the book, and thus it implies that
Galileo had been hearing theological objections elsewhere. Still later, at
the end of 1612, he received a report that Niccolo Lorini, who would
figure prominently in the later controversy, had attacked him on the
matter of Copernicanism and Scripture. 20 With the dinner at the court
in Pisa in December 1613, the issue was directly joined.
Castelli wrote to Galileo about the incident at once and received in
reply a long letter discussing the question Castelli had been forced to
debate, a letter that functioned as the first draft of the well known
"Letter to the Grand Duchess Christina".21 The Bible is the word of
God, Galileo argued, and as such it cannot err. Interpreters of the Bible
are human, however, and prone to error like all human beings. Nature
is also the revelation of God, and nature is its own interpreter. It is
immutable and never errs. Fallible men must not endanger the Church
by committing it on matters of natural philosophy, which it is not the
purpose of the Scriptures to teach, where scientific reason may one day
demonstrate the truth of conclusions that they have declared to be false
from the evidence of Scripture. What really made Galileo insufferable
was the clarity of his understanding, in this case greater clarity than the
combined assembly of theologians pursuing their own discipline.
For a year, nothing more happened, at least nothing overt, though
apparently copies of the letter to Castelli circulated. Galileo, who never
thought poorly of his own compositions and had the original letter after
Castelli returned it, must have been responsible for the copies. Then, in
December 1614, a Dominican friar, Tommaso Caccini, was preaching
in Florence on the book of Joshua. When he came to the famous passage, "Sun, stand thou still", he launched into a denunciation of Galileo,
Copernicans, and mathematicians in general. At this point, according to
56
RICHARD S. WESTFALL
Caccini's later deposition, Lorini, who was also a Dominican, brought

him a copy of Galileo's letter to Castelli. It is difficult to believe this
order of events, since the passage from Joshua that served as Caccini's
springboard for the initial attack, figured prominently in Galileo's letter
to Castelli. Lorini, who also taught at the University of Pisa, appears to
have been one of the primary instigators of the assault, a man busy in
spreading rumours about Galileo. He may well have taken Caccini the
copy of the letter to Castelli, but surely before Caccini's sermon. In any
case, Caccini denounced the letter to the Inquisition in Florence. When
nothing happened, Lorini sent the letter to the Inquisition in Rome in
February 1615.
This was no joking matter. Galileo was literally beside himself with
anxiety, as he had good reason to be. From the record of the Inquisition
we know that they pursued the charge, taking depositions. 22 For his
part, Galileo mobilized his influential friends in Rome. The details at
this stage need not detain us. Suffice it to say that late in 1615, as the
case neared its climax, Galileo travelled to Rome to defend himself. He
remained in the Papal city for nearly six months. It appears to me that
those six months must have been crucial in his relations with the Jesuits.
The period in question, from December 1615 to May 1616, came
after six years of the most spectacular success on Galileo's part. At this
point the trajectory of his career, mounting almost vertically, intersected another trajectory, that of the Society of Jesus. It is essential to
understand that their success had been equally spectacular. The Society
of Jesus had been founded three quarters of a century earlier, in 1540,
seventy-five years of almost uninterrupted triumphs. As one historian
of the order has asserted, the Jesuits had literally changed the course of
history ([11]128). When the order was founded, the Catholic Church
was falling back in disarray before the Protestant onslaught. By 1615,
the momentum had reversed its direction, and the only question left to
be decided was whether Protestantism would survive the counterattack.
More than any other factor, the Jesuits had determined that change of
direction in the course of events. Within the Catholic world their
prestige was enormous.
Moreover, the Jesuit enterprise was uniquely educational and intellectual. Such had not been Loyola's original conception of the order he
established. He had thought of the Jesuits as the shock troops of the
Pope in the battle against infidelity, wherever its challenge might be. In
the middle of the sixteenth century, there was not much doubt about
57
the principal challenge. It was the Protestant movement in Europe. In

relation to Protestantism, Loyola and other early Jesuits perceived
almost at once the necessity of an educational effort, for it was exactly
on the academic front that the Catholic religion was losing the battle.
The original Jesuit schools appeared during the first decade of the
order's history, and the undertaking underwent an incredible expansion.
In the early seventeenth century, more than three quarters of the Jesuit
houses were schools and more than four fifths of the Jesuits themselves
were teachers. By 1615, the Jesuits operated more than three hundred
colleges, not colleges in the twentieth-century American use of the
word, but rather gymnasia, though more than a handful of them had
university level divisions ([10]201-2; [12]57-8; [3]41-2). By the
seventeenth century, the Jesuits had established an educational and
intellectual hegemony within the Catholic world.
Special significance attached to one institution in the Jesuit educational enterprise, the Collegio Romano. Loyola himself had founded the
Collegio Romano in 1551 to be at once the crucible in which Jesuit
educational practices would be tried and the show piece of the order's
educational effort. To the Collegio Romano the order called its best
minds. Here Francisco Suarez, the greatest of the neoscholastic theologians, taught for a number of years. Here Roberto Bellarmino, the
greatest Catholic apologist, pursued most of his career. Suarez and
Bellarmino were by no means the only men of repute. In the early years
of the seventeenth century, the Collegio Romano was the most prestigious university in the Catholic world, or better, perhaps, in the
Christian world ([3]75-7; [15] passim).
It is necessary to add, moreover, that Jesuit success in the educational and intellectual world had not been tied to a dogged reassertion
of tradition but rather to a ready embrace of what in the new appeared
to be good. This was so especially in regard to the humanities. Loyola,
and the Jesuit order following him, rejected the concept of a tension
between the new learning and Christianity, and thought rather in terms
of a full accommodation between the two, indeed of a fusion of them
into a true Christian humanism. By the end of the sixteenth century, the
Jesuits had become the primary teachers of the humanities in Catholic
Europe. Nor had they excluded the exact sciences from their domain.
Jesuits schools placed more emphasis on mathematics and applied
mathematical sciences, such as astronomy, than any educational system
before them had done. When Gregory XIII had set out to reform the
58
RICHARD S. WESTFALL
calendar, he had turned to an astronomer at the Collegio Romano,

Christopher Clavius, who became the architect of the Gregorian
calendar ([3] passim, but especially 14-19, 222-8, 361-71).
It does appear to me that the Jesuit concept of a fusion of the
traditional and the new bore the seeds of a potential conflict with
Galileo, who was convinced that the old order, as far as natural
philosophy was concerned, had simply collapsed and needed to be
replaced. The Jesuits sought to employ the new learning to bolster the
old. In natural philosophy, they were wedded to Aristotle, whose
philosophy they saw as the foundation of Scholastic theology, the
intellectual rampart of the Catholic religion. Galileo, in contrast, was
the vitriolic critic of Aristotelian natural philosophy throughout most of
his career. Their difference in outlook was a ready source of probable
friction. In this essay, however, I am concentrating on a different level
of interaction. With the Jesuits the issue was not, as it was in Florence,
a competition for largesse. The Jesuits' ability to command resources
within the Catholic world was beyond challenge, and I have not seen
any evidence to suggest that Galileo's competition worried them in this
respect. As I indicated above, patronage involved prestige and standing
as well as money. What seems crucial to me is the fact that Galileo
saw himself as the prophet of the new intellectual order in natural
philosophy. The Jesuits, on the other hand, against the background of
their own history of success, were not inclined to sit quietly at lessons
under any human master.
The period of six months in 1615-16 was not Galileo's first contact
with the Jesuits. Far from it. By that time his relations with them
already stretched back over a quarter of a century. The earliest
surviving correspondence of Galileo is an exchange of letters with
Christopher Clavius, and subsequent letters indicate that the two had
stayed in contact. 23 Moreover, recent scholarship has established the
deep influence of the Collegio Romano on Galileo. The scholarship to
which I refer appears to me to constitute the most significant step
toward a fuller understanding of Galileo taken during the last decade.
We have long known about some early manuscripts, which were
partially published in the edizione nazionale under the title, "Juvenilia"
([6]1, 15-177). As a result of recent studies, we now recognize that
these manuscripts depended directly on the lecture notes of professors
of philosophy at the Collegio Romano, and because we can date the
manuscripts from what we know about the faculty at the Collegio, we
59
know that they were not juvenile exercises. Rather they were the work
of a mature young scholar who was either already a professor at the
University of Pisa or at least on the verge of assuming that position, and
who turned to the philosophers of the Collegio Romano to enhance his
understanding of logic and natural philosophy ([16], [1]). No doubt
there will be a long discussion about the interpretation of these
findings; the findings themselves appear to be established beyond the
possibility of rejection. On the basis of what has been published so far, I
am impressed by the case that has been made for the enduring
influence of the Collegio Romano on Galileo's outlook in the allied
fields of epistemology and method; I am less impressed with the present
case for their enduring influence on his natural philosophy. For this
essay, all that matters is the solidly substantiated conclusion that at
the beginning of his professional career, Galileo sought out the Jesuit
professors of the Collegio and willingly placed himself at their feet.
In 1610, Galileo published his short tract, Sidereus Nuncius, which
asserted the existence of hitherto unknown bodies in the heavens, on
the basis of observations made with an instrument that no-one else had.
Not surprisingly, there was scepticism about his assertion, and not
surprisingly there were some who claimed that the new celestial bodies
were delusions generated by the instrument itself. It was the Jesuit
astronomers at the Collegio Romano who obtained a telescope, confirmed Galileo's observations, and said as much publicly. Their support
was a critical factor in his initial success.
When Galileo came to Rome in 1611, the Collegio organized a
special assembly in his honour with a discourse on his discoveries. In
addition to the learned community of Rome, a number of counts, dukes
and prelates, including at least three Cardinals, attended the assembly,
which was little short of an official laying on of hands. Late in 1614, the
French cleric, Jean Tarde, visited the Collegio Romano. He recorded in
his diary that the Jesuit fathers accepted everything Galileo had discovered, including apparently the sunspots, although they agreed that
this evidence of mutability in the heavens was the source of some pain
([6]19, 591-2). Tarde's visit to Rome fell on the very eve of Galileo's
denunciation to the Inquisition. The denunciation was not the work of
the Jesuits, who appear on the whole to have been embarrassed by the
outburst of Dominican fundamentalism. At least one expression of
private support from their number has survived, and among Galileo's
letters there are numerous indications that he considered them to be his
60
RICHARD S. WESTFALL
allies. When fear that the copy of his letter to Castelli sent to Rome by
Lorini had been altered to his disadvantage led him to send a copy he
knew to be authentic, Galileo specified more than once that he wanted
Father Grienberger, the successor to Clavius, to see it. 24 The Jesuit
order was not so authoritarian that it contained no diversity of opinion.
Nevertheless, I find the evidence very strong that at this stage the
members of the Society in Rome who were most influential in such
matters were willing to accept Galileo's discoveries. Without denying
the basic difference in outlook mentioned above, I am convinced that
the origin of the conflict between Galileo and the Jesuits did not lie in
Galileo's discoveries themselves, despite the unsettling implications he
drew from them.
Against this background consider a lecture delivered at the Collegio
Romano two years later, on the three comets that had appeared in the
heavens in 1618. The man who delivered the lecture was Orazio
Grassi, the professor of mathematics at the Collegio. Grassi was the
Signor Sarsi of II Saggiatore, where Galileo reduced him to an object of
derision forever. But Grassi was not an object of derision. He was one
of the most respected scientists of the Jesuit order, who occupied the
chair in mathematics at the Collegio Romano for more than ten years.
When the canonization of Loyola led the Collegio to plan a new
church, St. Ignatius, intended to be second in magnificence in Rome
only to St. Peter's, it designated Grassi as the architect. Contrary to the
implication of Galileo's ridicule, Grassi was clearly a man of ability
much respected by his age.
His lecture on comets did not explicitly mention Galileo. He
obviously referred to Galileo several times, however, and the references
were all respectful. Nevertheless, Galileo regarded the lecture as an
attack, and it appears to me that Galileo was correct. The result of the
Church's actions in 1616 (to which I shall return) had been to separate
the realms of science and faith. As Bellarmine stated several times, it
was permissible to accept the Copernican system as a superior mathematical hypothesis. What was not permissible was the assertion that
Copernicanism is true, for passages in Scripture appeared to be in
contradiction to the heliocentric system. Increasingly, moreover, in the
opinion of the learned community Galileo had established himself as
the recognized master of the separate realm of science. What was the
import of Grassi's lecture? Grassi did not build his argument on
Scriptural citations. Rather he confined himself to the domain of
61
science and attempted to use the evidence of the comets to demonstrate

scientifically that the earth is at the centre of the universe - that is, that
the very position with which Galileo, in his numerous assertions about
the system of the world since 1610, had virtually identified his reputation in science was wrong.~5 Grassi even used the evidence of telescopic
observations, invading Galileo's unique territory and turning his own
instrument against him. Thus when Giovanni Rinuccini reported the
lecture to Galileo, he added that the Jesuits (and he did use their name
in the plural) were claiming that the lecture destroyed the basis of
CopernicanismY' As Galileo would state his reaction to Grassi's
endeavour later, he was "desirous of stripping me and entirely despoiling
me of every ornament of glory ... transported by the desire to obscure
my name in the field of science".27 In his report to Galileo, Rinuccini
also added that Galileo would comprehend how much he awaited his
opinion about the comets - surely the suggestion that someone beside
Galileo saw the lecture as an attack, and an attack not just by Grassi
but by the Jesuits.
What had happened? Galileo's erstwhile supporters were now
apparently opponents; and Galileo, with the one tone of disputation he
knew how to employ, would soon turn opponents into enemies. The
lecture of 1618, the first explicit episode in Galileo's relations with the
Jesuits after the charge before the Inquisition, takes us back to the
crucial period in Rome in 1615-16.
What had transpired in Rome? As for the charge of heresy against
Galileo, it was quietly set aside. As we shall see, the situation was in fact
much more complicated than that statement suggests. Nevertheless, the
record of the Holy Office contains nothing that explicitly resolved the
charge against him. In a letter to the court in Florence, Galileo reported
that he had been successful in convincing the Church of his own
integrity.2H I find the letter puzzling in that Galileo seemed to distinguish his integrity from his Copernican opinions. Personal integrity in
asserting opinions deemed heretical was not usually perceived as a
virtue. Nevertheless, Galileo's statement to the court does correspond
to the silence of the record about the charge against him. And the
record seems also to treat Copernicanism as a separate issue, for both
the Holy Office and the Congregation of the Index took actions on
Copernicanism as a result of the charge against Galileo. On 24 February
1616, the group of theologians consulted by the Holy Office found the
proposition that the sun stands motionless in the centre of the universe
62
RICHARD S. WESTFALL
to be absurd in philosophy and formally heretical because it expressly

contradicts the meaning of several passages of Scripture, and they
decided the proposition that the earth rotates daily on its axis and
moves in an annual orbit around the sun is also absurd in philosophy
and at least erroneous in theology. On 5 March, the Congregation of
the Index embodied these decisions in a formal decree which suspended
both Copernicus's De Revolutionibus and a work by a Spanish theologian Zuniga (which interpreted a passage in Job to say that the earth
moves) until corrected, and absolutely prohibited a recent book by the
Carmelite theologian Foscarini, which had been written in support of
Galileo ([6]19, 320-3). About any work by Galileo the Congregation
of the Index was silent.
The Holy Office and the Congregation were not unaware of Galileo,
however. They could scarcely have been unaware of him; his books and
the charge of heresy against him had been the immediate occasion of
their actions. On 25 February, the day following their decision that
Copernicanism was heretical, the group of theological consultants
informed the Holy Office of their judgment on what they called "the
propositions of the Mathematician Galileo", and the Holy Office in its
turn ordered Cardinal Bellarmine to call Galileo before him and to
warn him to give up the Copernican opinion. If Galileo should refuse
to obey, the Father Commissary of the Holy Office was instructed
formally to order him, before witnesses and a notary, not to hold,
discu~s or teach the Copernican doctrine in any way, and if Galileo did
not acquiesce in the warning, forthwith to imprison him. This action on
25 February, which focused on Copernicanism rather than Galileo and
looked toward the future rather than the past, was the closest approach
to a resolution of the charge that had been levelled against Galileo.
Cardinal Bellarmine allowed no more time to slip away than the Holy
Office itself had. On the next day, 26 February, he called Galileo before
him ([6119, 321-2). The interview has been the topic of much discussion because the document that purports to record it was not
properly signed. The order possibly given to Galileo that day figured
prominently in his trial in 1633, and understandably, since the testimony for it is suspect, there has been extensive discussion about what
took place at the interview.
The discussion seems to me largely to ignore the implications of the
action itself. Both the Holy Office and the Congregation of the Index
were acutely conscious of Galileo. His books, together with the animus
63
that he himself generated in some circles, had led directly to the action
taken against Copernicanism, as the theologians' identification of the
condemned propositions with him testifies. Those who were informed
understood that he stood at the source of the disturbance. Thus,
Antonio Querengo, in a letter to Cardinal d'Este, spoke of the decree
against Copernicanism as the resolution of "Galileo's disputes"/9 But
neither the Holy Office nor the Congregation struck Galileo. Apparently he was immune. Galileo's escape in 1616 has been attributed to
the influence of the Grand Duke and of Prince Federigo Cesi of the
Aeeademia dei Lineei. This attribution is what I have begun to doubt as
I have reflected on the multiple paeans of praise sung to Galileo in that
age. They were sung, not just by the Grand Duke and by Cesi, but by
the whole class of patrons, that is, by the ruling class, of Italy and of
Europe as a whole. Galileo was the illumination of the age. True, he
kept advancing paradoxical propositions that challenged the accepted
wisdom of the age and common sense itself. Continually he made his
patrons nervous, and they felt the need to rein him in. But almost to a
man, those who composed the Holy Office, men who by definition
belonged to the class of patrons, appear to have been profoundly
anxious not to extinguish the light of the age.
Recall that the Pope, Paul V, who was repeatedly described as
stolidly anti-intellectual, received Galileo in Rome - received him after
the publication of the decree by the Congregation of the Index! - and
assured Galileo of his favour. 3o Recall that when rumours began to
circulate that Galileo had been censured by the Holy Office, Cardinal
Bellarmine furnished him with an official, signed statement that no such
thing had occurred ([6)19, 342). Recall that in the immediate aftermath
of the events of 1616 even conservative Cardinals such as Borromeo
and Aldobrandini were eager to express their esteem for Galileo - and
to receive telescopes and copies of his books from him.31 Recall that
before he left Rome he had begun negotiations through the archlyconservative Cardinal Borgia to supply his method of determining
longitude at sea, another fruit of his genius, to the Spanish government,
and that during the following two years he remained in constant touch
with the Spanish crown on this matter. 32 And recall, what seems to me
the most revealing indication of all because it was unconscious of its
implication, the request that Galileo received from the court in Florence
while he was still in Rome in 1616, asking him to stay on until Cardinal
Carlo de' Medici arrived, since the court was concerned, "when people
64
RICHARD S. WESTFALL
of quality dine with him [the CardinalI, that he should have someone
present who would be able to please those Lords with his conversation
and discourse, for which Their Highnesses judge that you can be the
best one possible".33 It is hard to imagine how we could have a fuller
demonstration of the reputation Galileo enjoyed with every segment of
the patron class of Italy and of Europe.
If 1616 seemed to indicate that Galileo was immune from threat,
must we not say that patronage had made him immune? Such apparently was the conclusion of his enemies. In 1623, Caccini was spreading
his opinion around Rome that various "Princes" had defended Galileo
from the punishment of the Inquisition, almost that the "Princes" were
obstructing the Inquisition and protecting evil persons. 34
Galileo himself was chastened by the events of 1615-16. He did not
quickly forget the terror a denunciation to the Inquisition inspired, and
he remained rather quiet for a time. Of course he was not immune from
disciplinary action in any absolute sense. He was immune only if he
constrained himself within certain limits, and he was well aware of this
fact. It seems possible to me that in the late twenties, as he contemplated the enormous gamble the Dialogue involved, he reflected anew
on the outcome in 1616 and relied again on his apparent immunity.
This is speculation, of course, for he left no statement to that effect. The
outcome of the denunciation of 1615 is not speculation, however; it is
recorded fact.
Galileo's activities in Rome during the six months he was there were
recorded well enough that we can also speak of them as facts. By every
account he was brilliant, brilliant beyond compare. Everywhere he
could be found surrounded by opponents whom he engaged in discourse on the Copem;~an system and on natural philosophy, "now in
one house and then in another".35 In later years more than one person
recalled the impact of his discourse. Virginio Cesarini, a young prodigy
of learning in whom the Jesuits had vested great expectations, altered
the course of his intellectual life as a result of his discussions with
Galileo during these months and took up the study of mathematics and
natural philosophy which he had hitherto largely ignored. 36 He chose
not to enter the Society of Jesus, but he did become a member of Cesi's
Accademia dei Lincei. One external observer, who was not a member
of Galileo's circle, left an account of a disputation on Copernicanism at
the home of Federigo Ghisilieri. A whole phalanx of anti-Copernicans
were present, attacking him, almost in formation. Before Galileo replied,
he made it clear that he understood their position better than they did
65
by strengthening their arguments against himself until they seemed

overwhelming, and then he overturned the lot, making his opponents
look ridiculous. 37
The Jesuits could not have been unmoved by these events. They
would have found much on which to reflect. There was the subversion,
if one may use the word, of Cesarini. There were the discussions in
which they, as the resident intellectuals of Rome, must have participated. It is highly probable that more than one Jesuit, including
possibly Orazio Grassi, were present at the debate in the home of
Federigo Ghisilieri. How exquisitely did Galileo's steadfast supporters
enjoy finding themselves the butt of his ridicule? And beyond the realm
of personal impact, there was the most important event of all, the
outcome of the deliberations of the Holy Office, that is, the demonstration of the position Galileo had established among those who counted.
Copernicus was suspended. Zuniga was suspended. Foscarini was
prohibited. But Galileo, the cause of all the fuss, the very author of the
propositions that the theologians consulted by the Holy Office had
been called upon to judge, remained untouched. Here was this new
phenomenon on the intellectual scene, the scene the Jesuits had grown
accustomed to thinking of as their scene, riding high on the crest of
patronage, raised apparently above the ordinary rules. Had he raised
himself as well above the authority of the arbiters of Catholic intellectual life? The Jesuit order was not a monolith, and it is not wise to
speak of the order reaching a decision in such matters. Is it possible,
however, that influential members of the order decided at that point
that it was time to cut Galileo down to size, time to remind him who
was in charge? 38 As Grassi would say later, "If the Jesuits knew ... how
to reply to a hundred heretics they would also know how to do it to one
Catholic".39
Galileo himself was convinced in later years, after his trial, that his
troubles had stemmed from the challenge he offered to the Jesuits'
domination of Catholic intellectual life. "Having discovered many
fallacies in the philosophies commonly taught in the schools now for
many centuries", he wrote to Peiresc,
and having communicated some of them and also published some, I have aroused such
animus in the minds of those who want themselves alone to be known as learned that,
because they are very crafty and powerful, they have known how and been able to
grasp the means to suppress what I have found and published and to impede my
publication of what remains with me. 40
66
RICHARD S. WESTFALL
Interestingly, Descartes apparently assessed the situation in a similar

way and justified his own decision not to publish Le monde in relevant
terms. If he wanted to preserve his tranquillity, he told Mersenne, he
could not allow himself to nurse animosities or ambitions. Thus he
thought only of instructing himself and rejected the notion of teaching
others, "especially those who, having already acquired some credit
through false opinions, would perhaps fear to lose it if the truth were
discovered".41
We almost have to posit a decision at this time by influential Jesuits
to reassert their authority in the intellectual realm. Without it, the
watershed that 1616 constituted in Galileo's relations with them
becomes incomprehensible. No further chapters were added to the saga
of Galileo's cordial relations with the order; that era ended abruptly.
The close ties that he had formed with Cardinal Alessandro Orsini
dissolved with equal abruptness. Orsini had appeared suddenly in
Galileo's correspondence early in February. He had displayed, Galileo
wrote to the court in Florence, "a singular inclination and disposition to
protect and to favour" him, and Orsini appears to have been Galileo's
closest companion during the rest of the sojourn in Rome. When
Galileo presented his letter of introduction from the Grand Duke to
Cardinal Borghese, the Papal nephew, Orsini accompanied him, and
though Orsini had received the red hat only in December, after Galileo
arrived in Rome, he had been willing to risk Paul V's displeasure by
defending Copernicanism before the Papal throne. 42 Galileo addressed
the first written version of his treatise on the tides, which he composed
at this time and intended as a demonstration from terrestrial phenomena
that the earth moves, to Cardinal Orsini.43 Then, about the end of the
sojourn in Rome, as suddenly as he had appeared in Galileo's life, the
Cardinal disappeared, never to emerge again. The Jesuit connection
offers a possible explanation of the mystery, for the Orsini family had
close associations with the order, such that they would later become
Scheiner's patrons, and the Cardinal himself was known as a friend of
the Society, who would later lay aside the purple in order to become
one of them ([21257). One can expect Cardinal Orsini to have been
aware of decisions within the order, and to have conformed to them.
And two years after the events of 1616, Grassi delivered his lecture on
comets, not a quiet withdrawal in this instance, but a direct challenge to
Galileo's authority in science, the foundation of his position. From this
time on, conflict between Galileo and the Jesuits became the norm.
Without pretending that there was not an important divergence of
67
outlook between Galileo and the Jesuits, this analysis moves the origin
of their conflict from the realm of intellectual differences to another
context. It asks how Galileo, the successful client, impinged on others,
and it asks how those with established authority in intellectual life
viewed the arrival of a new man whose genius was the more compelling
because it enjoyed the support of the patron class.
The rest of the story is quickly told. Galileo had only one style of
controversy. To Grassi's lecture he replied with a Discourse on Comets.
The Discourse appeared over the name of Mario Guiducci; we know
from the manuscript that it was Galileo's composition, and at the time
everyone received it as such. Although the issue of comets had nothing
to do with sunspots, Galileo chose to open the Discourse with a charge
of plagiarism against Scheiner, deliberately couched in the most insulting
terms. Scheiner calls himself Apelles, Galileo said, though he does not
compare even with a mediocre painter ([616, 48; [8124). Can we
seriously wonder that Scheiner became his inveterate enemy? The
Discourse then proceeded to rip Grassi apart, not yet with the fullthroated ridicule of II Saggiatore to be sure, but in a sufficiently
humiliating manner nevertheless. In the reports that Galileo received
back from Rome there was no mistaking the reaction now. It was not
Grassi but the Jesuits who were offended, and the word they were using
in regard to Galileo was "annihilation".44 Grassi replied to the Discourse with the Libra Astronomica (ascribed to the pseudonymous
Lothario Sarsi). It is instructive to read the Libra. Despite the attack
that Galileo had made on him, Grassi replied with restraint - except
for one passage. That one passage offered, with unseemly joy, the
wicked insinuation that Galileo remained a closet Copernican. The
insinuation was wholly true, but three years after the condemnation of
1616, it could hardly be received as a friendly gesture. In due time, in
response to the Libra, Galileo composed II Saggiatore, one of the all
time masterpieces of sarcastic invective. Not even a saint would have
received II Saggiatore without hostility, and Grassi has not been
nominated for sainthood. The stage was now fully set for the final
drama.
After the tragedy of 1633, two Jesuits who were in a position to
know commented on it. Father Grienberger, the successor to Clavius in
the chair of astronomy at the Collegio Romano had this to say:
If Galileo had known how to retain the affection of the Fathers of this College, he
would have lived gloriously before the world, and none of his misfortunes would have
68
RICHARD S. WESTFALL
happened, and he would have been able to write as he chose about everything,
including the motion of the earth.45
Grassi's comment on Galileo was still more pointed:

But he has been ruined by himself, by being too infatuated with his own genius and by
wholly disdaining that of others; and therefore it is not surprising that everyone
conspires to injure him.46
Both comments seem true to me. And both agree that fully to understand Galileo's conflict with the Jesuits we need to look beyond the
intellectual issue that divided them and at the context within which they
pursued it, a context in which the system of patronage helped significantly to shape men's actions.
NOTES
1 Cf. Galileo to Diodati, 15 January 1633: "I understand from good sources that the
Jesuit Fathers have inserted in the principal heads the conviction that my book is to be
detested and is more dangerous to the Holy Church than the writings of Luther and of
Calvin ..." ([6]15, 25).
2 Cf. Naude to Gassendi, April 1633, and Micanzio to Galileo, 15 July 1634 ([6]15, 8,
and 16, 109). I could multiply the instances cited if there were need; I am convinced
that, if I wished to spend the time, I could multiply them nearly without limit.
3 Guiducci to Galileo, 21 May 1633 ([6]15, 131).
4 Bentivogli to Galileo, 21 September 1614 ([6]12, 99).
5 Gualdo to Galileo, 27 May 1611 ([6] 11, 117).
6 Galileo to Vinta, 19 March 1610 ([6]10, 297-9). The Tuscan court accepted the
proposal, and Galileo sent telescopes to a number of rulers in Catholic Europe with
letters from the Grand Duke to introduce them.
7 Faber to Cesi, 11 May 1624 ([5] 875).
8 Gassendi to Galileo, 19 January 1634 ([6]16, 21). See also Diodati to Gassendi, 10
November 1634, Diodati to Peiresc 10 November 1634, and Diodati to Schickhardt,
29 December 1634 ([6]16,153,184).
9 Galileo to Buonamici, 19 November 1629 ([6]14, 52-3).
10 Castelli to Galileo, 30 July 1638 ([6] 17,361-2).
11 Nozzolini to Monsignor Marzimedici, 12 Sept. 1612 ([6]4, 289), Galileo to Cesi, 5
January 1613, and Cigoli to Galileo, 1 February 1613 ([6]11, 461 and 476).
12 Risposta aile opposizioni ([6] 4, 727).
13 Considerazioni (]6] 4, 177-8).
14 Castelli to Galileo, 20 November 1613 ([6]11,596).
15 Risposta aile opposizioni ([6] 4, 465). Cf. a passage in Galileo's suggested dedication
to the allied "Errori ... commessi da Messer ... Coresio", in which he compared
himself to a famous painter, probably Andrea del Sarto, whose frescoes in the
courtyard of Santissima Nunciata were known to all of Florence, and his critics to
69
clumsy country imitators from the Umbrian town of Montelupo, who are no better than
house painters: "And who in our city, if he saw one of the meanest painters of
Montelupo running in fury to daub whitewash on the marvellous frescoes of Andrea
would not hasten there and with cries and remonstrances and, if they did not suffice,
with angry blows prevent this outrageous act? Sig. Galileo has handled this question so
superbly that I do not hesitate to say, and you will understand it very well, that
Archimedes himself would not have been able to explain it more ingeniously and
establish it on more solid foundations, and should a man not oppose one who, ruining
everything he touches, attempts to mutilate itT' (!6J 4, 285). (It appears to me that in
this jibe Galileo is conflating two passages from Vasari's life of del Sarto, one about a
monument to him in the church of the Servites made by Raffaella da Monte Lupo,
which the superintendents of works in the church mutilated through ignorance, and one
about frescoes depicting traitors to Florence that del Sarto painted during the seige of
the city, which were later covered with whitewash ([14J 293-4 and 298-9).) The
literature on Galileo is full of references to Galileo's enemies, but it has not frequently
bothered to consider what it was about him that aroused enmity. The provocative
display of egotism was a characteristic he never learned to suppress. A decade after the
controversy on bodies in water, he wrote the famous passage in the Assayer, comparing philosophers to eagles, which can only be read as a self-portrait, and one filled
with contempt for nearly everyone else: "Perhaps Sarsi believes that all the host of good
philosophers may be enclosed within four walls. I believe that they fly, and that they fly
alone, like eagles, and not in flocks like starlings. It is true that because eagles are rare
birds they are little seen and less heard, while birds that fly like starlings fill the sky with
shrieks and cries, and wherever they settle befoul the earth beneath them. . . . The
crowd of fools who know nothing, Sarsi, is infinite. Those who know very little of
philosophy are numerous. Few indeed are they who really know some part of it ..."
([7] 239). Nearly another decade later, Galileo began the dedication to the Dialogue
with a passage ostensibly about Ptolemy and Copernicus, which can again only be read
as his own portrait of himself: "Though the difference between man and the other
animals is enormous, yet one might say reasonably that it is little less than the
difference among men themselves. What is the ratio of one to a thousand? Yet it is
proverbial that one man is worth a thousand where a thousand are of less value than a
single one. Such differences depend upon diverse mental abilities, and I reduce them to
the difference between being or not being a philosopher; for philosophy, as the proper
nutriment of those who can feed upon it, does in fact distinguish that single man from
the common herd in a greater or less degree of merit according as his diet varies"
(!9] 3). What can we call this except hubris? Can we afford to ignore its impact on
others if we want to understand important episodes in Galileo's life?
16 Fragments by Galileo in response to Colombe and di Grazia ([6] 4, 443). There may
be no need further to establish what is a well known aspect of Galileo, but let me call
attention to the notes he made on Antonio Rocco's Esercitazioni filosofiche, a reply to
the Dialogue. Even in private notes Galileo could not curb the violence of his reaction.
Thus, "this beast has not fully understood a word of what I say here". Then, "0 wicked
and totally ignorant!". And ultimately, "blockhead!", "fathead". ([6] 7, 641, 645, 646,
668.) After 1633, Galileo did not venture to publish these comments. Before the trial,
the system within which he worked tended to encourage him to display his arrogance
rather than to conceal it.
70
RICHARD S. WESTFALL
Aggiunti to Galileo, 4 June 1633 ([6]15, 144).

Cigoli to Galileo, 16 December 1611 ([6]11, 241-2).
19 Galileo's letter does not survive, but we have Conti's reply and his second reply
when Galileo pursued the question further. (Conti to Galileo, 7 July and 18 August
1612 ([6]11, 354-5, 376).
20 See Lorini's denial when Galileo confronted him, 5 November 1612 ([6]11, 427).
21 Galileo to Castelli, 21 December 1613 ([6] 5, 281-8).
22 The records of the Inquisition that concern Galileo are published in [6]19, 275421. They include both the 1615-16 episode and trial of 1632-3.
23 Galileo to Clavius, 8 January 1588, and Clavius to Galileo, 16 January 1588 ([6]10,
22-5).
24 Dini to Galileo, 25 April 1616 ([6]12, 174); Galileo to Dini, 16 February 1615
([6] 5,292-5).
25 The lecture is printed, in translation, in ([8] 5-18).
26 Rinuccini to Galileo, 2 March 1619 ([6]12, 443).
27 The line is in the Assayer and refers explicitly to a passage in SarsilGrassi's Libra
astronomica ([8]270). The tone of Galileo's Discourse on Comets can hardly be
explained unless we assume a similar reaction to Grassi's original lecture.
28 As regards the charge against him, Galileo has been assured that "la determinazione
essere stata di haver toccato con mano non meno la candidezza et integrita mia, che la
diabolica malignita et iniqua volonta de'miei persecutori". Galileo to Picchena, 6
February 1616 ([6]12, 230). Cf. Galileo to Picchena, 13 February 1616 ([6]12,
233-4).
29 Querengo to d'Este, 5 March 1616 ([6]12, 243) .
.10 Galileo to Picchena, 12 March 1616 ([6]12, 247-8).
31 Borromeo to Galileo, 14 July 1617; Giggi (Borromeo's secretary) t6 GaJileo, 26 July
1617; Galileo to Borromeo, 23 December 1617; Giggi to Galileo, 27 December 1617;
Ciampoli to Galileo, 21 July 1618 ([6]12, 320, 332, 356-7, 362, and 399-400).
32 GaJileo to d' Argensola, 16 May 1616 ([6]12, 260-1). Related correspondence
continued at least until April 1618 ([6]12,268-384, passim.) See especially Galileo to
Orso d'Eki, the Tuscan ambassador in Madrid, June 1617, for his account of the
beginning of the negotiations ([6]12,327-8).
.13 Picchena to Galileo, 19 February 1616 ([6]12, 237). The letter was written in
response to Galileo's announcement that he had cleared his own name, but before the
decree against Copernicanism and before the extremely negative report on the visit
from the ambassador, Guicciardini. By May the court was increasingly anxious to get
Galileo out of Rome and back home .
.14 Reported in Castelli to Galileo, 6 December 1623 ([6]13,156).
35 Querengo to d'Este, 20 January 1616 ([6]12, 226-7).
36 Cesarini to Galileo, 1 October 1618 ([6]12,413-15). For another recollection of
discussions in Rome see Carlo Muti to Galileo, 7 September 1618 ([6]12, 411), and
for a hostile report that describes the discussions in which he engaged, Guicciardini
to Picchena, 13 May 1616 ([6]12, 259). Francesco Ingoli's tract, De situ et quiete
terrae contro Copernici systema disputatio ([6] 5, 403-12), which ultimately generated
Galileo's Letter to 1ngoli, arose from one of the discussions in Rome.
17
18
71
Querengo to d'Este, 20 January 1616 ([6]12, 226-7).

This is similar to the conclusion stated by Giorgio di Santillana after a different
analysis ([13] 206).
39 The comment was directed specifically at II Saggiatore; reported in Rinuccini to
Galileo, 2 December 1623 ([6]13,153-4).
4U Galileo to Peiresc, 16 March 1635 ([6]16, 235).
41 Descartes to Mersenne, February 1634 ([4]1, 282). Cf. substantially the same
opinion in Diodati to Bernegger, 6 January 1635 ([6]16, 194-6) and Micanzio to
Galileo, 5 May 1635 ([6]16, 264).
42 Galileo to Picchena, 6, 13, and 20 February 1616 ([6]12, 231-2, 235 and 238);
Guicciardini to Cosimo, 4 March 1616 ([6]12, 242).
43 "Discorso del flusso e reflusso del mare all'Illustrissimo e Reverendissimo Sig.
Cardinale Orsini" ([6]5,377-95).
44 Ciampoli to Galileo, 6 December 1619 ([6]12, 498-9). Cf. Ciampoli to Galileo, 12
July 1619 and Muti to Galileo, 24 September 1619 ([6]12, 465-6 and 492).
45 Galileo quoted a friend in Rome, in Galileo to Diodati, 25 July 1634 ([6]16,117).
46 Grassi to Bardi, 22 September 1633 ([6]15, 273).
37
38
REFERENCES
1. Carugo, Adriano, and Alistair Crombie, 'The Jesuits and Galileo's ideas of science
and of nature', Annali dell'lstituto e Museo di Storia della Scienza di Firenze, 8.2
(1983)3-68.
2. Colanna, Gustavo Brigante, Gli Orinsi (Milano: Ceschina, 1955).
3. de Dainville, Fran~ois, La naissance de l'humanisme moderne (Paris: Beauchesne,
1940) .
4. Descartes, Rene, Oeuvres de Descartes, 12 vols, eds. C. Adam and P. Tannery
(Paris: Vrin, 1964-76).
5. Gabrieli, Guiseppe, 'II carteggio linceo della vecchia Accademia di Federigo Cesi
(1603-1630)" Memorie della R. Accademia Nazionale dei Lincei. Classe di
scienze morali, storiche e filologiche, VI, 7. I and 7.2 (1938).
6. Galilei, Galileo, Le opere di Galileo Galilei, 20 vols in 21, ed. Antonio Favaro
(Firenze: Barbera, 1890-1909).
7. Galilei, Galileo, Discoveries and Opinions of Galileo, tf. Stillman Drake (Garden
City, N.Y.: Doubleday, 1957).
8. Galilei, Galileo, The Controversy on the Comets of 1618, tf. Stillman Drake and C.
D. O'Malley (Philadelphia: Univ. of Pennsylvania Press, 1960).
9. Galilei, Galileo, Dialogue Concerning the Two Chief World Systems, tr. Stillman
Drake (Berkeley: Univ. of California Press, 1962).
10. Harney, Martin P., The Jesuits in History (New York: American Press, 1941).
11. Hollis, Christopher, A History of the Jesuits (London: Weidenfeld and Nicolson,
1968).
12. Mitchell, David, The Jesuits. A History (London: Macdonald, 1980).
13. di Santillana, Giorgio, The Crime of Galileo (Chicago: Chicago Univ. Press, 1955).
72
RICHARD S. WESTFALL
14. Vasari, Giorgo, Lives of Seventy of the Most Eminent Painters, Sculptors and
Architects, 4 vols, eds. E. H. Blashfield, E. W. Blashfield, and A. A. Hopkins (New
York, 1897).
15. Villoslada, Riccardo G., Storia del Collegio Romano (= Analecta Gregoriano, 66)
(Rome: Universitas Gregoriana, 1954).
16. Wallace, William, Galileo and His Sources. The Heritage of the Collegio Romano
in Galileo's Science (Princeton: Princeton Univ. Press, 1984).
WILLIAM R. SHEA
DESCARTES AND THE ROSICRUCIAN

ENLlGHTENMENTl
Abstract. Descartes' biographer, Adrien Baillet, tells us that Descartes heard of the
Brothers of the Rosy Cross in Germany during the winter of 1619-20. This news
reached him "at a time when he was in the greatest perplexity concerning the way that
he should follow in the investigation of the truth", and he immediately attempted
to contact members of the Fraternity. Descartes failed to meet any of the alleged
Rosicrucians but he was rumoured to have become one of their members when he
returned to Paris in 1623, the year of the great Rosicrucian Craze that was to engross
some of the best French minds for several months. Baillet says that Descartes had no
difficulty in proving that he was not a Rosicrucian but the accusation stuck and was
occasionally revived during his lifetime and after.
In this essay, I examine Descartes' interest in the Rosicrucians against the general
background of the period, and I consider the reasons why he was subsequently accused
of belonging to the movement. I also indicate how some of Descartes' more daring
pronouncements in the works of his maturity can be traced to the hermetic and mystical
tradition that influenced his youthful quest for a new method that would unlock the
secrets of nature.
It is generally agreed that men donned new thinking caps in the
seventeenth century, and this can be brought home vividly by comparing Descartes' lucid manifesto of rationalism in the Discourse on
Method or the Principles of Philosophy with the outlook prevalent a
few years earlier. For instance in the concluding paragraph (article 64)
of the second part of the Principles of Philosophy, Descartes writes:
I recognize no matter in corporeal things apart from that which the geometers call
quantity, and take as the object of their demonstrations, i.e. that to which every kind of
division, shape and motion is applicable. Moreover, my consideration of such matter
involves absolutely nothing apart from these divisions, shapes and motions; and even
with regard to these, I will admit as true only what has been deduced from indubitable
common notions so evidently that it is fit to be considered as a mathematical demonstration. And since all natural phenomena can be explained in this way, as will become
clear in what follows, I do not think that any other principles are either admissible or
desirable in physics ([141 B.1, 78-9; [131 1,247).
Less than three decades earlier, texts such as the following were
widespread:
73
74
WILLIAM R. SHEA
The things which are perceivable by the senses are helpful in enabling us to conceive of
Olympian matters. The wind signifies spirit; movement with the passage of time signifies
life; light signifies knowledge; heat signifies love; and instantaneous activity signifies
creation. Every corporeal form acts through harmony. There are more wet things than
dry things, and more cold things than hot, because if this were not S,), the active
elements would have won the battle too quickly and the world would not have lasted
long ([14]10, 218; [13]1, 5).
Or again:
There is a single active power in things: love, charity, harmony ([14]10, 218; [13]1, 5).
These quotations embody, for us, two profoundly different ways of

looking at the world, and, if we perceive them as radically opposed, it is
largely because we are the heirs of the Cartesian reform in philosophy. I
chose them, however, not only to obtain clarification by contrast but
because they are, in fact, by the same man. The last two quotations are
also by Descartes, and they were written around 1619 when he was in
his early twenties. What happened between that time and 1637 when he
published his first book, the Discourse on Method, at the age of
forty-one, is a complicated story, and what I shall have to say about it
will be largely tentative and exploratory. Descartes was, in that felicitous, if sadly obsolete, phrase, "a gentleman of independent means",
and he did not have to publish for fear of perishing. In other words, he
could afford to wait before going to press, and he left no published
juvenile papers behind.
The last two quotations come from a notebook that Descartes kept
for many years and that was found among his possessions after his
untimely death in Stockholm in 1650. Descartes' papers were brought
back to France by the French ambassador Hector-Pierre Chanut and
turned over to his brother-in-law Claude Clerselier, a friend of Descartes. When Leibniz visited Paris in 1675, he was shown the manuscript and he took copious notes. These were discovered by Foucher de
Careil among the Leibniz papers in Hanover and were first published
as Cogitationes privatae or Private Thoughts in 1859 ([141 10, 21348).
It is, of course, not enough to state that "love, charity and harmony
are all a single power" to be a Rosicrucian, but then the Rosicrucians
never existed. All we find are sympathizers of all or some of the ideals
proclaimed in the so-called Rosicrucian tracts that appeared between
1614 and 1619, namely between Descartes' eighteenth and twenty-
DESCARTES AND THE ROSICRUCIAN ENLIGHTENMENT
75
fourth birthday, a period in life when the mind is still open to bold
ideas and grandiose schemes. These tracts include the Fama Fraternitatis (1614), the Confessio (1615), the Chemical Wedding of Christian
Rosencreutz (1616), and the Raptus philosophicus (1619) to mention
but a few. 2 The Fama and the Confessio, which appeared in several
modem languages, were easy to read, because they were short (together
they run to about twenty pages of printed text), and attractive, because
they attacked the Establishment. Short shrift is made of the acknowledged authorities: the Pope in religion, Aristotle in philosophy, and
Galen in medicine.
The hero of these tracts is a mysterious Rosencreutz who according
to the Confessio was born in 1378 and lived for 106 years. He travelled
in the east where he learned the "Magia and the Cabala" and entered
into "the harmony of the whole world".3 He returned to Europe to
found a society for the reformation of universal knowledge at a time
which brought forth such men as Theophrastus (Paracelsus) who was
''well-grounded in the aforesaid harmonia", although he was not a
member of the Fraternity ([43) 286). The author of the Fama recounts
the history of the fictitious Order culminating with the recent discovery
of the passage of the vault in which Brother Rosencreutz was buried in
Germany. This is seen as a symbol heralding the dawn of a new age:
"For like as our door was after so many years wonderfully discovered,
also there shall be opened a door to Europe (when the great wall is
removed) which already doth begin to appear, and with great desire is
expected of many" ([43) 290-1). The Fama ends by exhorting its
readers "to declare their mind ... in print" ([43) 296). Several did and
were contacted by others who had also read the Rosicrucian manifestos,
but there is no recorded instance of anyone ever meeting a member of
the alleged Fraternity.
Descartes' biographer, Adrien Baillet, tells us that Descartes heard
of the Brothers of the Rosy Cross in Germany during the winter of
1619-20. Baillet's account is particularly instructive because he had
access to Descartes' manuscript notes that have since been lost:
It was in conversations with his German hosts that he heard of a Brotherhood of
scientists established in Germany some time ago under the name of the Brothers of the
Rosy-Cross. They were praised in an astonishing manner ... and Descartes felt shaken.
He, who openly stated his genuine contempt of aU scientists because he had never met a
genuine one, began to accuse himself of haste and rashness. His desire to emulate them
was strengthened by the fact that news of the Brotherhood reached him at a time when
76
WILLIAM R. SHEA
he was in the greatest perplexity concerning the way that he should follow in the quest
for truth. He felt that he could not remain indifferent to them ([4J 1, 87; cited in [14110,
193).
As was to be expected, Descartes was unsuccessful in his attempt to

contact the Fraternity. "Peu s'en fallut qu'il ne mit la societe au rang des
chimeres" ([4]1, 90; in [40]10,196).
When Descartes returned to Paris in 1623, he was surprised to find
it rumoured that he was a Rosicrucian. Now 1623 is the year of the
great Rosicrucian Craze. We know this from the account of Gabriel
Naude in his Instruction a la France sur la verite de l'histoire des
Freres de la Rose-Croix (Paris, 1623), and the anonymous pamphlet
Effroyables pactions faites entre Ie Diable et les prhendus invisibles
(Paris, 1623), as well as from Fr. Franl(ois Garasse's voluminous (it
runs to 1025 pages) La doctrine curieuse des beaux esprits de ce temps
au pretendus tels et natamment celie des mechants belistres nommes les
Freres de la Croix de Roses (Paris, 1623).
According to Naude, Parisians awoke one morning to find the
following placard at various crossroads:
We, the delegates of the Main College of the Brothers of the Rosy Cross, are making a
visible and invisible visit to this city by the Grace of the Almighty toward whom the
hearts of the just are turned. We show and teach without books or signs how to speak
all kinds of languages of the countries where we wish to be in order to draw our
fellow-men from deadly error ([28127).
As Baillet puts it:

The chance of their arrival in Paris at the same time as M. Descartes might have had a
unfortunate effect on his reputation, had he concealed himself or lived in solitude in the
town, as he had been accustomed to do in his travels. But he confounded those who
wished to make use of this conjunction of events to establish their calumny. He made
himself visible to all the world, and particularly to his friends who needed no other
argument that he was not one of the Brotherhood or Invisibles. 4
If all the disproof that was needed was "to make himself visible",
then Descartes had an easy time of it! In any event, this was most
fortunate since Fr. Garasse had clear ideas about the treatment that
should be meted out to Rosicrucians:
I conclude that the Brothers of the Roses are guilty, evil, and condemned as sorcerers.
They are an evil grouping of good-for-nothings, dangerous to religion, the secular state
and sound morality .... I cannot imagine tortures great enough for these dogmatizers.
77
Garasse adds for good measure: "the doctrine of these allegedly fine
spirits, of which they speak in their cabala, is atheism". In this they
resemble Luther who "was a perfect atheist", and, hence, "what is best
for them is the rack or the noose".5
But why, we may ask, was Descartes suspected of being a Rosicrucian? There are a number of reasons, all circumstantial. I shall consider
five.
The first is that he came from Germany, the land from which
Rosicrucian tracts poured forth. We know that Descartes had no luck in
his quest for the elusive and illusory Brothers of the Rosy Cross, but
according to Baillet he met and impressed Johann Faulhaber in Vim in
1620. Prior to 1613, namely before the beginning of the Rosicrucian
craze, Faulhaber wrote mathematical treatises with forbidding but
philosophically unproblematic titles such as Arithmeticus cubicossicus
hortus (Tubingen, 1604), Usus de novo in vento instrumenti alicuius
Belgae (Augsburg, 1610), Novae geometriae & opticae inventiones,
aliquot peculiarium instrumentorum (Frankfurt, 1610), Speculum
mathematicum polytechnicum novum, tribus visionibus illustre (Vim,
1612). After 1613, we find the following publications: Ansa inauditae
novae & admirandae artis, quam Spiritus Dei aliquot propheticis &
Biblicis numeris ad ultima usque tempora obsignare & occultare voluit
(Nuremberg, 1613), Cae/estes arcana magia, sive cabalisticus, novus,
artificiosus & admirandus computus de Gog & Magog (Nuremberg,
1613) and, especially, the Mysterium arithmeticum sive cabalistica et
philosophica inventio, nova admiranda et ardua, qua numeri ratione et
methodo computantur. ... Cum illuminatissimis laudatissimisque Frat.
R. C. Famae viris humiliter et syncere dicata (Vim, 1615). The latter is
one of the first works dedicated to the Rosicrucians.
The spontaneous, enlightened comment is: "Yes, but Descartes was
interested in Faulhaber the mathematician, not Faulhaber the Hermeticist". Fair enough. But was this distinction, so obvious to us, equally
clear to the seventeenth century? Why are two distinguished editors
of Euclid, John Dee in England and Fran~ois de Foix de Candalle
in France, professed hermeticists? What should have driven a man
like Dee to write a lengthy Preface to Euclid as well as the Monas
hieroglyphica, or Candalle to comment on Euclid and translate the
Pimander of Hermes Trismegistus?
Naude gives a list of authors whose doctrine is embraced by the
Rosicrucians. This includes not only the names of Dee and Candalle
but also the Mertonians John Hentisbury and Richard Swineshead
78
WILLIAM R. SHEA
([28] 31). It would appear that students of mathematics, at least in

France and England, were sometimes accused of magic because of the
wonders they worked. John Dee complains of being falsely accused of
being a "conjuror" because of his mathematical skill and ability to
produce mechanical marvels. 6 Now Descartes was very much interested
in mechanical devices and optical illusions. Leibniz copied the following
passage from his notebook:
Shadows representing various shapes, such as trees, can be made in a garden .... In a
room, tongues of fire, chariots of fire, and other shapes can be made to appear in the
air. This is done by mirrors that collect the rays at certain points ([14]10, 215-16).
The source of these remarks is book 17 of della Porta's popular

Magia naturalis.7 Descartes never described himself as an assiduous
reader of della Porta but there are several indications that lead me to
believe that he spent more time between the covers of the Magia
naturalis than a genuinely rational (i.e., Cartesian) person would wish to
own. 8 When Isaac Beeckman mentioned to Descartes in 1629 that
Cornelius Agrippa, the author of De occulta philosophia, claimed that
letters could be inscribed on the face of the moon so that a message
could be read at some remote place on earth, Descartes replied that
della Porta ascribed this to the action of burning lenses. 9 Beeckman and
Descartes indulged in early science fiction and speculated that if we
could make telescopes powerful enough to observe the inhabitants of
the moon, and found that these inhabitants were as well equipped as we
are, then they might accept to signal to us what is going on in the
southern hemisphere. lO
To the French lens-maker, Jean Ferrier, whom he wanted to hire as
his assistant, Descartes wrote on 13 November 1629, "I venture to
hope that, with your aid, we will see whether there are living beings on
the moon" ([14]1, 61). In the Ninth Discourse of La Dioptrique,
published in 1637, he reaffirmed his conviction that,
if the skill of the workers do not fail us, we shall be able with this invention to see
objects as particular and as small in the celestial bodies as those we commonly see on
earth ([ 14[ 6, 206).
Descartes also hoped that Ferrier would help him develop "what I
call the science of miracles because it teaches how to use air and light
in such a way that we can achieve through their agency all the illusions
that Magicians are said to perform with the help of Demons"Y The
79
Meteores, published with the Discourse on Method and the Dioptrique

in 1637, bears witness to his abiding fascination with the "science of
miracles". He tells of a way of using his newly-acquired knowledge of
the formation of rainbows "to make signs appear in the sky that could
cause great wonder in those who were ignorant ofthe causes"P
Earlier in 1619, shortly before leaving Germany, Descartes had a
long conversation with a man in Dordrecht on the art of Raymond
Lullius, and he wrote to Beeckman, who had a copy of Cornelius
Agrippa's In Artem brevem Raymondi Lullii commentaria, asking
whether there was anything worth pursuing along those lines.u It might
be objected, of course, that Descartes was intent on discovering basic
truths and setting them out as clearly and distinctly as possible, and
that, therefore, he had no truck with the obscure writings of Lullius'
followers. But the Rosicrucians' manifestos also calied for simple and
straightforward language. The Confessio voices this concern in moral
terms:
We must earnestly admonish you, that you put away, if not all, yet most books written
by false Alchemists, who do think it but a jest, or a pastime, when they either misuse
the Holy Trinity, when they do apply it to vain things, or deceive the people with most
strange figures, and dark sentences and speeches, and cozen the simple of their money;
as there are nowadays too many such books set forth, which the Enemy of man's
welfare doth daily, and will to the end, mingle among the good seed, thereby to make
the Truth more difficult to be believed, which in herself is simple, easy, and naked
([411305).
The Raptus philosophicus of 1619 makes this point in its very

title: Raptus philosophicus, das ist Philosophische Ojfenbarungen gang
Simpel und Einfaltig gestellet, und an die HochlObliche and beruhmte.
Fraternitet R.c. unterthiinig geschrieben.
The Rosicrucian manifestos insist that we must put an end to the
obfuscation of the truth by pedantry and conceit. Or to quote the young
Descartes once again:
The sciences are now masked, but when the masks are lifted, they will be seen in their
beauty. Upon inspecting the chain of the sciences, it will not appear more difficult to
remember them than a series of numbers.14
This is why the Rosicrucians profess not a new but a restored

knowledge. "Our philosophy also", we read in the Fama, "is not a new
invention, but [is] as Adam after his fall hath received it, and as Moses
80
WILLIAM R. SHEA
and Salomon used it" ([41]295). Readers of Descartes' Rules for the
Direction of the Mind may recollect a passage in Rule Four:
So useful is this method that without it the pursuit of learning would, I think, be more
harmful than profitable. Hence I can readily believe that the great minds of the past
were to some extent aware of it, guided to it even by nature alone. For the human mind
has within it a sort of spark of the divine, in which the first seeds of useful ways of
thinking are sown, seeds which, however neglected and stifled by studies which impede
them, often bear fruit of their own accord. 15
In Descartes' early notebook, there is an even more forceful

statement:
It may seem surprising to find weighty judgements in the writings of the poets rather
than the philosophers. The reason is that the poets were driven to write by enthusiasm
and the force of imagination. We have within us the sparks of knowledge, as in a flint
[the Latin has "sunt in nobis semina scientiae, ut in silice", literally, "we have within us
the seeds of science, as in a flint"]: philosophers extract them through reason, but poets
force them out through the sharp blows of the imagination, so that they shine more
brightly ([14)10,217; [13)1, 4).
Descartes' account of the "revelation" that came to him on the night

of 10-11 November 1619 begins with the words: "10 November
1619, when I was full of enthusiasm and had found the foundation of
the admirable science" ([14]10, 179). He did not, therefore, make his
fundamental discovery "through reason" like the philosophers, but ''full
of enthusiasm" like the poets. This is assuredly not the Descartes of our
textbooks!
If the first reason for suspecting that Descartes was a Rosicrucian
was his recent sojourn in Germany, the second was provided by the
Rule of the Brotherhood as set out in the Fama Fraternitatis. It
comprised six short articles whereby the members were enjoined: (1) to
cure the sick free of charge; (2) to wear no special habit but follow the
custom of the country in which they happened to be; (3) to meet once a
year; (4) to find a worthy person to succeed them; (5) to use c.R. as
their seal; (6) to maintain the Fraternity secret for a hundred years.
I shall indicate briefly how, especially in the heated Parisian climate
of 1623, these Rules could have been seen as fitting Descartes'
behaviour.
Descartes was not a physician but he gave detailed medical advice to
his friends, usually with the request that they should not tell anyone he
had done so. In his biting satire, Nouveaux memoires pour servir d
81
l'histoire du cartesianisme published at the end of the seventeenth

century, Daniel Huet describes Descartes as the perfect Rosicrucian. "I
renounced marriage", he has him say, "I led a wandering life, 1 sought
obscurity and isolation, 1 abandoned the study of geometry and of the
other sciences to apply myself exclusively to philosophy, medicine,
chemistry, the cabala and other secret sciences" .16
Huet writing in 1692 is not a reliable source of information about
the working of Descartes' mind more than half a century earlier, but it
proves that the charge of being a Rosicrucian, levelled in 1623, stuck.
Nicolas Poisson, in his Commentaire ou remarques sur la methode de
Rene Descartes, published in 1670, goes out of his way to vindicate
Descartes of this accusation. His main argument, arrived at with all the
advantages of hindsight, is quite simply that Descartes ''was too
sophisticated to be a friend of these visionaries who rest all their
arguments on empirical evidence rather than on reasoning" .17
But Descartes not only practised medicine (albeit discreetly and free
of charge),18 he also believed in the possibility of prolonging life. He
confided to Constantijn Huygens that he hoped to live more than a
hundred years (the Rosicrucians generally aimed for a hundred and
twenty) and that he was writing a medical treatise to that end. 19 Of even
greater interest in the testimony of the Abbe Claude Picot, the French
translator of the Principles of Philosophy. Picot spent three months with
Descartes in Holland in 1647 and, upon his return to France
he resolutely gave up high living, to which he had not hitherto been opposed, and
adopted M. Descartes' diet in the belief that it was the only way of ensuring the success
of the secret method that he claimed that our Philosopher had found to make men live
four or five hundred years ([41 2,448).
According to Baillet, Picot was so convinced of the genuineness of

Descartes' discovery that he was ready to swear
that barring an unusual and violent cause (such as the one that put his machine out of
order in Sweden) he would have lived five hundred years, having discovered the art of
living for several centuries. 20
It would seem that some of Descartes disciples could have shared

with those of Robert Fludd the name of "longlivers"!
The Second Rosicrucian Rule enjoined to dress and live like
everybody else, and Descartes was careful to do just this. He was
anxious not to appear singular in his dress or his person.21 When he
82
WILLIAM R. SHEA
heard of Galileo's condemnation in 1633, he wrote to Mersenne that he

had decided not to publish his cosmological treatise Le Monde, in
keeping with his motto "bene vixit, bene qui latuit" ("he lives well who
hides well").22
The Third and Fourth Rule (enjoining to meet once a year, and to
find a successor) offer little indication. But the Fifth, namely that c.R.
should be the only seal or mark of the Brotherhood, has given rise to
much speculation, especially when conjoined with the following passage
from the Fama Fraternitatis:
After a time there will be a general reformation, both of divine and human things ... in
the meantime some few, who shall give their names, may join together, thereby to make
a happy and wished for beginning of our Philosophical Canons, prescribed to us by our
brother R.C. ([431 294).
R.c., of course, are the initials of Renatus Cartesius. But as Etienne

Gilson very sensibly remarked, "If Rene Descartes had been called
Pierre Gassendi, the argument from the seal would be singularly
stronger" ([18]278).
The Sixth and final Rule prescribed secrecy about the Brotherhood,
and is not particularly helpful since we can hardly conclude from
someone's silence that he was a member of the organization. Nevertheless there is a curious text in Descartes' early notebook that has never
been elucidated:
Actors, taught not to let any embarrassment show on their faces, put on a mask. I will
do the same. So far, I have been a spectator in this theatre which is the world, but I am
now about to mount the stage, and I come forward masked. 23
Why should the twenty-three year old Descartes want to wear a

mask? I shall return to this point presently.
His German connections and the Rules of the Brotherhood provided
two reasons for suspecting that Descartes had Hermetic leanings. A
third reason was his interest in a universal language, a plan which
appealed to reformers for whom it heralded the dawn of genuine
communication. In a letter to Mersenne, Descartes staies clearly,
I believe that this language is possible and that we can find the science upon which it
rests. By its means, peasants could be better judges of the truth of things than
philosophers at the present time. 24
83
Note how conveniently the peasants will be spared the tedious task
of reading what philosophers have written. The obsolescence of our
departments of philosophy awaits the discovery of a universal language!
A fourth reason for crediting Descartes with a willingness to
acknowledge non-Cartesian forces in nature could have been culled
from his Compendium musicae, composed in 1618, in which he avers
that:
The human voice is pleasant because it agrees with our dispositions. What makes the
voice of a friend more agreeable than that of an enemy is probably the sympathy or
antipathy that we feel. For the same reason, a drum covered with the skin of a lamb
ceases to vibrate and becomes silent if, as we are told, its initial sound sets up a
resonance in another drum covered with the skin of a wolf ([14]10,90).
The source of this story could again be della Porta who gives two
versions of it:
The Wolf is hurtful and odious to sheep after he is dead: for if you can cover a drum
with a wolf's skin, the sound of it will make the sheep afraid ... if you hang several
skins one against the other ... the Wolf's skin eats up the Lamb's skin (1301 bk. 1, ch.
14,pp.19-20).
There is Antipathy between Sheep and Wolves, as I said often, and it remains in all
their parts; so that an Instrument strung with Sheep strings, mingled with strings made
of a Wolf's guts, will make no music, but jar, and make all discords. 25
There is always, of course, the possibility of a mechanical explanation of such phenomena, as Descartes urged in 1644 in the Fourth Part
of his Principles of Philosophy.26 But, in 1619, I wonder how far his
explanation would have risen above the one della Porta offers for
"infected mirrors". Della Porta informs us in book I of the Magia
naturalis that a harlot possesses a "virtue" such that "if someone often
looks at himself in her mirror, or puts on her clothes, he will become as
insolent and lewd as she is".27 Seven books later, having kept us on
epistemological, if not moral, tenterhooks, he offers the following
explanation:
The polished mirror fears the look of an immoral woman, as Aristotle says, for her look
soils it and reduces its splendour. This is because the vapour of her blood coalesces on
the surface of the mirror. 28
84
WILLIAM R. SHEA
I hasten to add that we do not find this explanation in Descartes'

own writings; nonetheless, in a fragment dated 1631, we read:
Spirits are exhaled through the eyes as can be seen in menstruating women whose eyes
are said to emit vapours. The whole body of a woman is full of vapours when she has
her days. The heavier humour is purged through the vagina, the subtler humour is
purged higher up, namely through the eyes. 29
In the earlier notebook, we find this cryptic sentence about women

and science:
Science is like a woman: if she stays faithful to her husband she is respected; if she
becomes common property she grows to be despised. 30
We also know that about this time Descartes read Campanella's De

sensu rerum et magia !ibri quatuor, pars mirabilis occultae philosophiae,
ubi demonstratur mundum esse Dei vivam statuam beneque cognoscentem (Frankfurt, 1620) which he described, some fifteen years later, as
having left no impression on him except one of superficiality.31 By then,
Descartes was light years away from the intellectual atmosphere of his
youth and, as psychologists tell us, his memory had perhaps been
cleansed. Or it may be a case of complacent ladder-kicking-away
behaviour not unknown among middle-aged and tenured academics.
The fifth, and final reason, for associating Descartes with the
fashionable Rosicrucian movement is the title of a book that Leibniz
found in Descartes' notebook:
The mathematical treasure trove of Polybius, citizen of the world. This work lays down
the true means of solving all the difficulties in the science of mathematics, and
demonstrates that the human intellect can achieve nothing further on these questions.
The work is aimed at certain people who promise to show us miraculous discoveries in
all the sciences, its purpose being to chide them for their sluggishness and to expose the
emptiness of their boasts. A further aim is to lighten the agonizing toil of those who
struggle night and day with the Gordian knots of this science, and who squander their
intellectual resources to no avail. The work is offered afresh to learned men throughout
the world and especially to the distinguished brothers of the Rose Croix in Germany
([14J 10,213; [13J 1,2).
To date no one has been able to trace this book. Henri Gouhier has
suggested that it is the title of a projected work in which Descartes
would appear on the "theatre of the world ... wearing a mask". As a
resolute defender of Descartes' "rationality", Gouhier interprets the
dedication to the Rosicrucians as ironical. 32 But we need not share
85
Gouhier's qualms about admitting what Descartes was interested in in

the Rosicrucian movement, for what intelligent twenty-year old misses
the opportunity of jumping on an exciting bandwagon that trundles
through the campus? Descartes may have wished to contact the elusive
Rosicrucians by means of this publication and to gain their respect by
displaying his recent advances in mathematics. He does not seem to
have been reluctant, for instance, to speak of the Cabala of the
Germans in a mathematical note of uncertain date but prior to 1629.
Having explained some trigonometrical relationships, he concludes:
From these we can deduce an infinite number of theorems, and easily explain the
arithmetical progressions that include the bases or the sides of all triangles of this
nature, thereby imitating the Cabala of the Germans. 33
This may involve a reference to the book by Faulhaber that I

mentioned earlier, the Mysterium arithmeticum sive cabalistica et
philosophica inventio qua numeri ratione et methodo computantur
which is dedicated like the Thesaurus mathematicus of Polybius the
Cosmopolitan, to the Rosicrucians. This would explain the "denuo" of
the latter dedication. 34
The interesting question, of course, is whether this lengthy discussion
of the Rosicrucian atmosphere (or craze) can shed light on Descartes'
mature philosophy. I believe it can, at least, help us understand the
famous autobiographical passage in the Discourse on Method where
Descartes tells us that he discovered his new method in his winter
quarters in Germany:
At that time I was in Germany, where I had been called by the wars that are not yet
ended there. While I was returning to the army from the coronation of the Emperor,
the onset of winter detained me in quarters where, finding no conversation to divert me
and fortunately having no cares or passions to trouble me, I stayed all day shut up
alone in a stove-heated room, where I was completely free to converse with myself
about my own thoughts ([1416,11; [1311,116).
The wars are the Thirty Year War that did not end until the Peace of
Westphalia in 1648. The Emperor is Ferdinand, who was crowned in
Frankfurt in 1619. The exact location is not known but Daniel
Lipstorp, writing in 1653, three years after Descartes' death, suggests a
village near Vim, where the mathematician Faulhaber lived.35
"Finding no conversation to divert me and fortunately having no
cares or passions to trouble me ... I was completely free to converse
86
WILLIAM R. SHEA
with myself about my own thoughts". How detached and serenely

philosophical all this sounds! The mature Descartes' description of the
momentous discovery of his new method would have us believe that he
arrived at his insight in the posture a sculptor would select were he
asked to represent "the thinker". It is, in fact, a posture. In 1619,
Descartes described the change in his life in a language that is far
removed from the cool and calm prose of the Discourse on Method. It
is couched in the language of dreams and, although the manuscript in
which he wrote out a detailed account of his visionary experience has
vanished, it was seen by Leibniz during his visit to Paris in 1675-6 and
it was translated by Baillet in the first volume of his biography,36
Descartes records that on the night of the tenth to the eleventh of
November 1619 he had, in rapid succession, not one but three dreams
"which he imagined could only come from on high". In the first dream,
he was frightened by ghosts and buffeted by a strong wind that kept
him from advancing to where he wished to go. The imagery is vivid and
would have meant more to a seventeenth-century hermeticist than it
does to us. He felt, for instance, a weakness on the right side, a
whirlwind made him spin three or four times on his left foot, others
were straight and steady while he wavered, and so forth. Descartes
woke up in a fright, confessed his sins to the Almighty, and fell asleep
again. The second dream ended with a piercing noise like a clap of
thunder. On opening his eyes,
he per::eived a large number of fiery sparks all around him in the room. This had often
happened to him at other times; it was nothing extraordinary for him to wake in the
middle of the night and find his eyes sparkling to such a degree as to give him glimpses
of the objects nearest to him.37
Quite a feat, even for a philosopher!

The third dream, which followed fast upon the second, was peaceful
by contrast. He saw a book that he took for a dictionary, and a
collection of poems, which he opened at random. He fell on a poem by
Ausonius which began "Quod vitae sectabor iter?" At that moment an
unknown person handed him another poem which began with "Est &
Non". It then occurred to Descartes, in the midst of this dream, to ask
himself whether he was dreaming. He not only concluded that he was
dreaming but he started to interpret the dreams while still asleep. When
he awoke, "he continued the interpretation of the dream on the same
lines". Note the continuity between the sleeping and the waking state.
87
He judged the dictionary to be the "sciences gathered together", the

collection of poems, "the union of philosophy and wisdom", and the
poets assembled in the collection, the "revelation and inspiration, by
which he hoped to see himself favoured". The "Est & Non" of the
poem he interpreted as "the Yes and No of Pythagoras", meaning truth
and error in human knowledge and the secular sciences. The clap of
thunder that he heard in the second dream he took to be "the signal of
the Spirit of truth descending to take possession of him". Lest this new
Pentecost be greeted like the first one with jeers ("They are filled with
new wine", Acts 2, 13), Descartes, like Peter before him, affirms that he
was not drunk, having "passed the evening and the whole day in a
condition of complete sobriety", and, for that matter, not having
touched wine for three months!
I do not believe that Descartes had forgotten all this when he wrote
the Discourse on Method. Granted that we remember what we want to
remember and that we cut and edit our memories with the ruthlessness
of a Russian censor, we cannot seriously entertain the notion that, from
the rationalist standpoint he had reached in 1637, the past was so
distorted that a breakdown occurred somewhere along the memory
line. Descartes was not the kind of person psychoanalysts love to see
lying on their couch and, more cogently, he kept the Dream Manuscript
with him for thirty years.
I believe it is much more profitable to raise the question whether
Descartes deliberately chose the dream as a literary device that allowed
the use of symbols that would appear incongruous if the narrator was in
the waking state. Every educated person in the seventeenth century was
familiar with the Dream of Scipio and knew from both the biblical
and classical sources that God communicates with men in dreams. As
a poetico-philosophical device it is not uncommon in the sixteenth
century.38 But what about the seventeenth century, and, to return to our
topic, what about the Rosicrucian treatises? I have already mentioned
Rodophilus Staurophorus' Raptus philosophicus published in German
in 1619. This is the account of the dream of a young man at the
crossroads. He wonders what path to take and - one readily guesses chooses the straight and narrow. After several dangerous incidents, he
meets a young woman who asks him, "Where are you going? What
Spirit brings you here?", and shows him a book "that contained all that
is in earth and in heaven but not ordered methodically". A young man,
dressed in white, then reveals to him that this woman "is Nature ... at
88
WILLIAM R. SHEA
the present time unknown to scientists and philosophers" (quoted

[2]160-1).
Henri Gouhier acknowledges the resemblances between this work
and Descartes' dream but he dismisses them as "une influence purement omamentale" ([19]140). In the light of Descartes' subsequent
intellectual development this sounds plausible, but one cannot help
wondering whether Gouhier is not making too little of what was a brief
but not necessarily a superficial phase in Descartes' evolution. Baillet
explicitly states that Descartes had been expecting significant dreams
for several days.
He [DescartesJ adds that the Genius which had been exciting in him the enthusiasm
with which, as he felt, his brain had been inflamed for several days had predicted these
dreams to him prior to his retiral to rest, and that the human mind had no share in
them ([14J 10, 186;[21J 38).
Our acquaintance with the vast Renaissance literature on dreams is

too slight to allow us to see Descartes' account against the background
that was probably so familiar in his day as not be be considered worth
mentioning. Francesco Trevisani has drawn attention to Cardano's
symbolic interpretations of dreams, and there are scores of such works
to be explored ([38]). In 1599, Giovanni Battista Nazari published his
Della tramutatione matellica sogni tre in which the author hears a
rustling of weeds calling out: "Quo viator iter, tu avaritia dementis"
([29110). A famous contemporary of Descartes, J. B. van Helmont,
gives 1610 as the date of his dream of enlightenment:
In the year 1610 after a long weariness of contemplation, that I might acquire some
gradual knowledge of my own minde, since I was then of opinion, that self-cognition
was the Complement of wisdom, fallen by chance into a calm sleep, and wrapt beyond
the limits of reason, I seemed to be in a Hall sufficiently obscure. On my left hand was
a table, and on it a faire large Vial, wherein was a small quantity of Liquor: and a voice
from that liquor spake unto me: "Wilt thou Honour and Riches?" ([20J 123).
Descartes went on to become the father of modern rationalism but

he never wavered in his belief in the role of inspiration. To Princess
Elizabeth with whom he entertained a free and easy correspondence he
expressed himself, as late as 1646, clearly on the subject. After
mentioning that a happy frame of mind makes things appear brighter,
he pursues:
89
I even make bold to believe that inner joy has a secret force to make Fortune more
favourable. I would not write this to people with weak minds lest they should be led to
superstition, but, in the case of your Highness, I rather fear that she will laugh at my
credulity. But I have a very large number of experiences, as well as the authority of
Socrates, to confirm my opinion. My experience is such that I have often noticed that
what I undertake gladly and without an interior repugnance generally goes well even in
games of chance where Fortune alone holds the sway.39
Descartes not only believed in his inner voice, he had "a very large
number of experiences" (the French has "une infinite d'experiences")
that confirmed its reliability.
The young Descartes' interest in hermeticism helps us understand
why he attached such deep significance to his threefold dream. I believe
it also casts light on his ready and willing enthusiasm for a cosmic
scheme that would embrace the stars. Descartes was to become more
cautious as he grew older, but as late as 1632, when he was thirty-six
years old, he could write to Mersenne as someone who is about to fulfil
the abiding dream of the astrologers:
In the last two or three months, I have penetrated deeply into the heavens and, having
satisfied myself about its nature and that of the celestial bodies that we see, as well as
many other things that I would not even have dared to hope a few years ago, I have
become so bold that I now dare to seek the cause of the position of each star. Even if
they appear to be strewn haphazardly in the heavens, I do not doubt that there is a
regular, fixed and natural order among them. Knowledge of this order is the key and
the foundation of the highest and most perfect science that man can have about natural
things. The more so, since, by its means, we could know a priori all the varied forms
and essences of terrestrial bodies, whereas, without it, we have to rest content with
guessing them a posteriori and by their effects.40
After stating that the observation of comets and an historical

catalogue of the position of heavenly bodies would be of assistance,
Descartes reverts, but this time with qualms in his voice, to his
grandiose project:
I think it is a science that surpasses human understanding, but I have so little wisdom
that I cannot help myself from dreaming about it, even though I believe it would only
be a waste oftime. 41
The grand scientific scheme, the great instauration that Descartes

nurtured, was something that he shared with the authors of several
Rosicrucian tracts. His answer was novel but the goal and the quest
were less so. The mechanical philosophy had to fit, however uneasily,
90
WILLIAM R. SHEA
into a framework of other things that Descartes, like most of his

contemporaries, believed he understood, and about which he spoke but
little.
But granted that Descartes shared concerns (many of which were far
from purely intellectual) with individuals whose assumptions, from our
vantage point, are at variance with the main thrust of his mature
philosophy, this does not automatically imply that he was influenced by
hermetic tenets.
It is always difficult to trace the intellectual development of someone
who has reconstructed his past as lucidly as Descartes did in the
Discourse on Method. I would like to suggest, however, that there is at
least one passage in his notebook that helps us understand why he
developed and resolutely maintained the absolute incorrigibility of the
view that some motions are instantaneous. This is the statement that
"instantaneous activity signifies creation" which occurs in a note (cited
above on page 74) that purports to indicate "how things which are
perceivable by the senses are helpful in enabling us to conceive of
Olympian matters" ([14) 10,218; [13) 1,5).
The notion of instantaneous activity plays a fundamental role in
Descartes' metaphysics as well as in his natural science. The basic
intuition of the self, the celebrated "cogito, ergo sum", is not arrived at
by a deduction but is perceived in an instantaneous flash of insight. 42
God can be said to be the cause of himself (causa sui) because the
notion of instant excludes any temporal priority.
In physics, Descartes not only repeatedly affirmed that light was
propagated instantaneously but he stressed that it was essential to his
system. In a letter written in 1634, he states, "this is so certain, that if
anyone could prove it false, I am ready to confess that I know nothing
at all in philosophy". A few lines later, he repeats that he is prepared to
stake his entire system on the outcome of an experiment to determine
whether the velocity of light is finite, "I say that if such a delay were
found, the whole foundation of my philosophy would be overturned".43
Instantaneity reappears in Descartes' analysis of falling bodies and
precludes a correct understanding of Galileo's famous law of free fall (s
ex: t 2) which rests on the assumption that an accelerated body goes
through all degrees of speed. According to Descartes a falling body
"has a determined speed from the first moment".44 To Mersenne, who
had asked him how a body could acquire a given speed in the first
91
moment of descent, he offered an embarrassed reply in which he

claimed that
it follows from certain principles that are obvious to me although, as I have often told
you, I could not explain them without a long discourse that I shall probably never
write. 45
It is interesting, in the context of our inquiry, to note how the

instantaneous acquisition of a degree of speed appeared to Descartes to
be obvious yet unexplainable.
The principle of instantaneity governs the rules of motion that
Descartes formulated in the Principles of Philosophy. Rule one, for
instance, states that if two identical bodies, moving at the same speed,
collide, they are reflected backwards along their initial paths not
because of an elastic collision but because their directions are instantaneously changed. The problematic status of the concept of instantaneity
and its attendant notion of "quantum jumps" of speed is revealed if we
compare rules four and five. 46 Rule four states that a small body,
however great its speed, will never be able to move a larger one. Rule
five states that a large body, however small its speed, will always move a
smaller one. Now let us (a) increase the small body in rule four until it
is almost the same size as the larger one, and (b) increase the smaller
body of rule five until it is virtually as big as the large one. We should
then expect the collision in the two cases to be similar. But according to
Descartes, the (however infinitesimally) smaller body considered in rule
four will still backtrack while the (however infinitesimally) larger body
considered in rule five will continue to move forward and impart half
of its motion to the (however infinitesimally) smaller body. Leibniz
pointed out that this violated the law of continuity in nature and only
makes sense if motion can be both instantaneously and arbitrarily
transmitted ([231 4,378).
However difficult the principle may have been to explain, Descartes
consistently maintained the reality of instantaneous activity. This was
one way of dispensing with the Aristotelian notion of potency on which
he poured scorn. There is no unfolding or deployment in the Cartesian
universe. Time is pulverized into instants and reduced to "parts that are
mutually independent". Indeed nothing would exist for more than an
instant unless God "continuously, as it were, produced it again, that is,
preserved it".47 To Descartes it is self-evident that the distinction
92
WILLIAM R. SHEA
between preservation and creation is only a conceptual one.48 Hence

God preserves, i.e., creates, things precisely as they are at a given
instant without considering their prior state.49 There is, as such, no
dynamism in matter. The law of inertia is a consequence of God's
instantaneous activity and the resulting cosmic actualism. This is clear
from The World that Descartes wrote between 1629 and 1633 and in
which the law is formulated for the first time. A body normally moves
in a curved path, says Descartes, yet "each of its parts individually tends
always to continue moving along a straight line". This rule depends
solely on God preserving each thing "not as it may have been some
time earlier but precisely as it is at the very instant that he preserves it".
It follows, adds Descartes, that only motion in a straight line is simple
and comprised in an instant:
For in order to conceive such a motion, it suffices to think that a body is in the process
of moving in a certain direction, and that this is the case at each determinable instant
during the time it is moving. By contrast, in order to conceive circular motion, or any
other motion, it is necessary to consider two of its instants, or rather two of its parts,
and the relation between them,5!'
Descartes then offers an analysis of motion in a sling, and concludes:

According to this rule, then, it must be said that God alone is the author of all the
motions in the world in so far as they exist and in so far as they are rectilinear; but it is
the various dispositions of matter which render them irregular and curved.
Circular motions would appear to bear the mark of cosmic sin! In

the very next sentence, they are compared to moral misdeeds:
Likewise, the theologians teach us that God is also the author of all our actions, in so
far as they exist and in so far as they have some goodness, but it is the various
dispositions of our wills that render them evil. 5 I
The instantaneous activity of God is also invoked in the establishment of the imaginary world, the fable that is meant to be truer to
nature than the allegedly realistic accounts of the philosophers. The
matter that God creates is perfectly solid and homogeneous. 52 But such
a matter is essentially inert and unable to give rise to change. For
motion to spread and produce a division in the matter it would already
have to be fluid, but matter can only become fluid through motion! The
whole thing looks circular! Not so for Descartes since God "from the
first instant" divides matter into parts that are distinguished by their
93
various motions. 53 But the paradox of matter as an homogeneous block

and the instantaneous production of motion is not fully lifted since, a
few pages later, Descartes repeats that we must imagine original matter
"as the hardest and most solid body in the world" while maintaining
that it was set in motion "at the same instant".54 The homogeneous
block is reduced to a pedagogical fiction. More radically, since what
really happens, happens instantaneously, all models of development can
only be likely stories. Strictly speaking, God has no memory, and
history does not unfold.
It would seem, therefore, that the early hermetic passage "instantaneous activity signifies creation" played some role in the genesis of
Descartes' system. This does not mean that Descartes was an unrepentant Rosicrucian but that his rational philosophy was broader and more
comprehensive than the watered-down versions of some of his wouldbe followers. There are more things in the Cartesian heaven and earth
than are dreamt of in some philosophies.
NOTES
I The author wishes to thank the Social Sciences and Humanities Research Council of
Canada for its generous support of his research. He is also grateful to the editor of the
Annali dell' lstituto e Museo de Storia della Scienze for permission to use material that
appeared in his journal.
2 The Farna and the Confessio are conveniently reprinted in the English translation
given by Thomas Vaughan in 1652 as an appendix to [43]279-306. Lenglet de
Fresnoy lists 43 works concerning the Rosicrucians published between 1613 and 1619,
all in Germany. In 1619-20, 15 mor eappeared. The first French works date from
1623-4 (Histoire de fa philosophie herrnetique (Paris, 1742) 3, pp. 279-87, numbers
651-705). On the relation of these tracts to the Paracelsian movement see the
scholarly [12]1, 211ff. For the climate in Italy see [39]. For an excellent account of
Descartes' mature views in the light of recent developments in philosophy of science,
see [10]. The chapter on Descartes in Gerd Buchdahl's [6] is essential reading. This
essay stems from Buchdahl's suggestion that we should sometimes look under the
furniture ...
3 Farna Fratemitatis in [43] 284.
4 [4]1, 107. Descartes was always a very retiring person. When he visited Paris he
usually stayed at the house of Nicolas Le Vasseur, Seigneur d'Etioies, a friend of his
father, but in 1628 he sought lodgings elsewhere to avoid importunate visitors and
"make himself visible to only a very small number of friends" ([4]1, 153). Later, in the
Netherlands, he became notorious for his love of privacy. Here is what a fellow
Frenchman has to say about Descartes' visit to Leyden to supervise the printing of the
Discourse on Method in 1637: "He has been in town since they began printing his
94
WILLIAM R. SHEA
book, but he hides and only shows himself very rarely. In this country, he always lives
in some small and remote town. Some say that this is how he got the name d'Escartes
for he used to be called something else" (letter of Claude de Saumaise to M. de Puy, 4
April 1637, cited in [14]10, 555-6). When he first arrived in Holland, Descartes
called himself du Perron, e.g., in his correspondence with Isaac Beeckman in 1619
([14] 10, 153, 160, 161, 164, 166). The physician Vopiscus-Fortunatus Plempius states
that (around 1629) Descartes lived in the Kalverstraat in Amsterdam "unknown to
anyone (Nulli notus)" ([14]1, 401).
5 Fran<;:ois Garasse, "La doctrine curieuse des beaux esprits de ce temps", cited in
[2]137-8.
6 John Dee, The Mathematical Preface to the Elements of Geometrie (1570), quoted in
[42] 31.
7 See [30] Book 17, chapters 2, 4, 5, 6, 7, and 10 discuss this kind of optical tricks.
The first edition of Magia naturalis sive de miraculis rerum naturalium in four books
appeared in Naples in 1558 and was reprinted at least thirteen times between 1560 and
1588, not counting the numerous reprints of the Italian, French and German versions.
The second edition in twenty books was published in Naples in 1589, and was
frequently reprinted, as well as translated into Italian, French, German and English. See
[27].
H Compare, for instance, the following passages from the Cogitationes privatae and the
Magia naturalis: (1) [14]10, 209 and [30] bk. 20, ch. 9, p. 408 on making bystanders
appear coloured; (2) [14]10, 244 and [3D] bk. 16, ch. 2, p. 341 on making invisible
writing legible by heating the paper; (3) [14]10, 244 and [30] bk. 15, ch. 5, p. 332 on
fishing with a submerged candle; (4) [14]10, 232 and [30] bk. 20, ch. 10, p. 409 on a
mechanical dove that can be made to fly.
~ After musing on the possible applications of parabolic mirrors that "burn at an
infinite distance", della Porta concludes: "I have observed, that we may use this Artifice
in great anG wonderful things, and chiefly by inscribing letters in a full Moon" ([30] bk.
17, cit. 17, p. 376). Agrippa speaks of raising letters and reading them on the disk of
the moon "as was done of old by Pythagoras" but he does not claim that he knows how
to do it (Henricus Cornelius Agrippa of Nettesheim, De occulta philosophia, bk. 1, ch.
6, cited in [14]10, 347 from the Opera omnia (Lyons, 1600) 1, 347. De occulta
philosophia was first published in 1531.
10 See the diary of Isaac Beeckman cited in [4]10, 347. In the same year, Mersenne
expressed to Galileo the hope that the new telescope would reveal whether living beings
existed on the moon (letter of 1 February 1629 ([26] 2, 175-6.
11 Letter to an unknown correspondent, September 1629 (?) ([14]1,21).
12 Les Mtiteores, 8th Discourse ([14] 6,343). See 131]461-74.
L1 Letter to Beeckman, 29 April 1629 ([14]10, 165) and Beeckman's reply ([14]10,
167-8). Later, in the Discourse on Method, Descartes gave no inkling of his early
interest but disparaged "the art of Lully" as merely serving "for speaking without
judgement about matters of which one is ignorant" ([14] 6, 17; [13]1, 119). Further
evidence of Descartes' interest in the art of memory is a long note on Lambert
Schenkel, a popular writer on mnemonics ([14]10, 230). Paolo Rossi remarks that
Descartes accepts both Schenkel's terminology and his way of raising the problem
([32]154-5), and Frances Yates adds that Descartes' "new idea of organizing memory
on causes sounds curiously like a rationalization of occult memory" ([41]374).
95
Cogitationes privatae ([14)10, 215).

[14)10,373; [13)1, 17. A few pages later, Descartes reiterates his conviction "that
certain primary seeds of truth naturally implanted in human minds thrived vigorously in
that unsophisticated and innocent age - seeds which have been stifled in us through
our constantly reading and hearing all sorts of error" ([14)10, 376; [13)1, 18). In the
Second Set of Replies to the queries raised about his Meditations, Descartes explains
that ancient geometers used synthesis but not analysis in their published works not
because they were ignorant of the latter but "because they had such a high regard for it
that they kept it to themselves like a sacred mystery" ([14]8, 156; [13)2, 3). In the
Discourse on Method, Descartes states that he derived the general principles or first
causes of everything "only from certain seeds of truth which are naturally in our souls"
([14) 6, 64; [13)1,44). On the sources ofthe notion of "seeds oftruth" see [19) 93-4.
16 Nouveaux memoires pour servir Ii [,histoire du cartesianisme par M. G. de 1'A.
(initials of Gilles de I'Aunay) (No place, 1692),42, cited in [19)128.
i1 [14)10, 197 note a. The first suggestion to appear in print that Descartes might be a
Rosicrucian occurs in a vitriolic attack on him entitled Admiranda methodus novae
philosophiae Renati Descartes published in Utrecht in 1643. The book is anonymous
but is likely the work of Martin Schook who wrote the Preface. The author says that he
does not believe the imputation because Descartes is too vain to accept the rule of
silence ofthe Brotherhood ([14)8.2, 142 note b).
18 Descartes gave medical advice to Mersenne for their mutual friend Claude Clerselier
but requested that it be imparted to those who attended him ''without their knowing in
any way that it comes from me (letter to Mersenne, 23 November 1646 ([14)4, 566.
In the same year he gave advice to an acquaintance who suffered from nose-bleeding
with the same proviso ([14)4, 698).
19 Letter to Huygens, 25 January 1638 ([14)1, 507).
20 [4)2, 452-3. In 1629 Descartes turned seriously to the study of medicine (letter to
Mersenne, 18 December 1629 ([14)1, 102 which he hoped to "base on infallible
demonstrations" (letter to Mersenne, January 1630 ([14)1, 106. When he wrote the
Discourse on Method in 1637 he expressed great hopes in the future of medicine: "if it
is possible to find some means of making men in general wiser and more skilful than
they have been up till now, I believe we must look for it in medicine" ([14)6, 62; [13)1,
143). The following year, 1638, he was at work on a treatise of medicine (letter to
Huygens, 25 January 1638 ([14)1, 507, and, in 1645, he confided to the Marquis of
Newcastle: "The preservation of health has been at all times the main goal of my
studies" ([14)4, 329). According to Des Maizeaux, Kenelm Digby visited Descartes in
the Netherlands around this time and urged him to devote his efforts to finding means
of prolonging life. To which Descartes replied "that he had thought about the matter,
and that although he could not promise to render man immortal he was certain that he
could render his life as long as that of the Patriarchs" (Des Maizeaux, La Vie de
Monsieur de Saint-Evremond, written in 1706, cited in [14]11, 671). Since, with the
exception of Enoch who departed from the earth at 365 years of age, the ages of the
Patriarchs vary from 777 (Lamech) to 969 (Methuselah), Descartes may be said to have
had high expectations! These were not fulfilled and by the Spring of 1648, when he was
interviewed by the young Frans Burman, he had given up the hope of reaching such a
ripe old age: "How life was so vastly extended before the Flood is something that is
beyond philosophy" ([14]5, 178). He never abandoned, however, his conviction that
14
15
96
WILLIAM R. SHEA
life could be extended: "We must not doubt that human life could be prolonged if we
knew the proper art" ([14]5, 178). When Descartes died in Stockholm at the age of
fifty-three, the Gazette of Antwerp scoffed at his alleged claims of longevity: "A fool,
who said he could live as long as he wished, has died in Sweden" (quoted in a letter of
Christiaan Huygens to his brother Constantijn, 12 April 1650 ([14]10, 630. The
prolongation of life was discussed by several authors since Roger Bacon in the
thirteenth century. Bacon referred to Scripture to prove the possibility of living almost
a thousand years. Like Descartes he emphasised a sane and moderate diet. For a short
survey of Bacon's views, see his chapter "De retardatione accidentium senectutis, et de
prolongatione vitae humanae" in his Epistola de secretis operibus artis et naturae et de
nul/itate magiae ([3]1, 538-42).
21 "He was never untidy, and he especially avoided dressing like a philosopher" ([4]2,
447).
22 Letter to Mersenne, April 1634 ([14]1, 286).
23 Cogitationesprivatae ([14]10, 213; [13]1, 2).
24 Letter to Mersenne, 20 November 1629 ([14]1, 81-2). On the history of such
attempts see [32], [22], [36].
25 [30] bk. 20, ch. 7, p. 403. Della Porta ascribes this view to Pythagoras. The earliest
source that I have come across is Fracastoro's book on Sympathy and Antipathy of
1550 where he writes that "striking a drum made of the skin of a wolf will, they say,
break drums made of the skin of lambs" ([17] 22). There are many variants on this
theme. For instance Burton applies it to Jan Zizka, the fifteenth-century national hero
of Bohemia: "the great captain Zisca would have a drum made of his skin when he was
dead, because he thought the very noise of it would put his enemies to flight" ([7]1, 38).
26 After stating that he has explained all the properties of magnets and fire by "the
shape, size, position and motion of particles of matter", Descartes adds: "And anyone
who considers all this will readily be convinced that there are no powers in stones and
plants that are so mysterious, and no marvels attributed to sympathetic and antipathetic
influences, that are so astonishing, that they cannot be explained in this way" (Principles
of Philosophy, part 4, article 187 ([14]8.1, 314-15; [13]1, 279. The French version
of the Abbe Picot adds the following marvels not mentioned in the original Latin
version and said to result from the motion of parts of the first element: "making the
wounds of a dead man bleed when the murderer draws near, or exciting the imagination
of those who are asleep, or even awake, and giving them ideas that warn them of what
is happening far away" ([14] 9.2, 309).
27 [30] bk. 1, ch. 13, p. 19. I have modernized the translation.
28 [30] bk. 8, ch. 14, p. 230. I have modernized the translation.
29 [14]11, 602. The passage in Aristotle that is the source of the remarks of both della
Porta and Descartes occurs in On Dreams II, 459 b 28-31: "If a woman looks into a
highly polished mirror during the menstrual period, the surface of the mirror becomes
clouded with a blood-red colour" ([1]357).
30 Cogitationesprivatae ([14]10, 214; [13]/, 2).
31 Letter to Huygens, March 1638 ([14]2, 48).
32 [191 110. I find no irony in the following remark which Descartes jotted down in his
notebook: "The sciences are at present masked, but if the masks were taken off, they
would be revealed in all their beauty. If we could see how the sciences are linked
together, we would find them no harder to retain in our minds than the series of
numbers" ([14]10,215; [13]1, 3).
97
33 Excerpta mathematica ([14]10, 297). On the problems that Descartes experienced

in articulating a "universal mathematics", see [34]. For Descartes' mathematics prior to
the Discourse on Method, see [11]27-59. On the "uncertain" metaphysics of the Rules
for the Direction of the Mind, see [24].
34 Unfortunately, I have not been able to locate a copy of Faulhaber's work, whose
earliest bibliographical appearance would seem to be Kooss' Bibliographie der Freimaurerei (Frankfurt, 1844) under number 2.452 (according to [19]132). On the
probability of a personal acquaintance between Descartes and Faulhaber, see
[25]196-200. Pierre Costabel has shown that Descartes could have been inspired by
Faulhaber's method of the gnomon when he wrote, sometime between 1620 and 1628,
his highly original essay on polyhedra ([16]52-6,89-90). In the Cogitationes privatae,
Descartes mentions the mathematicians Peter Roth and Benjamin Bramer who were
acquaintances of Faulhaber and worked on the same kinds of mathematical problems
([14]10, 242).
35 Daniel Lipstorp, Specimina philosophiae Cartesianae (Leiden, 1653), pp. 78-9,
mentioned in [15]156. Adrien Baillet, in his two volume Vie de Monsieur Des-Cartes
published in 1691, does not identify Descartes' winter retreat in 1619, but has him
sojourn in Ulm from the end of June to the beginning of September ([4]1, 96). In the
abridged version that he published a year later, he states, without further explanation,
that Descartes took up his winter quarters in the Duchy of Neuburg in October 1619
([5]39; [5a] 33). Neuburg is not near Ulm but is situated on the Danube in Northern
Bavaria a few kilometers to the west of Ingoldstadt.
36 [4]1,81-8; in [14]10,181-8. This passage is translated in [21]33-40. Descartes'
original Latin manuscript bore the title "Olympica". It is not clear what "Olympica"
meant for Descartes, but the word belongs to the hermetic and Paracelsian tradition. In
a Latin book on the magic of the Ancients published under the name of Astrabe1 in
Basel in 1575, we find a list of nine kinds of magic. These include Olympian magic
along with Hesiodic, Pythagorean and Hermetic ([37]6, 457).
37 [14]10, 182; [21]35. In the Optics, Descartes states that light streams out of the eyes
of cats, and he implies that this is also possible in men who rise above the ordinary namely exceptional people like himself (Dioptrique, Discours 1 ([14]6,86. According
to Sextus Empiricus, "Tiberius could see in the dark" (Outline of Pyrrhonism, bk. 1,
chs. 14,84 ([35]1, 51.
38 See [8]147-8, and notes on 154; [40]. In Cicero's widely read De divinatione,
Quintus, Cicero's interlocutor remarks: "the Stoic view of divination smacked too
much of superstition. I was more impressed by the reasoning of the Peripatetics, of
Dicaearchus, of ancient times, and of Cratippus, who still flourished. According to their
opinion there is within the human soul some sort of power - 'oracular' I might call it by which the future is foreseen when the soul is inspired by a divine frenzy, or when it
is released by sleep and is free to move at will" ([9]483).
39 Letter to Elizabeth, November 1646 ([14]4, 529).
4() Letter to Mersenne, 10 May 1632 ([14]1, 250-1).
41 [14]252. Note how Descartes cannot help himself from dreaming ("je ne s~aurois
m'empescher d'y rever").
42 The most explicit passage occurs in the Second Set of Replies to the Meditations
([14] 7,140; [13]2,100).
43 Letter to Beeckman (1), 22 August 1634 ([14]1, 307-8). See [11]77-86; [33]112.
98
WILLIAM R. SHEA
44 Letter to Mersenne, 25 December 1639 ([14]3, 630). Descartes had made the same
point in an earlier letter to Mersenne, 11 October 1638 ([14)3, 399).
45 Letterto Mersenne, 4 November 1630 ([14]1,176).
46 Principles of Philosophy, part 2, arts. 46, 49,50 ([14]8.1, 68-9).
47 Principles of Philosophy, part 1, art. 21 ([14]8.1, 13).
48 Third Meditation ([14] 7,49; [13]2, 33).
49 In the case of motion, Descartes explicitly says "nulla habita ratione ejus qui forte
fuit paulo ante" (literally, "without considering the motion that was perhaps shortly
before") (Principles of Philosophy, part 2, art. 39 ([14]8.1, 64.
50 Le Monde ([14]10, 45; [13]1, 97).
51 [14] 10,46;[13] 1, 97. Underlining mine.
52 [14]10, 31; [13]1 omits this passage.
53 ]14]10,34; [13]1, 9l.
54 [14]10,49; [13]1 omits this passage.
REFERENCES
l. Aristotle, On the Soul. Parva naturalia. On Breath, tr. W. S. Kett (London:

Heinemann, 1975).
2. Arnold, P., La Rose-Croix et ses rapports avec la Franc-Mat,;onnerie (Paris: G.-P.
Maisonneuve et Larose, 1970).
3. Bacon, Roger, Opera quaedam hactenus inedita, 3 vols, ed. J. S. Brewer (London,
1859).
4. Baillet, Adrien, Vie de Monsieur Des-Cartes, 2 vols (Paris, 1691).
5. Baillet, Adrien, Abrige de la vie de M. Descartes (Paris, 1692; 5a: reprinted [ParisI:
La Table Ronde, 1946).
6. Buchdahl, Gerd, Metaphysics and the Philosophy of Science: The Classical Origins,
Descartes to Kant (Oxford: Blackwell, 1969).
7. Burton, Robert, The Anatomy of Melancholy (1st ed. 1621; 6th ed. 1651 reprinted
in 3 vols, London: Dent, 1932).
8. Chaste!, Andre, Marsile Ficin et l'art (Geneva: Droz, 1954).
9. Cicero, M. T., De senectute. De amicitia. De divinatione, tr. W. A. Falconer
(London: Heinemann, 1923).
to. Clarke, Desmond, M., Descartes' Philosophy of Science (Manchester: Manchester
Univ. Press, 1982).
11. Costabel, Pierre, Demarches originales de Descartes savant (Paris: Vrin, 1982).
12. Debus, Allen G., The Chemical Philosophy, 2 vols (New York: Science History
Publications, 1977).
13. Descartes, Rene, The Philosophical Writings of Descartes, 2 vols tr. J. Cottingham,
R. Stoothoff, and D. Murdoch (Cambridge: Cambridge Univ. Press, 1985).
(Paris: Vrin, 1964-76).
15. Descartes, Rene, Discours de la methode, ed. Etienne Gilson (3rd ed., Paris: Vrin,
1967).
16. Descartes, Rene, Exercices pour les elements des solides. De solidorum elementia,
tr., ed., and with a commentary by Pierre Costabel (Paris: Vrin, 1987).
99
17. Fracastoro, H., De sympathia et antipathia rerum (Lyons, 1550).

18. Gilson, Etienne, Etudes sur Ie role de la pensee mMievale dans la formation du
systeme cartesien (Paris: Vrin, 1967).
19. Gouhier, Henri, Les premieres pensees de Descartes (Paris: Vrin, 1958).
20. van Helmont, J. 8., Ternary of Paradoxes, tr. Walter Charleton (London, 1650).
21. Kemp Smith, Norman, New Studies in the Philosophy of Descartes (London:
Macmillan, 1952).
22. Knowlson, James, Universal Language Schemes in England and France 1600-1800
(Toronto: Univ. of Toronto Press, 1975).
23. Leibniz, G. W., 'Animadversiones in partem generalem principiorum Cartesianorum', in Die philosophische Schriften, 7 vols, ed. C. I. Gerhardt (Berlin and
Leipzig, 1875-90).
24. Marion, Jean-Luc, Sur l'ontologie grise de Descartes (Paris: Vrin, 1975).
25. Marion, Jean-Luc, Sur la theologie blanche de Descartes (Paris: Presses Univs. de
France, 1981).
26. Mersenne, Marin, Correspondance du Pere Marin Mersenne, 13 vols, ed. C. de
Waard (Paris: Presses Univs. de France, and Centre National de la Recherche
Scientifique, 1945-77).
27. Muraro, L., Giambattista Della Porta mago e scienziato (Milan: Feltrinelli, 1978).
28. Naude, Gabriel, 1nstruction Ii la France sur la verite de l'histoire des Freres de la
Rose-Croix (Paris, 1623).
29. Nazari, Gio. Battista, Della tramutatione metallica sogni tre (Brescia: Pietro Maria
Marchetti, 1599).
30. della Porta, John Baptista, Natural Magick, tr. anon. (London, 1658; reprinted in
facsimile New York: Basic Books, 1957).
31. Rodis-Lewis, Genevieve, 'Machineries et perspectives curie uses dans leurs rapports avec Ie Cartesianisme', XVIle Siecle 32 (1956),461-74.
32. Rossi, Paolo, Cia vis Universalis (Milan: Riccardo Ricciardi, 1960).
33. Sakellariadis, Spyros, 'Descartes' experimental proof of the infinite velocity of light
and Huygens' rejoinder', Archive for History of Exact Sciences 26 (1982), 1-12.
34. Schuster, John A., 'Descartes' Mathesis Universalis: 1619-28', in Descartes:
Philosophy, Mathematics and Physics, ed. Stephen Gaukroger (Brighton: Harvester
Press, 1980), 41-96.
35. Sextus Empiricus, Sextus Empiricus, 4 vols, tr. R. G. Bury (London: Heinemann,
1933-49).
36. Slaughter, Mary M., Universal Languages and Scientific Taxonomy in the Seventeenth Century (Cambridge: Cambridge Univ. Press, 1982).
37. Thorndike, L., A History of Magic and Experimental Science, 8 vols (New York:
Columbia Univ. Press, 1923-58).
38. Trevisani, Francesco, 'Symbolisme et interpretation chez Descartes et Cardan',
Rivista Critica di Storia della Filosofia 30 (1975), 27 -4 7.
39. Va soli, Cesari, Profezia e Regione (Naples; Morano, 1974).
40. Wagner, lean-Marie, 'Esquisse du cadre divinatoire des songes de Descartes',
Baroque 6 (1973), 81-95.
41. Yates, Frances, A., The Art of Memory (Chicago: Chicago Univ. Press, 1966).
42. Yates, Frances, A., Theatre of the World (Chicago: Chicago Univ. Press, 1969).
43. Yates, Frances, A., The Rosicrucian Enlightenment (London: Paladin, 1975).
STEPHEN W. GAUKROGER
DESCARTES' CONCEPTION OF INFERENCE 1
Abstract. Descartes rejected traditional syllogistic and attempted to replace it with a

doctrine of method which construes questions of inference in terms of the natural light
of reason. Taken with his doctrine that eternal truths are created by, and not merely all
known by, God, the upshot seems to be that Descartes has grossly misunderstood the
nature of deductive inference and logical truth, and because of the great influence of his
views on method over the next two centuries, he is often blamed for leading logic
astray. The argument of this essay is that, if we take care to identify precisely the
notions of inference that Descartes was opposing - namely humanist and especially
late scholastic notions - then we can see that this is not the case, and that the doctrine
of eternal truths in fact allows Descartes to provide the first account of inference which
makes any serious attempt to deal with both the cognitive and the logical aspects of
inference.
INTRODUCTION
In common with many of his contemporaries, but with more success

than any of them, Descartes was concerned to replace traditional syllogistical, topical, dialectical - modes of reasoning with a method of
discovery, and he found a model for this method in the mathematical
procedure of analysis. The relation of Descartes' conception of method
to those of his predecessors and contemporaries, and especially his
successors, has received a good deal of attention, as has the relation
between this method and the classical procedures of analysis and
synthesis. Despite the fact that there is still much that we do not know
in these areas, and much disagreement in interpretation, the important
point is that the groundwork has been laid. 2 This much cannot be said
of Descartes' conception of inference, despite its obvious importance
for his methodological and epistemological concerns. My aim in this
essay is to try to remedy this situation.
It is of paramount importance here that we understand the context
within which questions of inference are raised between the sixteenth
and the nineteenth centuries. This period is usually written off (with an
exception being made for Leibniz) as an interregnum in the history of
logic 3 yet, while it is true that technical results are virtually non-
101
102
existent, there are profound changes in conceptions of inference which

are central to how we have come to think about logic. It is one such
change, which I believe to be the most fundamental one in this period,
that we shall be concerned with. This is the shift from a discursive to a
cognitive conception of inference, and my argument will be that
Descartes has a major role to play in this shift.
PROOF AND TRUTH
One of Descartes' principal objections to traditional syllogistic is that it

is an impediment to the conduct of our reasoning. In Rule 4 of the
Regulae, we are told:
But if our method rightly explains how intellectual intuition should be used, so as not to
fall into error contrary to truth, and how one must find deductive routes in order that
we may arrive at knowledge of all things, I cannot see anything else is needed to make it
complete; for I have already said that the only way science is to be acquired is by
intellectual intuition or by deduction. Method cannot be extended further so as to show
how these operations themselves should be effected, because they are the most simple
and primary of all, to the extent that, unless our understanding were already able to
make use of them, it could comprehend none of the precepts of that very method, not
even the simplest. As for the other operations of the mind, which dialectic claims to
direct by making use of these two, they are quite useless here; rather they are to be
accounted impediments, because nothing can be added to the pure light of reason
which does not in some way obscure it ([7110,372-3).
This "light of reason", or "light of nature" as it is called in Rule 10,

apparently cannot mislead us, as "none of the mistakes which men
make ... are due to faulty inference; they are caused merely by the fact
that we build upon the basis of poorly comprehended experiences, or
because hasty or groundless propositions are put forward" ([7110, 365).
What the light of reason does in the first instance is to allow us to
grasp the truth of clear and distinct ideas. But of course on some
occasions we have to connect such ideas inferentially, and then we
require demonstration or deduction. Descartes' account of this process
is, however, modelled upon intellectual intuition (intuitus):
Thus if, for example, I have first found out, by distinct mental operations, what relation
exists between the magnitudes A and B, then what between Band C, between C and
D, and finally between D and E, that does not entail that I will see what the relation is
between A and E, nor can the truths previously learned give me a precise idea of it
unless I recall them all. To remedy this I would run over them many times, by a
DESCARTES' CONCEPTION OF INFERENCE
103
continuous movement of the imagination, in such a way that it has an intuition of each
term at the same time that it passes on to the others, and this I would do until I had
learned to pass from the first relation to the last so quickly that there was almost no
role left for memory and I seemed to have the whole before me at the same time ([7[
10,521).
One way in which this passage has been taken is as a claim that
deduction has no real role to play in knowledge. Ian Hacking takes it in
such a way, assimilating Descartes' view to that of the mathematician G.
H. Hardy, who thought of proofs as "gas, rhetorical flourishes designed
to affect psychology ... devices to stimulate the imagination of pupils"
([19]).
Hacking supports his reading by appeal to the doctrine of eternal
truths. This doctrine, first elaborated in three letters to Mersenne of 15
April, 6 May and 27 May 1630, offers an account of God's grasp of
truths. The second letter presents the essentials of the doctrine:
As for the eternal truths, I say once more that they are true or possible only because
God knows them as true or possible. They are not known as true by God in any way
which would imply that they are true independently of Him. If men really understood
the sense of their words they could never say without blasphemy that the truth of
anything is prior to the knowledge which God has of it. In God willing and knowing are
a single thing in such a way that by the very fact of willing something He knows it and it
is only for this reason that such a thing is true.4
The central claim is elaborated upon in the third letter in these terms:
You ask what necessitated God to create these truths: and I reply that just as He was
free not to create the world, so He was no less free to make it untrue that all the lines
drawn from the centre of a circle to its circumference are equal. And it is certain that
these truths are no more necessarily attached to His essence than other creations are.
You ask what God did in order to produce them. I reply that from all eternity He willed
and understood them to be, and by that very fact He created them. In God, willing,
understanding, and creating are all the same thing without the one being prior to the
other even conceptually ([ 1O[ 15).
Hacking takes the doctrine of intuition and the doctrine of eternal

truths together as illustrations of an underlying conception of the
irrelevance of proof to truth. Construed in this context, the import of
the doctrine of eternal truths is that eternal truths depend upon the will
of God, who has no need of deduction (proof); He knows truths in
virtue of having created them (i.e., willed them), so proof is clearly
irrelevant. This doctrine then seems to mirror the doctrine of intuition
which, on Hacking's interpretation, maintains that we need only
intuition, and not deduction, in grasping truths.
104
There are a number of problems with this association of the two

doctrines. In the first place, they are developed independently. The
earliest appearance of the doctrine of intuition is Rule 3 of the Regulae,
which dates from around 1619.5 The doctrine of eternal truths, on the
other hand, only makes an appearance in 1630, in the letters to
Mersenne. Moreover, although the term intuitus tends to disappear
after the Regulae, the doctrine itself does not - it is to be found as late
as the 1640s in the Search after Truth,6 for example - yet this doctrine
is not altered after 1630 in any way which would suggest that it had a
connection with the new conception of eternal truths. Secondly, while
Descartes holds both doctrines after 1630, he never discusses them
together or even in the same context. As well as in the three letters to
Mersenne of 1630, the doctrine of eternal truths is discussed or
mentioned in letters to Mersenne of 17 May 1638 and to Mesland of 2
May 1644, and in the Replies to the Fifth and Sixth sets of Objections
to the Meditations and in the Principles (I, arts. 22-4 and 48-9). It
is hard to believe that, if the doctrines were simply part of the one
underlying conception, Descartes would have made no effort to discuss
them together or indeed to make any explicit connection between them.
Third, not only is there no textual reason to associate the doctrines in
the way that Hacking suggests, there are other grounds for believing
such an association to be mistaken. Hacking points out that Leibniz's
God knows all truths because He knows all proofs, whereas we only
know some because we only know some proofs, and we are in any case
restricted in our grasp of proofs to those which are finite whereas God
is not. But what is the parallel with Descartes here? Consider the
doctrine of intuition. The parallel that suggests itself on the basis of this
conception is one on which God has an intuitive grasp of all truths, but
we only have an intuitive grasp of a few. We would then be able to
conclude, as Hacking does, that, in general terms, proof is constitutive
of truth for Leibniz and irrelevant to truth for Descartes. But the whole
thrust of the doctrine of eternal truths is precisely that we cannot
compare what knowledge for us consists in and what knowledge for
God consists in. We are simply unable, on Descartes' view, to make any
connection at all between our intuition and God's cognitive grasp.
In discussing the doctrine of eternal truths, Descartes never raises
the question of deduction or proof, and this is the crucial point. He
nowhere maintains that proof is irrelevant to truth for God; rather, he
provides us with an account whereby God wills truths into existence, an
105
account which, if construed in a logical context, does indeed have this

as a consequence. But it is far from clear that Descartes thinks such
questions can be construed in a logical context, since it appears that we
can say nothing at all about what God's grasp of truth consists in.
Hence, if we are to understand the conception of inference that
Descartes offers, we must focus our attention on what I have called the
doctrine of intuition.
On the face of it, this is not a particularly attractive doctrine, and
even if we dissociate it from the doctrine of eternal truths, it has two
features which may appear to lend support to Hacking's low view of
Descartes' general conception of inference. First, in the limiting case,
deduction tends towards what is in effect the model for all reasoning,
intuition. The point of the exercise seems to be to reduce our inferential
steps altogether, so that one grasps the premisses and conclusion in the
one intuition. The role of demonstration or proof on this conception is
obviously problematic. Secondly, for Descartes, knowledge which we
have in an intuition is an immediate grasp of clear and distinct ideas
which Descartes construes explicitly as thoughts, thoughts which are
grasped in the first instance in their own right without any reference to
whatever extramental correlates they may have. So not only is deduction construed (in some way that we have yet to elucidate) in terms of
intuition, but intuition, and hence deduction, is construed psychologistically. Psychologism has not generally been taken seriously as a basis for
logic since Frege's famous attack on it ([131 paras. 26-7), and its faults
now seem as obvious to us as the faults of syllogistic seemed to
Descartes. What we need to come to terms with in understanding
Descartes is not just his psychologism, however, but more importantly
the issues that underlie his advocacy of psychologism.7 Psychologism is
simply the form taken by Descartes' attempt to provide what, I shall
argue, is a cognitive basis for inference. To appreciate what is at issue
here we need to take a broad view of the development of conceptions
of inference up to Descartes' time. Fortunately, this is not incompatible
with taking a selective view, and I shall be very selective. We will look
first briefly at Aristotle's conception of inference, and at how the
Aristotelian conception comes to be transformed in the early Middle
Ages, and then at the views of Descartes' immediate predecessors and
contemporaries. Although this means ignoring the very important Stoic
and terminist conceptions of logic, as well as many other less important
theories, I believe the selection provides us with a broad outline of the
106
central development, which I shall argue lies in a shift from discursive

to cognitive conceptions of inference. Seen in this light, Descartes is the
first to make a serious attempt to come to terms with a novel and
important but especially intractable problem about the cognitive basis
of inference.
CONCEPTIONS OF INFERENCE BEFORE DESCARTES
Aristotelian syllogistic was misunderstood in many respects in the

seventeenth century, both by its detractors and its ever-decreasing
number of advocates. The charges of circularity and question-begging
which were levelled against the syllogism, e.g., by Descartes and Locke,
depended to a large extent upon its being taken as an instrument of
discovery, which is something that Aristotle never intended.8 For
Aristotle, the demonstrative syllogism in particular was primarily a
didactic and expository device which provided an explanation of a
conclusion which was known in advance. The procedure for finding
such explanations was provided not by syllogistic, the concern of which
was formal and systematic presentation, but by the "topics". The topics
work by supplying strategies for classifying or characterizing problems
in such a way that they can be solved using set techniques. These set
techniques are techniques of argument or disputation and they are
initially developed in the context of dialectical argument, where they
function somewhat like the Sophists' procedures for disputation: they
help to enable one to discover what distinctions are to be made, what
route is to be followed, etc., if one is to get one's opponent to yield to
the case one is defending. But as Aristotle becomes progressively more
concerned with the formal properties of arguments and with scientific
demonstration, the topics come to be supplemented by a formal
account of the structure of arguments: syllogistic. They retain their role
as an instrument of discovery, but are superseded in many other
respects by syllogistic.
The pioneering work of Lukasiewicz ([32]) and others showing, from
the perspective of modem logic, the formidable formal strengths of
Aristotelian syllogistic, has tended to open up a gulf between the early
dialectical concerns of the central Books 2 to 7 of the Topics and the
concerns of the mature Analytics, and this shift of interest is very easily
seen as a shift from a concern with discursive reasoning to a concern
with "pure" patterns of inference. But Aristotle's syllogistic grows out of
107
the dialectic of the Topics and the De sophisticis elenchis, and it retains
important traces of its dialectical origins. Kapp has given a particularly
insightful account of this discursive context of syllogistic reasoning
in his now classic article on syllogistic in Pauly-Wissowa's RealEncyclopiidie. 9 Kapp's argument is that the syllogism should be seen as
a real process in which two people participate. We have already noted
that the conclusions of Aristotelian syllogisms are not sought but given
prior to the construction of the syllogism. What are sought are the
premisses which will yield those conclusions in the requisite way. The
path to be followed in such a search is clearly the reverse of syllogistic
inference. If, following Kapp, we let A seek the premisses, then upon
finding them by this reverse path A is in a position to construct a
syllogism, and to present this syllogism to B who, in grasping that
syllogism, moves inferentially from premisses to conclusion. The process described in Aristotle's definition of the syllogism - namely, that
certain things (the premisses) being stated, something other than what is
stated (the conclusion) follows of necessity from the truth of those
things alone (An. pro AI, 24b18-22) - occurs as an intellectual
process in B. But the syllogism itself is not to be identified with B's
mental activity; A and not B is responsible for the syllogism which B
grasps. That syllogism is therefore in an important sense independent of
B, who can only accept or reject it. In other words, the context of
syllogistic is a thoroughly discursive one. This is true not only of the
paradigmatic case of the dialectical syllogism - where A and Bare
opponents, and where the point of the exercise is for A, by employing
dialectical skills, to get B to accept something contentious - but
equally so of the demonstrative syllogism, where A and B are teacher
and pupil respectively, the point of the exercise now being for A to
convey information to B in the most effective and economic way.
The fact that it is the topics that provide the discursive model for
syllogistic is interesting in the light of their subsequent history. The
topics underwent a number of changes after Aristotle, with Themistius
and Cicero providing their own systems of topics, and Boethius providing what was to be the definitive system of antiquity as far as the
Middle Ages was concerned. Yet while there is on the face of it a
fundamental gulf separating Aristotle and Boethius - their lists of
topics differ considerably and are organized in different ways, as well as
offering different procedures by which to find arguments by means of
these topics 10 - there is one crucial question on which they are in
108
agreement, and which distinguishes the topical systems of antiquity

from those of the Middle Ages. The topics were above all dialectical in
antiquity. They are explicitly concerned with the art of disputation in
Aristotle, and this concern is retained throughout antiquity. Boethius'
account of the topics, for example, is firmly within the context of
arguing by question and answer, and in developing arguments for and
encouraging belief in conclusions. There is a stark contrast between this
and the medieval approach. The difference is apparent in the very
earliest extant medieval logical text - Garlandus Compotista's Dialectica,
composed probably in the early eleventh century - where the focus
is not principally upon the discovery of arguments but upon their
confirmation, with a special emphasis on enthymemes (see [46]). The
context of disputation is merely perfunctory, as indeed it is also in the
case of the standard medieval account of the topics, that provided two
centuries later by Peter of Spain in his Tractatus. Peter does not
conceive of the topics in terms of questions or of inducing one's
opponents to believe something, but rather in terms of supplying
explanations and justifications of correct but enthymematic inferences
(see [451235-6).
Peter of Spain's work lies at the heart of subsequent developments in
logic up to Descartes, and the two most influential conceptions of logic
in the sixteenth and early seventeenth centuries can be distinguished in
terms of the attitude that they take to Peter of Spain. The humanist
view, which in this period takes the form of Ramism, takes its starting
point from Agricola's rejection of Peter's conception of logic. The
scholastic view of logic on the other hand, which in this period
principally takes the form of the Jesuit theory of directio ingenii or
"directions for thinking", is a development, albeit a considerably revised
one, of Peter's account. These are not the only views which flourished
in the period but regressus theory, for example, had no influence in the
seventeenth century, and little outside Padua in the sixteenth century,
and the only other influential school - the so-called "systematics"
(Keckermann, Buscherus, Livarius, Alsted and Timpler) 11 - were
concerned to reform scholastic logic in the light of Ramist criticisms, so
need no separate attention here.
The humanist interpretation of logic has two landmarks which
deserve our attention: Rudolph Agricola's De dialectica inventione libri
tres, first published in 1515 but circulated in manuscript form from the
1480s, and the writings of Peter Ramus, and his collaborator Omar
109
Talon, from the 1540s onwards. The De dialectica inventione, although

undeniably indebted to earlier humanist writings, especially Lorenzo
Valla's Dialecticae disputationes (written in 1439, published in 1499)
(see [24]), was virtually synonymous with logic or dialectic in the
first part of the sixteenth century, and with the derivative works of
Melanchthon and Caesarius it quickly replaced Peter of Spain and Paul
of Venice as the standard textbook on dialectic, being overshadowed in
the later sixteenth century only by Ramus' work, which owes a great
deal to Agricola. Logic or dialectic must be understood broadly here.
As a component of the trivium, dialectic was theoretically an equal
partner with grammar and rhetoric, but it was usually defined in such
broad terms that it overshadowed the other two. Peter of Spain and
Lambert of Auxerre, enlarging on the Aristotelian definition (Top. A2,
101b3), define it as "the art of arts, the science of sciences, possessing
the path to the principles of all methods". Agricola's conception of
dialectic is a development of Peter of Spain's (see [35] chs. 4, 5) and it
involves dialectic taking over everything except actual delivery from
rhetoric, which in tum is reduced to ornamentation. Parallel with this
there is what can only be called a homogenization of dialectic. Aristotle
had distinguished between various forms of syllogism - dialectical,
eristic, demonstrative - and had conceived of discourse being directed
towards scientific, dialectical, rhetorical and other ends, and Aquinas
had elaborated upon the different forms of argumentation and the
different ends of discourse. But as far as scholastic thinking about
dialectic was concerned, it was Peter of Spain's broad conception, not
that of Aristotle or Aquinas, that held sway, and the humanists
capitalized on this broad undifferentiated conception. For Agricola, all
dialectic, which now effectively comprises a general theory of discourse,
has a single aim, and that aim is teaching. Cicero had distinguished
teaching, moving and pleasing as the three objectives of discourse (Opt.
gen. II), but Agricola points out that we can teach without moving or
pleasing but not vice versa (De inv. dial. Lib. I, cap. i), and concludes
that teaching is the only universal and intrinsic function of speech (Lib.
II, cap. iv). On Agricola's account, whether our imm~diate ends are
rhetorical or scientific or whatever, we are always ultimately engaged in
teaching. Indeed, one looks in vain for the logic of discovery or
"invention" mentioned in the title of Agricola's work: the whole
purpose of logic or dialectic is the ordering of material so as to convey
it to an audience. Ramus draws on this conception and gives the topics
110
the central role of sorting ideas into appropriate groups, but the topics
in turn are conceived in a completely pedagogic fashion. The structure
of knowledge is dictated in Ramus by the pedagogic classification of the
arts and sciences; as Ong puts it, "Ramus assumes that the primary
units which the mind 'contains' are the objects in the curriculum" ([35]
197), i.e., the curriculum subjects. In this respect, Ramism can be seen
as an extreme version of Aristotle's mature preoccupation with the
question of organizing and presenting already attained knowledge, an
attitude reinforced in both cases by a belief that learning is virtually
complete and remains only to be conveyed (see [16]) although there is
an added factor in Ramus: there is a Platonic element in his thinking,
more explicit in his earlier writings,12 whereby the ideas in the mind are
prior to the empirical world, and it is in this context that he advocates
clarity and distinctness as the criterion of the truth of ideas.
There is no role for demonstration, if by this we mean logical
inference, on this conception. The "principles of the arts", Ramus tells
us, "are definitions and divisions; outside of these, nothing": to "demonstrate" something is simply to define itP Even geometry, on Ramus'
view, consists not of demonstrations properly speaking but of definitions and rules. Because Ramus treats knowledge in terms of mapping
ideas accurately according to their definitions in the mind, his treatment
of reason effectively reduces it to the operation of memory, and the
problem of "method" and that of memory becomes identical. There had
been a very active medieval concern with memory, which continued to
flourish in the sixteenth century, according to which the topics were
construed in terms of places (loci, the Latin translation of the Greek
r:61lOt) in the mind where ideas were to be found by employing
mnemonic devices displaying the structure of those places (see [48]
passim). But this is both too arbitrary for Ramus - because the
mnemonic systems, which typically worked with an image of a city or a
building intimately known to the subject, so that items in that city or
building could be associated with items of knowledge, need in no way
reflect the pedagogic ordering of knowledge - and too complex for
him, and, taking his cue from Quintillian, he abolishes the loci and
images and replaces them with the division and definition of one's
subject matter. 14
In sum, there are three elements in the humanist reformulation of
logic or dialectic. The first is the extension of the scope of dialectic to
cover everything except actual delivery and grammar, thereby trans-
111
forming what in antiquity was a theory of inference into a general

theory of discourse. In Ramus, this general theory of discourse, guided
by the all-encompassing "method" that it employs, covers without
distinction geometry, natural philosophy, poetry, military strategy,
biography, and so on (see [35] 30). The second is the destruction of the
differentiations within logic, so that the distinctions between probabilistic and conclusive inference, inferences designed to convince
opponents and those designed to convince pupils, inferences directed
towards practical ends and those directed towards knowledge, all of
these distinctions are obliterated, and dialectic given a single aim:
teaching. Thirdly, the space traditionally occupied by inference now
comes to be occupied by classificatory and mnemonic devices, as
knowledge comes to be conceived in a thoroughly pedagogic fashion.
Once we conceive of proof as a means of getting others to grasp what
we already know, the move to conceiving of dialectic in purely
pedagogic terms is a natural one. The point of the exercise is then to be
able to find one's way around an already constituted body of knowledge, and for this one needs to be familiar with the structure (i.e.,
classification) of knowledge. Then, when we are faced with a new
problem - in natural philosophy, geometry, public speaking, military
strategy, metaphysics or whatever - we can establish a connection
between that problem and the storehouse of ancient wisdom which we
have access to via the procedures of division and definition, which
replace the cumbersome old mnemonic devices. Such procedures
effectively take the place of syllogistic in that they lead us to knowledge.
They are aids to knowledge and are in no way constitutive of
knowledge, just as syllogistic was conceived in the Middle Ages and
Renaissance as an aid to knowledge, albeit an unsuccessful one on the
humanist view, and therefore to be replaced by something more
efficient.
It is crucial that we grasp this conception of logic as an aid to
knowledge if we are to understand how the humanists could conceive
of replacing the Aristotelian organon with classificatory and mnemonic
schemes. The idea of the organon providing an aid to knowledge is not
one peculiar to the humanists, however, and it is encouraged by a traditional distinction between the incorporeal intellect and powers such as
imagination (phantasia), reason (cogitatio) and memory (memoria)
which were thought to have their site in the three cerebral ventricles,
damage to which, the ancient medical writers argued, could be asso-
112
ciated with specific psychological disorders. There were two issues in

dispute however in late antiquity and the Middle Ages: (1) whether
these corporeal faculties exhausted the workings of the mind, as the
Stoic and medical traditions maintained, or whether there was an
incorporeal intellect; and (2) if there were such an intellect, what its
relation to the corporeal faculties was. IS It was Aquinas who gave the
definitive answers to these questions, at least as far as his immediate
successors were concerned. While concerned to establish the existence
of an incorporeal intellect, Aquinas was equally concerned to avoid
Manichaeism, and he rejects accounts which separated the intellect
completely from the body, whether in the version of Averroes, who
ends up with only one intellect in which all bodies share, or A vicenna,
who at least allows individualized souls. Aquinas' solution is to argue
that the material on which the intellect works must derive from our
corporeal faculties: the body, via the senses, provides the phantasia
which are the basis of all knowledge. But Aquinas draws a sharp
distinction between the kinds of cognitive grasp afforded us by the
intellect or understanding (intellectus), and the reasoning (ratio) which
is the cognitive activity of our corporeal faculties. The intellectuslratio
dichotomy is a complex one in Aquinas, but the general thrust of the
distinction is to mark out a form of direct intuitive grasp of truth from a
limited, piecemeal and often unreliable form of cognitive activity, which
is the only route we have to understanding, but which is far from being
an infallible route to such understanding. Moreover, and this is an even
more important point, when it does lead to understanding, ratio
annihilates itself: it has served its purpose and disappears in favour of
true knowledge, which is conceived on an intuitive basis. 16 So the
central contrast is between direct intuition on the one hand, and the
ratiocinative processes of imagining, remembering and inferring on the
other.
On the face of it, the notion of intellectus here seems somewhat like
Aristotle's V01)~, which is also a cognitive grasp somehow qualitatively
different from the actual procedure which enables us to come by that
grasp. But there is a crucial difference. For Aristotle, the knowledge
which constitutes vov~ is not independent of the procedure that yields
it. In the case of explicitly syllogistic knowledge, for example, there may
be many syllogisms which yield a proposition, and many that yield it in
a formally valid way, but only one will yield it in such a way that the
attribute is shown to inhere in the subject universally and necessarily,
113
and unless we can construct that syllogism we will not have true
understanding. There can be little doubt that Aquinas wishes to adopt
an Aristotelian solution to the problem, but the constraints he is
operating under render this impossible. These constraints are, on the
one hand, the belief in the existence of pure spirits - God and the
angels - who know and understand, but who have no corporeal
faculties. On the other hand, the medical tradition from Galen onwards
had shown that damage to the brain and nervous system affected the
workings of reason, so it was known that our reasoning was in some
way connected with the functioning of the cerebral organs. One could
yield to one or the other of these constraints, either by maintaining that
knowledge and reason were purely functions of the cerebral organs, so
that knowledge for us and knowledge for God, who knows without
recourse to a corporeal organ, would be quite different; or one could
separate our intellect and our corporeal organs as much as possible,
holding, on neo-Platonist grounds for example, that true understanding
transcended anything we could achieve merely on the basis of the
exercise of corporeal faculties. The first of these is clearly heir to the
tradition of the via negativa, and the second to the tradition of via
affirmativa. Aquinas offers a third option, still within the tradition of
the latter, but which attempts to capture the idea that while we cannot
attain to knowledge without the use of our corporeal faculties the
successful exercise of those faculties yields something which is not
wholly different from the understanding available to pure spirit, and the
connection between the two is captured not in terms of identity but in
terms of analogy.
I want to return to these questions below. For the moment, it is
sufficient to note that they form a crucial key to the context within
which conceptions of inference are formulated in the Renaissance, and
in particular that the conception of reasoning as the exercise of a
corporeal faculty tied logic and inference closely to one's understanding
of psychological processes. This is even more marked in the representatives of the scholastic tradition in logic or dialectic in the sixteenth and
seventeenth centuries than it is in the Ramists. Unfortunately, unlike
Ramism, this late scholastic tradition has not been studied closely, but it
has been outlined by Wilhelm Risse, and I shall cull from Risse's largely
pioneering work a sketch of the principal features of this tradition and
some of the representative texts making it Up.l?
The most authoritative logic textbooks in the late scholastic tradition
114
were those of Franciscus Toletus (Introductio in dialecticam Aristotelis

(1561 and Petrus Fonseca (Institutionum dialecticarum libri octo
(1564, both of which were reprinted many times up until the midseventeenth century. They were standard texts in Jesuit schools and
the former was almost certainly amongst the textbooks from which
Descartes learned his logic at La Fleche. IS More sophisticated than the
Ramist textbooks and less concerned with reducing logic to pedagogic
devices, they offered a version of Aristotelian/Thomist logic which construed its subject matter as a practical enterprise based on Aristotelian/
Thomist psychology. Logic on this conception is an explicitly normative
theory of thought, a theory of the regulation of the functions of
cognition. Toletus and Fonseca were not the only commentators to
treat logic in these terms, but they were easily the most influential, and
through the efforts of Fonseca's followers at Coimbra, who developed a
full-scale treatment of logic as a practical theory concerned with
guiding acts of the understanding, the approach had become one with a
wide circulation by the end of the sixteenth century. A few examples
will suffice to give the flavour of this development. Suarez (Disputationes metaphysicae (1597 distinguishes metaphysics, which deals
with being as such, and logic, which directs acts of the understanding,
and is therefore concerned with the process of knowing and not with
what is known. Josephus Blanch (Commentarii in universam Aristotelis
logicam (1612 considers this process as a real psychological thought
process, and Antonius Casilius (Introductio in Aristotelis logicam
(1629 presents it as an actio vita lis, thereby effectively tying logic to
medical theory. Chrysostomus Cabero (Brevis summularum recapitulatio
(1623 poses the question of inference in terms of whether logic
exercises a natural constraint or norm which is morally binding on
thought. Finally, Raphael Aversa (Logica (1623 takes a step which is
latent in this whole development and, construing logic in a way suggestive of medical conceptions of the healthy functioning of the body,
maintains that logic is that ability which remedies the natural weaknesses of reasoning by establishing rules for coming by knowledge.
THE NATURAL LIGHT OF REASON
The framework for Descartes' conception of inference is shaped by

Ramism and late Scholasticism, and it has a number of significant
features. On the one hand, there are two conceptions which the Ramists
115
and the late Scholastics held in common. They conceive of inference as

an aid to knowledge, i.e., it is not constitutive of knowledge in any
sense. Secondly, inference is conceived as a function of corporeal
faculties, on a par with memory and imagination. On the other hand,
there are specific claims that distinguish the two schools. The Ramists
maintained that rules of inference were to be replaced by or reduced to
classificatory and mnemonic techniques. The late Scholastics argued
that inference is a psychological process to be distinguished from
understanding, which is dependent upon that psychological process but
is something over and above it. Bearing in mind this quite specific
context, Descartes' own views can be summarized in three points. First,
scientific knowledge is arrived at by "intuition" and "deduction", and
there is no need for syllogistic or rules of inference. Second, these
operations require no explication since they are simple and primitive.
Third, the pure light of reason is in any case only obscured by
attempting to supplement it in any way. Let us look at these in turn.
Descartes claims that his method explains how scientific knowledge
is arrived at by "intuition" and "deduction". This method was as much
as anything else an alternative to Ramism, although the opposition was
not made explicit by Descartes: it was left to Arnauld's Port-Royal
Logic of 1662 to do this.19 Ramus, as we have seen, construed method
in pedagogic terms and, having defined dialectic in the traditional way
as "the art of disputing well", divorced the method regulating dialectic
from empirical considerations, tying it instead to classification and
memory. Descartes wants m:!thod to serve as a logic of discovery, and
he wants it to be empirical. Ramus' method refers all questions back to
an already existing storehouse of knowledge, whereas Descartes is
reluctant to accord the contents of this storehouse the title of knowledge at all. Descartes' concern, then, is to develop a method which will
enable us to come by new and genuine knowledge. This method is not
unlike the Aristotelian topics in one respect, in that it purports to
provide us with a procedure for formulating questions relevant to the
inquiry at hand. 20 Moreover, Descartes is not as hostile to experimentation and induction as his more programmatic statements might suggest,
and these can be incorporated into the method. Descartes' conception
of method is, however, far too abstract to provide us with any secure
guidance at this level. All that it tells us is that the route to be followed
is that of "intuition" and "deduction". As regards the latter, it might
appear that Descartes is inconsistent in maintaining on the one hand
116
that deduction is part of the process of attaining scientific knowledge,

and on the other that we require no rules of inference. But Descartes
does think of deduction as being something that requires no regulation.
In Rule 2 of the Regulae we are told that mistakes in reasoning are
never due to faulty inference, the implication being that the latter is just
not possible, and in the Replies to the second set of Objections to
the Meditations it is maintained that "the proper deduction of consequences ... may be performed by anyone, even the inattentive,
provided they remember what has gone before" ([71 7, 157). Descartes
uses the Latin terms deducere and demonstrare and their French
equivalents deduire and demontrer with abandon, and they may mean
explanation, proof, induction or justification, depending on the context. 21 The shared core of meaning here is no more specific than the
comparison of one item with another, or the relating of one item to
another. That this is indeed the intended core of Descartes' conception
is made clear in Rule 14 of the Regulae:
In every train of reasoning it is merely by comparison that we attain to a precise
knowledge of the truth. Here is an example: all A is B, all B is C, therefore all A is C.
Here we compare with one another what we are searching for and what we are given,
viz., A and C, in respect of the fact that each is B, and so on. But because, as we have
pointed out on a number of occasions, the forms of the syllogism are of no aid in
perceiving the truth about things, it will be better for the reader to reject them
altogether and to conceive that all knowledge whatsoever, other than that which
consists in the simple and pure intuition of single independent objects, is a matter of the
comparison of two things or more with each other. In fact practically the whole task set
the human reason consists in preparing for this operation; for when it is open and
simple, we need no aid from art, but are bound to rely upon the light of nature alone, in
beholding the truth which comparison gives us ([7]10, 439-40).
The difference between intuition and deduction lies in the fact that
whereas the latter consists in grasping the relations between a number
of propositions, intuition (intuitus) consists in grasping one proposition
or in grasping a necessary connection between two propositions, and it
is equated with clear and distinct perceptionP In the limiting case, as
we have seen, deduction reduces to intuition: we run through the
deduction so quickly that we no longer have to rely on memory, with
the result that we "have the whole in intuition" before us at a single
time. So in the limiting case, knowledge consists not in intuition and
deduction as such, but simply in intuition.
Notice, however, that as well as consisting in a grasp of a necessary
117
connection between two limiting terms, which is what deduction

reduces to, intuition can also consist in a grasp of a single proposition.
On the face of it, it might seem that the first alone is relevant to
Descartes' conception of inference. But the second is if anything even
more revealing for, given the way in which Descartes presents the
distinction between intuition and deduction, the obvious model is a
geometrical one, in which we grasp certain axioms etc. and deduce from
these geometrical theorems. One problem with axiomatic systems whether in geometry, logic or any other formalized domain - is that
one might be misled into thinking that axioms are indispensible, serving
a special role for which rules of inference alone would be inappropriate. Since Gentzen, we know this to be false, and the various forms
of "natural deduction" and other axiomless systems have distinct
advantages over axiomatic systems (see [26], [18]). Descartes saw
matters very differently. It is not just a case of axioms being necessary;
Descartes clearly thinks that for something to be an axiom it must have
special intrinsic properties, such as self-evidence and indubitability,
which enable it to play the role it does. Propositions meeting these
requirements are grasped by intuition, not deduction, and form the
basis for any subsequent deduction. Although intuitus disappears from
Descartes' vocabulary in his later writings, this general conception does
not, and indeed its crowning achievement is the cogito. The cogito
is effectively an intuition of a basic premiss which, because of its
indubitability and self-evidence, can be grasped independently of anything else, including rules of inference. It forms the starting point for
knowledge and the paradigm for knowledge in that, while it is a grasp
of a single proposition, to get to other propositions one grasps
necessary connections between this and the others, remembering that,
in the limiting case, this grasp should itself take the form of an intuition.
In construing deduction in terms of intuition rather than rules of
inference, one thing that Descartes is doing is ruling out any attempt at
analysing inferential steps: in the limiting case, there are no such steps.
Inference cannot be analysed on Descartes' view because it is simple
and primitive. He gives us no details of what he has in mind here, but
he makes the same kind of claim about truth in a letter to Mersenne of
16 October 1639, and here he does spell out what he means. Discussing Herbert of Cherbury's De veritate, which replies to scepticism by
providing a general account of truth, on the grounds that if we understand what truth is we will be able to show that scepticism rests upon a
misunderstanding of truth, Descartes writes:
118
In the general plan of the book the author takes a route very different from the one I
have followed. He examines what truth is; I have never thought of doing so, because it
seems a notion so transcendentally clear that nobody can be ignorant of it. There are
many ways of examining a balance before using it, but there is no way to learn what
truth is, if one does not know it by nature. What reason would we have for accepting
anything which could teach us the nature of truth if we did not know that it was true,
that is to say, if we did not know truth? Of course it is possible to tell the meaning of
the word to someone who did not know the language, and tell him that the word truth,
in the strict sense, denotes the conformity of thought with its object and that when it is
attributed to things outside thought, it means only that they can be the objects of true
thoughts, whether in our minds or in God's. But no definition of logic can be given
which will help anyone to discover its nature. I think the same holds of many other
things which are very simply and naturally known, such as shape, size, movement,
place, time and so on: if you try to define these things you only obscure them and cause
confusion .... The author takes universal consent as the criterion of his truths; whereas
I have no criterion for mine except the light of nature ([7[ 2, 587; [10[ 65-6).
That is to say, while we can define truth, such a definition could not be
explanatory, for nothing can be clearer than truth: we can explain what
the word means in the sense of explaining that this is the word that we
use of a certain phenomenon, but not in the sense of giving an account
of that phenomenon in other terms which are better understood. The
reasoning behind this is that unless we had a prior understanding of
truth, we could not understand a definition of it, for we would have to
be able to grasp that the definition itself was true if we were to
understand it. Unless we had already grasped the difference between
truth and falsity, it would be wholly obscure what role definitions could
play. Of course, a great deal here depends on whether we are interested
in expressions extensionally equivalent to "... is true", or what truth
consists in, or what it is that distinguishes true from false sentences, or
what we recognize as tests for truth, or what the connection between
truth and other semantic notions is, and so on. Truth can be taken as
primitive in some respects, but not in others. But if one takes Descartes'
own example, the conformity of a thought with its object, then, whether
one construes that object as being an intentional object or whether, as
with the correspondence theory of truth, one takes it as a real object (or
state of affairs), then Descartes is surely right. To say that truth consists
in such a relation is to say that it is true that it consists in that relation.
We can gain no enlightenment about what truth consists in in this
fashion.
Nevertheless, to say that truth is primitive and simple is not to say
119
that we have a primitive and simple way of determining, for any

sentence, whether it is true. This is where the problem in the closely
related case of inference arises. The parallel between inference and
truth is not merely one of analogy. If it were, then Descartes could
simply deny that one can define inference in terms which are better
understood. But he does not do this. Quite the contrary. He effectively
provides just such a definition in maintaining that, in the limiting case,
inference comes down to the intuitive grasp of a necessary connection
between premiss and conclusion. What Descartes denies is that this
grasp can be justified, on the grounds that anything which would justify
it would have to presuppose it. It is here that we have the parallel with
truth, and in fact it turns out to be more than merely a parallel, for our
intuitive and instantaneous grasp of inferential connection is an intuitive
and instantaneous grasp of a truth. But how do we know that what we
grasp is in fact a truth? To say that the "light of nature" or "light of
reason" must be our guide is unhelpful without some specification of
how this "light" works. Does it enable us to recognize some intrinsic
quality possessed only by truths, or perhaps to partition propositions
on the basis of some other criterion? It is interesting here to note just
how wide the gulf is between Descartes' solution to the problem and
the paradigmatic discursive justification of an inferential principle:
Aristotle's justification of the law of non-contradiction. In Metaphysics
f4, Aristotle points out that proofs must come to an end somewhere,
otherwise we could be involved in an infinite regress. Hence there must
be something that we can rely upon without proof, and he takes as his
example the law of non-contradiction. The law is justified by showing
that an opponent who denies it must, in denying it, actually assume its
truth, and by showing that arguments which apparently tell against it e.g., the Protagorean relativist arguments which deduce from the fact
that a thing may seem sweet to one person and bitter to another that it
is both sweet and bitter (Le., both sweet and not sweet) - cannot be
sustained. It is here that the discursive conception of inference shows its
mettle, and Descartes can offer nothing analogous. It is something
ambiguously psychological - the "light of reason" or the "light of
nature" - that stops the regress in Descartes' conception.
There is another aspect of Descartes' argument, however. In rejecting
the idea that inference is to be guided by rules, what he is concerned
with are the rules of reasoning offered by Ramus' method and the Jesuit
"directions for thinking" (directio ingenii). His argument is that we
120
cannot be taught what an inference is: we cannot be taught to reason.

Descartes, of course, offers his own "rules for the direction of the
mind" and "discourse on the method of rightly conducting reason", but
these presuppose not only that one can reason but that one never in
fact makes mistakes of inference, as we have seen, and hence tend to be
negative, often consisting of little more than trivial hints about how to
avoid various errors due to inattentiveness, unnecessary complexity etc.
Unlike the Ramist and Jesuit theories, they are not designed to instruct
one how to think. This is evident, for example, from his remarks in the
Search after Truth by the Light of Nature, written in 1640s:
I cannot prevent myself from stopping you here, ... [toJ make you consider what
common sense can do if it is well directed. In fact, is there anything in what you have
said which is not exact, which is not legitimately argued and deduced? And yet all the
consequences are drawn without logic or a formula for the argument, thanks to the
simple light of reason and good sense which is less subject to error when it acts alone
and by itself than when it anxiously tries to follow a thousand diverse rules which
human art and idleness have discovered, less to perfect it than to corrupt it. 2]
In rejecting "rules of inference" Descartes is not concerned with logical

laws as such, but with rules which purport to teach one how to think
properly. The broad way in which dialectic had been conceived in the
Renaissance led to a conflation of these two. In the "Conversation with
Burman", Descartes says that his criticisms of logic (in the Discourse on
Method) are really criticisms of dialectic, rather than criticisms of logic
proper:
This really applies not so much to Logic, which provides demonstrative proofs on all
subjects, but to Dialectic, which teaches us how to hold forth on all subjects. In this way
it undermines good sense, rather than building on it. For in diverting our attention and
making us digress into the stock arguments and headings, which are irrelevant to the
thing under discussion, it diverts us from the actual nature of the thing itself ([9J 46).
While this explicit distinction is an afterthought on Descartes' part, the

implicit distinction is there in the earlier writings. It is an important
distinction, and if we adhere to it we can separate out with greater
precision the issues to which Descartes' criticisms are directed.
We can distinguish between the question of the justification of basic
logical principles and the justification of particular inferences, and we
can break this last question down into two further ones: what inferences
do we count as canonical, and what is the relation of other inferences to
these? In putting the question in this way, a direct comparison with
121
Aristotle is possible. In the Prior analytics, Aristotle classified syllogisms into three figures, and the following can serve as examples of the
general forms:
Barbara (Fig. 1)
A holds of all B
B holds of all r
Cesare (Fig. 2)
N holds of no E
N holds of all M
E holds of no M
A holds of all
Darapti (Fig. 3)
holds of all P
n holds of all P
holds of some
Aristotle maintains that first-figure syllogisms are perfect or complete

(rtAfLO~), whereas those of the second- and third-figures are not, and
he provides techniques for converting the latter into the former.
Second- and third-figure syllogisms can be formally valid, yet Aristotle
is not completely satisfied unless they can be converted into a canonical
first-figure form. The reason for this, as Patzig has argued in detail, is
that there is an obvious transitivity of connections in the first-figure
syllogism which is lacking in the others.24 We can, as it were, see at a
glance that the syllogism is valid. Is this very different from Descartes'
procedure? Descartes tells us that in the case of lengthy inferences we
must go through the inferential steps more and more quickly so that in
the end we grasp the premisses and conclusion in one instantaneous
step. In doing this we assimilate inference to the canonical case of
intuitus. There are differences, of course. Descartes is not concerned
with inferences which are problematic for reasons other than their
length, whereas for Aristotle the number of steps in an inference is not
a logical problem. A much more important difference, however, lies in
the criteria by which canonical forms are singled out. Aristotle's aim is
to find an argument-form which proves irresistable to an opponent, and
this parallels his account of the justification of basic principles, which as
we have seen is conceived on similarly discursive lines. Descartes'
criterion is provided in both cases by some form of psychological
clarity experienced by the knowing subject. But there is some common
ground of problems between Aristotle and Descartes, despite the fact
that these problems are posed in a discursive context in the one case
and in a psychological one in the other. This "psychological context"
requires further clarification before we can assess it fully, but two
points can be made which may go some way to dispelling the idea that
the resort to posing questions of inference in such a context is totally
damning.
122
First, there is the question of the respective merits of the cognitive

and discursive models where questions of the logic of discovery are
concerned. Questions of discovery are intimately tied to general
questions of inference in the seventeenth century, and Aristotelian
syllogistic was rejected largely because it was expected to, and failed to,
provide a logic of discovery. This expectation was mistaken - if the
seventeenth-century natural philosophers were looking for a method of
discovery in Aristotle they should have turned their attention not to
syllogistic but to the topics - but mistaken or not it spelled the end of
syllogistic. Now in the context of deductive inference, the choice is
basically that between convincing oneself, on the cognitive model, and
convincing others, on the discursive model. There is no clear advantage
for one side or the other here. But in the context of discovery, there is
an immense advantage for the cognitive conception. The discursive
conception requires common ground between oneself and one's opponents, and in seventeenth-century natural philosophy that would not
have been at all forthcoming. That is to say, the case against conceiving
of inference in a discursive way links up strongly with the case against
appealing in one's inquiries to what is generally accepted rather than to
what is the case. It is, of course, from this that the immense polemical
strength of Descartes' attack on syllogistic derives.
Second, Descartes managed to pose questions central to the nature
of inference which are literally inconceivable in Ramist thought, with its
inability to give any account of relations between propositions not
germane to pedagogical classification, and in late Scholastic thought,
where a psychological reduction robs inference of any specifically
logical features. Of central importance here is the issue of logic, and
inference generally, as an "aid to knowledge". Both the Ramists and the
late Scholastics, as I have indicated, are committed to a conception of
logic/dialectic as an aid to knowledge, i.e., as something not constitutive
of knowledge in its own right. By making inference in the limiting
case a form of intuition, which for him is knowledge par excellence,
Descartes takes the ground from under this conception. The result is
that he can raise questions of inference in a rudimentary but recognizably logico-philosophical context. The difference between Descartes
and his contemporaries is that, for Descartes, inference is not something that our corporeal organs engage in so that the information
provided thereby can be passed on to the incorporeal intellect, which
unfortunately cannot get its information in any other way. Rather, this is
123
what our intellect, when it is acting through an intuitus, tells us is

knowledge. Descartes is able to effect this radical rethinking of inference because of his doctrine of eternal truths, to which we now tum.
ETERNAL TRUTHS: A HUMAN MODEL FOR COGNITION
At first glance, the doctrine of eternal truths appears to threaten, rather

than complement, the doctrine of intuition. It commits one to the view,
for example, that there is, at least at one level, no real distinction to
be made between necessary and contingent truths for, even though
Descartes is not claiming that we could actually conceive of a world in
which necessary truths are false, the fact is that no truths are necessary
as far as God is concerned, and the effective upshot of this is that, for
God, all truths are contingent. And since, after all, it is God who
provides us with our truths in the first place on Descartes' view, this is
somewhat disconcerting. Moreover, if God is free to change all truths at
will, then even those truths which we grasp in an intuitus are called into
question. On the face of it, the doctrine of eternal truths has the
potential to bring down Descartes' whole conception of knowledge, and
a fortiori of inference. If we are to throw light on the bearing of this
doctrine on the issue of inference, there are two questions that we must
answer. First, what motivates Descartes to adopt a doctrine so counterintuitive that not one of his predecessors, contemporaries or successors
was even tempted by it? Secondly, to what extent is our grasp of truths,
whether inferential or not, affected by the fact that the cognitive
faculties that enable us to exercise that grasp do not allow us to
comprehend those truths in a way which could register any understanding of their creator's comprehension of them?
Marion has recently shown in detail that Descartes' doctrine of
eternal truths is a reaction to two currents of thought about the relation
between our knowledge and God's knowledge. 25 The principal figure in
the first current, which is that of scholastic philosophy, is Suarez, and
the evidence indicates that much of Descartes' account is specifically
directed against Suarez (see [33127-8). Suarez's account is a revision
of Aquinas' doctrine. Aquinas had developed the standard scholastic
compromise on the question of whether attribution of properties to
God and creation was univocal or equivocal. Starting from a theologically motivated assumption of equivocality, he develops an account in
which this equivocality is bridged by conceiving of the relation between
124
creation and God analogically. Underlying this analogical conception is

the doctrine of exemplarism, according to which divine ideas are
exemplars or patterns, on the models of which God created the world,
but such exemplars are imperfectly exemplified in creation. Marion
shows how the ontological basis of exemplarism subsequently comes to
be replaced by an epistemological emphasis, so that eternal truths, for
example, are no longer construed as exemplars proper, patterns on
which creation is modelled, but rather as objects to be known by both
God and us. In this way exemplarism becomes transformed into the
problem of whether our ideas can represent these eternal truths, and in
this changed context a new problem comes to the fore, which undermines the basis for the Thomist doctrine of analogy. It is Duns Scotus
who points out that, insofar as we are concerned in metaphysics with
the question of being-qua-being, analogy is not enough: we must have a
single unitary conception of being logically prior to the distinctions
between (and any analogies between) created and uncreated being, and
finite and infinite being. Suarez, on the basis of this type of argument,
takes univocity as his starting point and deploys analogy in a restricted
range. In particular, he is happy to allow that there are general
constraints on representing objects to any intellect, whether human or
divine. While a full comparison with Aquinas is not possible here,
because of the shift of context from an ontological concern with
exemplarism to an epistemological concern with representation, there is
one central overwhelming difference between Aquinas and Suarez
which, for our limited purposes, can be abstracted from context, and
this is that whereas Aquinas conceives of our knowledge of eternal
truths and God's knowledge of these truths on the basis of analogy,
Suarez conceives of them on the basis of univocity. And Descartes
conceives of them on the basis of equivocality. In fact his doctrine is
advocated as a response to the problematic and unstable nature of
Suarez's compromise. Although eternal truths are understood univocally and hence are the same for God as they are for us, Suarez tells
us explicitly that we can have no insight into how God knows them to
be true. This is what Descartes specifically objects to. Here his position
is indeed the exact contrary of Leibniz, in the sense that Descartes and
Leibniz can be seen as taking up different horns of Suarez's dilemma.
Descartes' understanding of eternal truth as equivocal turns on his
accepting that we cannot have any insight into how God knows them to
be true, so we cannot then say that such truths are the same for God as
125
they are for us. Leibniz's position can be understood as the exact
opposite of this, as an advocation of univocity on the basis that he takes
as given that eternal truths must be the same for us and for God, and
hence we must have some insight into how God knows them to be true:
and we do have such insight, in that we can say that God knows them
to be true because He knows their proofs. The second current of
thought that Descartes is reacting against is really the precursor of this
Leibnizian view. This second current is the nascent tradition of mathematical physics, and Kepler, Mersenne and Galileo all take the view
that our grasp of mathematical truths is no different from that of God.
It would take us too far from our topic to attempt to follow through
the theological, metaphysical and other considerations underlying all
these different accounts. The crucial point is to recognize that the
context in which Descartes' account is formulated is not mathematical
or logical but in the first instance theological: it is a response to a
clearly unstable theological conception of eternal truths, a conception
which pulls us in two opposing directions, complete univocity and
complete equivocality.26 However, there is an epistemological element
in this theological question, as we have seen, and the consequences, for
the question of cognition, of holding to the equivocality conception are
worth drawing out. On Descartes' account, our having a merely human
intellect does not mean that we have the same kind of knowledge as
God, only in a reduced degree - as Kepler, Mersenne and Galileo
were arguing, at least in the case of mathematics. Descartes had
touched upon questions raised here - two years before he first
elaborated the doctrine of eternal truths - in Rule 14 of the Regulae:
If in the magnet there be something the nature of which our understanding has never
grasped, we cannot hope ever to be able to know it through reasoning; for this we
would need some new sense or a divine intellect; what the human mind can achieve in
this area is achieved when we perceive with complete distinctness that combination of
already known essences or natures which produces those effects found in the magnet
([7J 10,439).
Our knowledge is not knowledge in a reduced degree but rather

knowledge of a different kind from God's, since our route to that
knowledge must of necessity differ from God's. Employing this conception, Descartes is able to give an uncompromising answer to the
traditional cognitive problem of how to reconcile the belief that our
reasoning is in some way a function of our cerebral organs, on the one
126
hand, and the belief that there are pure spirits, such as God and the
angels, who reason yet have no corporeal faculties, on the other. His
answer is to deny flatly that we can say anything about those creatures
who reason without recourse to corporeal faculties. In this he is surely
right. Another consequence of this conception is that it enables
Descartes to naturalize cognition and epistemology generally to a large
extent; not to the extent of advocating a materialist theory of mind, as
one commentator has argued ([4)), but to a very considerable extent
nonetheless (for details see [28), [34), [44), [14)). This is made possible
by dissociating our knowledge from God's, and Descartes can thereby
free himself of the constraint of trying, per impossible, to model human
knowledge on a wholly inappropriate divine prototype. This of course
leaves the problem of how creatures with our corporeally limited and
constrained cognitive faculties can have any confidence that those
corporeal faculties actually yield knowledge. But the cogito provides a
ready answer to this problem, for no matter how limited, constrained,
systematically misleading, error-prone etc., our corporeal faculties are,
they could not mislead us in this respect. The cogito has an inbuilt
guarantee: whatever my cognitive faculties, I cannot be mistaken about
my own existence, since it is just not possible for them to mislead me in
this respect. We simply do not need God's knowledge as a model, only
God's guarantee for our knowledge, and this is not such a high price to
pay when we realize that it takes us away from a model for knowledge
which is not only inappropriate but incoherent. 27
This approach is taken further in Leibniz. His univocal model of
reasoning should not be seen as something which simply contradicts
Descartes' equivocal model; rather, it builds upon it and goes beyond it
in certain crucial respects. In attempting to understand proof in terms
of intuition, so that we can move directly from premisses to conclusion
in the one step, Descartes is raising an issue which Leibniz will deal
with much more successfully in his account of algebra as a system in
which "we cannot err even if we wish ... the truth can be grasped as if
pictured on paper with the aid of a machine".28 What Leibniz is doing
here is getting rid of the need to think through all the steps in a proof
by making one's traversal of these steps not instantaneous, as was
Descartes' solution, but mechanical, something which requires no
thought yet compells intellectual assent. Moreover, Leibniz, apparently
taking it as given that we cannot say anything about cognitive processes
different from ours, proceeds to ascribe to God a reasoning process
127
modelled upon our own. We can have an understanding of God's grasp

of truth because we can provide a mechanical model for such a grasp.
But it must be remembered here that Leibniz's solution models all our
reasoning on this a priori, deductive and mechanical prototype. The
advantage of such a prototype in the restricted domain of genuinely
deductive inference is very significant, and it has proved to be superior
to the psychological model developed by Descartes and taken up in the
tradition of psychological thinking about logic from the Port-Royal
Logic to Mill's System of Logic. It is with Leibniz's conception that we
find a model for inference which is a serious rival to the discursive
model, allowing logical questions about what inference consists in to be
posed, but arguably richer than its predecessor in recognizing a cognitive dimension of inference. But when Leibniz's account is taken in its
full intended domain, the picture looks rather different. The idea that
all truths have an a priori proof, for example, verges on unintelligibility
(see [20]). Moreover, one of the most pressing problems of the seventeenth to the nineteenth centuries was that of "scientific" inference, i.e.,
the competing claims of the inductive and hypothetical models of
reasoning. 29 In this context, Leibniz's conception has nothing to offer.
His model, intended as a general model for all reasoning, was implausibly a priori and simply inapplicable in the areas considered to be of
the greatest philosophical and scientific importance. Descartes' conception, on the other hand, with its accompanying open-ended conception
of method, was able to provide some kind of general framework for
these questions, and this is undoubtedly the basis of its great success.
CONCLUSION
The period between the demise of the medieval logical tradition in the
early sixteenth century and the development of modern logic in the
work of Boole and especially Frege is generally seen, as I noted earlier,
as an interregnum in the development of logic. But the gulf that
separates the two is conceptual as well as chronological. Something has
happened in the intervening period, and if we are to grasp the main
features of the transition from ancient and medieval logic to modern
logic, then it is clearly crucial that we understand exactly what it was
that occurred in the intervening period: a period which, although it may
be an interregnum in terms of technical results, plays a crucial transitional role in terms of the development of conceptions of inference. My
128
argument has been that the central transition was from a discursive to a
cognitive conception of inference, and that Descartes played a critical
role in this transition by providing an account of how inference can
both be constitutive of knowledge and yet a cognitive process in which
our corporeal faculties engage. It is by rejecting the notion that
inference is an aid to knowledge that he is able to do this, and this
rejection depends upon his being able to treat our cognitive faculties as
being productive of knowledge in their own right, which in turn is only
possible if we do not model them on God's faculties. This last point is
secured via the doctrine of eternal truths, which thereby plays a
fundamental role in Descartes' conception of inference.
I am not suggesting, of course, that Descartes' conception is essentially the modern one, for it is not: although, like Hacking, I do believe
that Leibniz's conception has some claim to be seen in this way. What I
have tried to establish is that without the Cartesian moves that I have
outlined above, we would not have the modern conception that we do
have. It must not be forgotten, however, that this modern conception is
in many respects as problematic and unstable as any of its predecessors.
The fundamental problem of how deductive inference can be both
necessary and informative is one that we have not yet solved, for we
have not yet been able to give a fully integrated account of both the
logical and cognitive features of inference (see [11), for example).
Indeed, in the face of this and related problems, some philosophers
have advocated models of inference which are reminiscent of the older
pre-Fregean conceptions: Hintikka's game-theoretical semantics (see
[43)) bears a striking resemblance to the discursive conception, for
example, whereas Ellis' "rational belief systems" approach ([12)) is in
many respects a continuation of the psychologistic tradition. Nevertheless, even the former programme is designed to capture logical and
cognitive aspects of inference. This is constitutive of modern approaches
to conceiving of inference, and it is something that we owe to
Descartes.
NOTES
I I am grateful to Charles Larmore, Lloyd Reinhardt, and John Yolton for helpful
comments on a draft of this essay.
2 Much of this groundwork has been laid by Gerd Buchdahl. See, for example, II], 12],
13]
129
3 See, for example, 1271. An exception to the neglect of seventeenth- and eighteenthcentury logic is Howell's work (122], 123]), although this is unfortunately largely
descriptive.
4 Translated in 110]13-14. In this, as in the next quotation, words in italics designate
Latin phrases.
5 On the question of dating see 147].
6 See 17]10, 521. This is a passage that I shall look at below. On the dating of the
Search after Truth see 18] 2, 1101-104.
7 Charles Larmore has a good account of some of the issues involved here (129]). He
construes Descartes' psychologism as comprising a theory of assent, according to which
we are compelled to hold as true a proposition we recognize as certain, and then he
shows how this is connected to the doctrine of eternal truths via the argument that what
we recognize as certain may nevertheless be false. This presupposes that God's basis for
assent and ours must be different, an issue that I shall focus on below.
M For further discussion ofthe issues raised in this paragraph see my 116].
9 Now translated as 125].
10 See 145]159-261 on the changes in the topics in antiquity and the early middle
ages. On the development of the use of the topics in rhetoric in this period, see 16].
lIOn the "Systematics" see 122] and 140].
12 The first (1543) version of the Dialecticae institutiones has explicitly Platonist
elements, which are discarded from the second (1546) edition onwards. On the
development of Ramus' doctrines see 135] chs. 8-12.
13 Arist. animo (1543), fols. 58, 60. Cited in 1351188.
14 See 142] 135f, especially 140; also [48] ch. 10. Division and definition are versions
of the Platonic procedures of tJta{ewu; and Oetap6~.
15 A good historical survey of these questions is provided in 121]. For more detail on
antiquity, cf. 138].
16 137] remains the standard account of this question.
17 See 140] and Risse's 139]. My summary here is based largely on pages 284-9 of the
latter. On the texts, Risse has provided an indispensible bibliography in his 141].
1M See Etienne Gilson's discussion of the authors whom Descartes would have studied
at La Fleche in 117] 5-33.
19 On this whole question see 122] ch. 6.
20 The procedure is, however, at least partially modelled on what Descartes takes to be
the mathematical procedure of analysis. See the first part of my [14].
21 For details see 151 63-74,207-10.
22 See 15] 58-63 for full references and an invaluable discussion of intuitus.
23 17]10, 521. See also Descartes to Mersenne, 27 February 1637 (110]30), where
Descartes insists that the aim of the Discourse on Method is not to teach method but to
describe it.
24 See [36] passim. I have ignored many details, such as the fact that some noncategorical first-figure syllogisms are not "perfect", because these have no bearing on
our current concerns.
25 133]. The two currents are discussed on pages 27-159 and 161-227 respectively.
26 Marion ([33]455-6), perhaps inspired by Gilson in this respect, argues that the
original Thomist solution, which depends on the doctrine of analogy, is the path that
130
Descartes should have taken. I find this baffling since so much of his account shows
how the shift from an exemplarist to a representational context robs analogy of its
original value and motivation. To keep analogy we would have to return to exemplarism, and it is difficult to imagine what grounds anyone could have for suggesting
that this would be a move in the right direction.
27 I have argued elsewhere ([15]) that we can make no sense of the idea of knowing
truths in virtue of having made or created them.
28 Leibniz to Oldenburg, 28 December 1675 ([31J 166).
29 There are illuminating discussions of this issue throughout Gerd Buchdahl's writings.
See also [30J.
REFERENCES
1. Buchdahl, Gerd, 'Descartes's anticipation of a "logic of discovery"', in Scientific
Change, ed. A. C. Crombie (London: Heinemann, 1963).
2. Buchdahl, Gerd, 'The relevance of Descartes's philosophy for the modern philosophy of science', British Journal/or the History o/Science I (1963),227-49.
3. Buchdahl, Gerd, Metaphysics and the Philosophy of Science. The Classical Origins,
4. Caton, Hiram, The Origin of Subjectivity. An Essay on Descartes (New Haven:
Yale Univ. Press, 1973).
5. Clarke, Desmond M., Descartes' Philosophy of Science (Manchester: Manchester
Univ. Press, 1982).
6. Curtius, Ernst Robert, European Literature and the Latin Middle Ages, tr. W. R.
Trask (Princeton: Princeton Univ. Press, 1973).
(Paris: Vrin, 1964-76).
8. Descartes, Rene, Descartes: Oeuvres philosophiques, 3 vols, tr. F. Alquie (Paris:
Garnier, 1963-73).
9. Descartes, Rene, Descartes' Conversation with Burman, ed. John Cottingham
(Oxford: Clarendon Press, 1976).
10. Descartes, Rene, Descartes: Philosophical Letters, tr. Anthony Kenny (Oxford:
Blackwell, 1980).
11. Dummett, Michael, 'The justification of deduction', Truth and Other Enigmas
(London: Duckworth, 1978).
12. Ellis, Brian, Rational Belief Systems (Oxford: Blackwell, 1979).
13. Frege, G., The Foundations of Arithmetic, tf. 1. L. Austin (2nd ed., Oxford:
Blackwell, 1959).
14. Gaukroger, Stephen, 'Descartes' project for a mathematical physics', in Descartes:
Philosophy, Mathematics and Physics, ed. Stephen Gaukroger (Sussex: Harvester
Press, 1980),97-140.
15. Gaukroger, Stephen, 'Vico and the maker's knowledge principle', History of
Philosophy Quarterly 3 (1986), 29-44.
16. Gaukroger, Stephen, 'Syllogistic, circular reasoning and the logic of discovery',
forthcoming.
17. Gilson, Etienne, La liberte chez Descartes et la theologie (Paris: Alcan, 1913).
131
18. Hacking, Ian, 'What is logic?', The Journal of Philosophy 76 (1979), 285-319.
19. Hacking, Ian, 'Proof and eternal truths: Descartes and Leibniz', in Descartes:
Press, 1980), 169-80.
20. Hacking, Ian, 'A Leibnizian theory of truth', in Leibniz: Critical and Interpretive
Essays, ed. Michael Hooker (Manchester: Manchester Univ. Press, 1982), 18595.
21. Harvey, E. Ruth, The Inward Wits: Psychological Theory in the Middle Ages and
Renaissance (London: The Warburg Institute, 1975).
22. Howell, Wilbur Samuel, Logic and Rhetoric in England, 1500-1700 (Princeton:
Princeton Univ. Press, 1956).
23. Howell, Wilbur Samuel, Eighteenth-Century British Logic and Rhetoric (Princeton:
Princeton Univ. Press, 1971).
24. Jardine, Lisa, 'Lorenzo Valla: Academic scepticism in the new humanist dialectic',
in The Sceptical Tradition, ed. Myles Burnyeat (Berkeley: Univ. of California
Press, 1983),253-86.
25. Kapp, E., 'Syllogistic', in Articles on Aristotle: Volume 1. Science, eds. Jonathan
Barnes, Malcolm Schofield, and Richard Sorabji (London: Duckworth, 1975),
35-49.
26. Kneale, William, 'The province of logic', in Contemporary British Philosophy,
Third Series, ed. H. D. Lewis (London: George Allen and Unwin, 1956).
27. Kneale, William and Martha Kneale, The Development of Logic (Oxford: Clarendon Press, 1962).
28. Larmore, Charles, 'Descartes' empirical epistemology', in Descartes: Philosophy,
Mathematics and Physics, ed. Stephen Gaukroger (Sussex: Harvester Press, 1980),
6-22.
29. Larmore, Charles, 'Descartes' psychologistic theory of assent', History of Philosophy Quarterly 1 (1984), 61-74.
30. Laudan, Larry, Science and Hypothesis: Historical Essays on Scientific Methodology (Dordrecht: Reidel, 1981).
31. Leibniz, G. W., Philosophical Papers and Letters, ed. L. E. Loemker (2nd ed.,
Dordrecht: Reidel, 1976).
32. Lukasiewicz, J., Aristotle's Syllogistic from the Standpoint of Modern Formal Logic
(Oxford: Oxford Univ. Press, 1957).
33. Marion, Jean-Luc, Sur la theologie blanche de Descartes: Analogie, creation des
verites eternelles et fondement (Paris: Presses Univs. de France, 1981).
34. Maull, Nancy L., 'Cartesian optics and the geometrization of nature', in Descartes:
Press, 1980),23-40.
35. Ong, Walter J., Ramus, Method, and the Decay of Dialogue: From the Art of
Discourse to the Art of Reason (Cambridge, MA: Harvard Univ. Press, 1960).
36. Patzig, Gunther, Aristotle's Theory of the Syllogism: A Logico-Philological Study of
Book A of the Prior Analytics, tr. Jonathan Barnes (Dordrecht: Reidel, 1968).
37. Peghaire, J., Intellectus et ratio selon S. Thomas d'Aquin (Paris and Ottowa: Vrin
and Institute d'Etudes Medievales d'Ottowa, 1936).
38. Pigeaud, Jackie, La Maladie de {'lime: Etude sur la relation de {'lime et du corps
dans la tradition mMico-philosophique antique (Paris: Societe d'Edition "Les
Belles Lettres", 19 81 ).
132
39. Risse, Wilhelm, 'Zur Vorgeschichte der cartesischen Methodenlehre', Archiv fUr
Geschichte der Philosophie 45 (1963), 269-91.
40. Risse, Wilhelm, Der Logik der Neuzeit. Band 1: 1500-1640 (Stuttgart and Bad
Cannstadt: Frommann and Holzboog, 1964).
41. Risse, Wilhelm, Bibliographia logica. Verzeichnis der Druckschriften zur Logik mit
Angabe ihre Fundarte. Band 1: 1472-1800, (Hildesheim: George Olms, 1965).
42. Rossi, Paolo, Clavis Universalis: arti mnemoniche e logica combinatoria da Lullo a
Leibniz (Milan and Naples: Ricardo Ricciardi, 1960).
43. Saarinen, Esa (ed.), Game- Theoretical Semantics (Dordrecht: Reidel, 1978).
44. Schuster, John A., 'Descartes' Mathesis universalis, 1619-1628', in Descartes:
Press, 1980), 41-96.
45. Stump, Eleonore, Boethius' de topicis differentiis (Ithaca: Cornell Univ. Press,
1978).
46. Stump, Eleonore, 'Garlandus Compotista and dialectic in the eleventh and twelfth
centuries', History and Philosophy of Logic 1 (1980), 1-18.
47. Weber, J.-P, La Constitution du texte des Regulae (Paris: Vrin, 1964).
48. Yates, Frances, The Art of Memory (London: Routledge and Kegan Paul, 1966).
GEORGE MACDONALD ROSS
THE DEMARCATION BETWEEN METAPHYSICS AND

OTHER DISCIPLINES IN THE THOUGHT OF LEIBNIZ
Abstract. The terms "science" and "philosophy" have undergone radical changes in
meaning since the seventeenth century; it is therefore more profitable to consider the
relation between metaphysics and other disciplines. This essay discusses the various
ways in which Leibniz uses the term "metaphysics", and concludes that he had two
radically distinct conceptions: the one as the science of general principles and concepts,
the other as the science of immaterial reality. On both conceptions physics was
grounded in metaphysics, but in very different ways: in the former case through the
dependence of specific principles and concepts on more general ones; in the latter case
through the dependence of matter on the immaterial realm of God and monads. Only
through Leibniz's mystical belief in a structural parallel between the hierarchy of
concepts, and the universe as generated out being and nothingness, do the two
conceptions fuse together into a unitary science of metaphysics.
1. INTRODUCTION
In July 1979, at a symposium of the G. - W -Leibniz-GesellschaJt at

Reading, Gerd Buchdahl read a paper on "The Interaction between
Science, Philosophy and Theology in the Thought of Leibniz" ([5]). I
fully support his thesis that there is a close interaction between the
scientific, philosophical, and theological aspects of Leibniz's thought.
There is, however, one self-confessed lacuna in Buchdahl's treatment of
his topic, namely that he would not "stop and explain in any great detail
what we are to understand in such a context by science, by metaphysics,
and by God" ([51 75). It is largely this lacuna that I wish to fill here.
In today's terms, the most important and controversial line of
demarcation is that between science imd philosophy.l I have, however,
refrained from using these words in my title, since the meanings of both
have changed radically since Leibniz's period. In so far as they differed
at all, "science" (and its equivalents in other languages) meant any
systematic body of knowledge; whereas "philosophy" meant the totality
of knowledge, including both science and philosophy in their modern
senses. There is therefore little point even in posing the question of how
Leibniz might have dis.criminated between what we know as science
and what we know as philosophy.
On the other hand, there was a reasonably sharp distinction between
133
134
metaphysics, and special sciences such as physics, mechanics, and

mathematics. It is not the same as our distinction, since seventeenthcentury metaphysics is not twentieth-century philosophy, and the scope
and character of the special sciences have changed almost beyond
recognition. However, many of the demarcational issues involved are at
least similar, and we have the benefit of explicit statements about them.
I shall therefore proceed with a relatively extended discussion of what
Leibniz meant by "metaphysics", and then move to the question of how
he conceived its relation to the special sciences.
2. SOME SUBSIDIARY SENSES OF "METAPHYSICS" IN LEIBNIZ
Before embarking on Leibniz's conception of metaphysics itself, it will

be helpful to get out of the way a number of peripheral uses of the term
"metaphysics" or its cognates, since these may otherwise confuse the
general picture.
First, Leibniz quite often uses "metaphysical" to mean "strict", or "in
a strict sense". In Primae veritates he writes:
In metaphysical rigour [in metaphysico rigorel it can be said that no created substance
exerts a metaphysical action or influence upon another . ... And what we call causes
are only concomitant requisites in metaphysical rigour ([101 521). 2
Here, "metaphysical action" presumably also means "action in a strict

sense", as contrasted with the apparent action of objects in the world of
appearance. Again, in an undated letter to de VoIder ([11]2, 195) he
uses in metaphysico rigore merely as a variant for accurate loquendo
(cf. also [14]325). Other such expressions are metaphysice loquendo
([10]21), philosophice loquendo ([11]2, 304-5), dans une certaine
rigeur metaphysique ([12]6.6, 195), dans la rigeur metaphysique
([12]6.6,210,227), and metaphysico sensu ([13]3, 575).3
A comparable use is in the expression "metaphysical necessity",
which means strict logical or mathematical necessity, as opposed to
hypothetical, moral, or physical necessity. Thus in the Theodicy of
1710 he writes:
But truths of reason are of two sorts: one sort are what are called eternal truths, which
are absolutely necessary, so that the opposite implies a contradiction; and such are
truths of which the necessity is logical, metaphysical, or geometrical, which one could
not deny without being led into absurdities ([111 6, 50, cf. 11 0117, 20, 21; 1111 2, 170,
7,389; [141274, 287).
DEMARCATION IN THE THOUGHT OF LEIBNIZ
135
More rarely we find the corresponding notion of metaphysical possibility. As Leibniz explains in a note of 1698:
Metaphysical possibility, apart from presupposing something physical, is a pure figment,
since it can exist only in concept, and not in actuality or really ([ 14] 392).4
Another special use is in the expression "metaphysical good" and

"metaphysical evil". Leibniz defines metaphysical evil in the Theodicy:
One can understand evil in a metaphysical, a physical, and a moral sense. Metaphysical
evil consists simply in imperfection, physical evil in suffering, and moral evil in sin
([11] 6, 115).5
Normally, however, Leibniz simply uses the terms "perfect" and

"imperfect", and takes it for granted that his readers will understand
him as meaning them in the sense of having more or less "quantity of
reality", or "fullness of being"." He may have equated "metaphysical
good" with "perfection" because in its strict and etymologically primary
acceptation, "perfection" means complete realization rather than moral
or aesthetic excellence. It is this sense of the word which is presupposed
by the ontological argument for the existence of God, since it is indeed
difficult to conceive how a fully realized being could fail to exist.
However, I think it is more likely that Leibniz had in mind the idea of
metaphysics as the science of being (which I shall come to in 3.7
below), so that "metaphysical" means "in an ontological sense".
As a result of the anti-scholasticism both of the Renaissance
humanists and of experime!1tal scientists, the term "metaphysical" had
already acquired pejorative overtones in many quarters. Although
Leibniz himself never questioned the importance of metaphysics, and
frequently chided others for disparaging it, 7 he was ready to admit that
much previous metaphysics was mere words. As he said to Clarke:
Metaphysics has usually been treated as a purely verbal discipline, like a philosophical
dictionary, without arriving at any discussion of things ([11] 7,395).
This is no doubt why he often described his own metaphysics as "real",

i.e., as concerned with things, rather than with mere words. 8 Alternatively, whereas the scholastics were concerned with mere distinctions,
he was concerned with definitions. As he put it to Tentzel:
I refer to the words of the metaphysicians, none of which has a certain and fixed
definition, or the full meaning of which can be expressed sufficiently conveniently in
136
ordinary language. And above all I recoil from scholastic distinctions, which are
calculated to confuse the understanding of things; whereas one good definition removes
the empty subtlety of all their distinctions, just as bringing a light to bear dispels
darkness ([12J 1.9, 485).
All the same, Leibniz does sometimes seem to use the term "metaphysical" in a purely pejorative sense. For instance, in his disputation
On the Principle of Individuation he writes:
So it is obvious that to be numbered among the greatest chiefs of the scholastic sects is
the man who, when it came to deciding the dispute about the principle of individuation,
had recourse to materia signata, drawn from no other source than the deficiencies of
pagan metaphysics ([12J 6.1, 7).
Again, in the Hypothesis physica nova of 1671:

Doubtless the dead remains of metaphysical notions were lying heavily on Digby, when
he ascribed this restitutive force of things that have been compressed or distended, to
some unknown innate appetite (l12J 6.2,247).
More specifically, "metaphysical" could mean something like "unreal",

or "existing in name only". This is clearly what Leibniz has in mind
when he says in the De arcanis sublimium of 1676:
God is not something metaphysical, imaginary, incapable of thought, will or action, as
some make him .... But God is a particular substance, a person, a mind ([12J 6.3,474).
Again, in a reply to Bayle of 1698, he writes:

I do not remember having said this either, and it can only be said by means of a
metaphysical fiction, as if one were to suppose God annihilating a material object in
order to create a vacuum. Both suppositions are equally contrary to the order of things
([111 4, 519).Y
Elsewhere the position is less clear. In another passage from the De

arcanis sublimium, he talks with approval of a "metaphysical vacuum":
A metaphysical vacuum is an empty space as small as you like, but true and real. A
physical plenum is compatible with an infinitesimal metaphysical vacuum ([1216.3,
473).
As a solution to the problem of how matter can be both continuous and

made up of discrete points, this is dangerously close to the sort of
appeal to metaphysical entities which he was so ready to criticise in
DEMARCA nON IN THE THOUGHT OF LEIBNIZ
137
others. He soon moved to a phenomenalist position, according to which

the material world is always a plenum, however far you divide it, and
the ultimate units are not in space at all, and hence do not need to be
separated by infinitesimal vacuums.
Because of the possible implication of unreality, Leibniz was generally reluctant to apply the adjective "metaphysical" to the basic items of
his system. One rare instance from his mature period is the Systeme
nouveau of 1695, where he describes monads as "metaphysical", albeit
with some hesitation:
There are only atoms of substance ..... One could call them metaphysical points: they
have something vital and a sort of perception, and mathematical points are their points
of view for expressing the universe .... Physical points are only apparently indivisible:
mathematical points are exact, but they are only modalities: it is only metaphysical
points, or points of substance (consisting of forms or souls) that are exact and real
(111]4,482-3).
As will become clear later, he meant by this that they belonged to the
underlying realm of immaterial reality, rather than to the material realm
of appearance; but he was evidently unhappy about this mode of
expression. Later, in a letter to des Bosses of 21 July 1707, he implies
that it was an aberration. He criticises Father Perez for the obscurity of
his use of the expression "metaphysical indivisibles", but then admits:
I could use this expression of his for denoting my monads, which I remember once
calling metaphysical atoms, and also substantial atoms ([II] 2,336).
Similarly, he had already repudiated the characterization of his monads

as "metaphysical" in a letter to Varignon of 20 June 1702:
Between you and me, I think Fontenelle ... was joking when he said he would derive
metaphysical elements from our calculus. To tell the truth, I myself am far from
convinced that our infinites and infinitesimals should be considered as anything other
than ideals, or well-grounded fictions .... I believe I can prove that there do not, and
never even could, exist any infinitely small things ([13]4,110).
Leibniz's reluctance to describe things he believed in as "metaphysical" has some bearing on the vexed question of whether or not
he was serious about the vinculum substantia Ie, which he proposed in
his correspondence with des Bosses. For some commentators it is an
essential part of his system, allowing him to maintain the reality of
corporeal substances despite his apparent phenomenalism. For others it
138
is an optional extra, metaphysical in the sense of being conceivable but

not grounded in reality, and which Leibniz suggested simply in order to
accommodate the Catholic doctrine of transubstantiation. 1O In his letter
of January 1710, Leibniz describes the vinculum or union as "metaphysical", and in a context which suggests the latter sense, since it
transcends satisfactory explanation:
I do not see what would prevent spirits from being animals, but much superior to the
ones we know. Whether they would constitute a person with a body depends on the
nature of the union, which is something metaphysical, and which cannot always be
satisfactorily explained by us ([111 2, 400).
3. LEIBNIZ'S CHARACTERIZATIONS OF METAPHYSICS
Having dealt with some peripheral uses of the word "metaphysics" in

Leibniz's writings, I shall now consider how he defined metaphysics
itself. As we shall see, he characterized it in a wide variety of ways, not
all of which are obviously consistent with one another. However, most
of them are quite traditional, and are to be found collected together in
one of Leibniz's student texts, which he read in 1663-4, when he was
about seventeen. The text is Daniel Stahl's Compendium metaphysicae
of 1655; and it contains the following discussion of the definition of
metaphysics: II
Anyone undertaking a systematic exposition of metaphysics must take account of two
aspects:
(1) its name, of which
(a) its etymology is this: that it is mefa fa physica, because it was discovered after
physics, or the science of natural things. Others think that the meta can also be
explained as meaning "above" or "because of", since things dealt with in this science
are superior to natural things.
(b) Its meaning: it is a contemplative discipline, distinct from physics and the
mathematical sciences, in the way that a universal science is distinct from particular
ones .... The synonyms and titles by which Aristotle characterizes it ... are: first
philosophy ... philosophy without qualification ... the science of the philosopher
... wisdom ... theology ... the absolute chief of the sciences ... the mistress of all
... the most honoured science.
(2) The thing itself, where one must propound:
(a) its definition, which consists of
(i) its genus, which is contemplative science, and
(ii) its differentia, which is to be found in its object, which is being, in so far as
its predicates are existentially and essentially abstracted from sensible matter.
139
(b) Its division into parts, which are

(i) universal metaphysics, which deals with being in its full scope, and in so far as
it is being; and
(ii) particular metaphysics, which descends to concepts relating to species of
being, such as substance and accidents.
Leibniz wrote the following note on the text:

Gassendi thinks that Aristotle did not give the name to his books on metaphysics, but
that he called it "first philosophy" (for which there are many citations), but that those
particular books are not extant. However, once the compilers and systematizers of
Aristotle's books had systematized his Physics, they seem to have gathered together
what they could not give a definite order to or attribute to a book, because they were
related in their treatment of abstract terms. They called them meta ta physica because
they were systematized after the Physics. Metaphysics, or first philosophy, is a system of
theorems. A theorem is a proposition which would be true even if nothing existed Le., only hypothetical, or resolvable into hypothetical propositions. This is how first
philosophy is defined by Honoratus Fabri, whose Science of Universal Reasons was
published by Mosner. And by Thomas Hobbes, who divided his work De corpore into
two parts, namely first philosophy (or philosophy abstracted from existence), and
physics (or the causes of things existing in the world). Metaphysics is the product of
pure reason, and derives from definitions; the foundations of physics are laid by the
senses ([12]6.1, 21-2).
Leibniz's note deserves some comment. First, although he clearly

backed the right horse on the etymology of "metaphysics", the sequel
will show that he was in fact heavily influenced by the idea of
metaphysics as a "meta"-discipline, in the sense of one that provides a
foundation for other disciplines. Secondly, the note was written during
the period when Leibniz was especially influenced by Hobbes. In
particular, Leibniz embraces the Hobbesian view that metaphysical
truths depend ultimately on definitions. This is a position which he
never abandoned, and which was reinforced by his ideal of a "universal
characteristic", which would enable all true propositions to be demonstrated by means of definitions in terms of simpler and simpler
concepts. In his youth he was optimistic about the time it would take to
provide such a reduction. As he wrote in a note of about 1680:
I think a few select people could complete the task [of producing a philosophical and
mathematical encyclopaedia] within five years; and within two years they will set out in
an irrefutable calculus the disciplines which are more embedded in everyday life,
namely morals and metaphysics ([ 11] 7, 187).
In later life, this confident prediction faded into a longer-term ideal,
140
and he accepted that, for the time being at least, metaphysics could
have only a provisional or hypothetical status, justified by its results. 12
To return to the passage from Stahl, we can extract a number of
distinct themes about the nature of metaphysics. Metaphysics is:
(i) superior to other sciences;
(ii) contemplative (abstracted from sensible matter);
(iii) universal;
(iv) first philosophy;
(v) wisdom;
(vi) theology;
(vii) about being in general.
I shall now consider how far these themes are taken up in Leibniz's
later pronouncements.
3.1. Metaphysics as Superior to Other Sciences
Leibniz sometimes explicitly employs the metaphor of superiority when

talking about metaphysics. For example, in his Dissertation on the
Combinatory Art of 1666, he begins: "Metaphysics, to start from the
very top ..." ([111 4, 35). And much later, in the Principles of Nature
and of Grace of 1714, he writes: "So far we have been speaking only as
mere physicists: now we must ascend to metaphysics" ([111 6, 602). It
should, however, be noted that the Latin word altus means "deep" as
well as "high", and it can be ambiguous which metaphor is being used.
Leibniz often talks of other sciences as being "founded in" metaphysics,
which would put metaphysics at the "deepest" rather than the "highest"
level. But, metaphors apart, metaphysics was always the principal
science for Leibniz.
3.2 Metaphysics as Contemplative and Abstract
The characterization of metaphysics as contemplative and abstracted

from sensible matter is of more significance than might appear at first
sight. We are used to the idea of metaphysics as a purely a priori
discipline which can then, supposedly at least, be brought to bear on
the world of experience. It might therefore be tempting to interpret the
above characterization of metaphysics as saying no more than that
metaphysics is a priori and non-experiential. This would, however, be a
mistake, since the emphasis is not on the apriority of knowledge, but on
the immateriality of its object.
141
It is always important to bear in mind the extent of Platonic
influences on Leibniz's thought. Plato's epistemology was dominated by

the principle that "like is known by like": that whereas the material
senses know material things by means of sensation, the immaterial soul
knows immaterial things by means of the intellect. Leibniz fully
accepted this Platonic position, and often talked as if the defining
characteristic of metaphysics was that it was the science of immaterial
things, such as thought, soul, spirit, form, and primitive active power; as
contrasted with physics, mechanics, geometry etc., which, however a
priori, were concerned with extended matter. Thus in an undated note
of about 1683 ([10)556), he writes: "Metaphysics is the science of
intellectual things"P And in another undated paper:
Universal mathesis should handle some method of exact determination by means of
things which fall under the imagination, Le., what one might call a logic of the
imagination. Therefore metaphysics will be excluded from this, as dealing with purely
intelligible things, such as thought and action ([ 10] 348).
Then in the Discourse on Metaphysics of 1686:

It seems more and more that, although all particular phenomena of nature can be
explained mathematically or mechanically by those who understand them, nevertheless

the general principles of bodily nature and of mechanics itself are metaphysical rather
than geometrical, and belong rather to certain indivisible forms or natures as the causes
of appearances than to bodily mass or extension ([11]4, 444).
Again, in a letter to the Electress Sophie of Hanover of 3 September

1694:
I also agree with Mr. van Helmont's opinion when he criticises the Gassendists and the
Cartesians for adhering exclusively to the corpuscularian philosophy, which explains all
things in nature by matter or by extension. I myself have shown that one must also
bring in the principle of force, in which consists so to speak the connection between
spiritual and bodily things. For I hold that the laws of nature and the principles of
physics can only be explained using the metaphysical principles which one needs in
order to understand properly what force is ([12]1.10, 60).
And in the Nouveaux essais:

As for real metaphysics [i.e., not just concerned with words], we have made some sort
of a beginning at establishing it, and we find in it important truths grounded in reason,
and confirmed by experience, which apply to substances in general. I also hope that I
have advanced a little the knowledge in general of the soul and of spirits. Such a
metaphysics is what is asked for ([12]6.6,431).
142
Leibniz's stress on the immateriality of the objects of metaphysics

rather than on the apriority of our knowledge of them is brought out by
his admission of the possibility of a direct knowledge of them by
acquaintance - closely analogous to Plato's belief that the disembodied
soul could have an immediate experience of the Forms. Admittedly,
Leibniz sometimes insisted that metaphysical truths could be known
only by demonstration. For example, in a note written in about 1677,
during his Hobbesian period, he says:
All metaphysical and geometrical truths, and all others that can be demonstrated on the
basis of the terms used, are necessary per se; but all historical propositions, or so to
speak propositions of fact, which can be known to us not by demonstration but by
experience, are contingent per se, and necessary only per accidens ([ 14] 274).
On a number of occasions, however, he implies that we sometimes

have direct confirmatory ex;,erience of metaphysical truths in this life,
and that we can look forward to more such experiences after death.
Thus in an undated note:
And this is the main reason why we have advanced further in mathematics than in
metaphysics. In the former, reasons are accompanied by certain perceptual tests and
experiences confirming the results; whereas in the latter, relevant experiences are rare,
and sometimes have to be deferred to another life ([14] 31).
In another undated note:

But the trouble is that experiments in physics are difficult and expensive; and in
metaphysics they are impossible - unless God were to perform a miracle out of love
for us, to let us know immaterial things separate from us ([10]154, cf. 336).
And from a paper of about 1686:

As far as metaphysics is concerned, experience is sometimes completely impossible in
this life ([10J 176).
3.3. Metaphysics as Universal

Particularly in his earlier writings, Leibniz often referred to metaphysics
as the universal or the most general science. Thus in his De arte
combinatoria of 1666:
So, since number is something utterly universal, it rightly belongs to metaphysics - if
you accept metaphysics as the discipline concerned with those things that are common
to every genus of beings ([12J 6.1, 170).
143
And in the Preface to Nizolius of 1670 he describes metaphysics as

"the science of the totality of things" ([12]6.1, 429); and in his Analysis
lingua rum of 1678 it is "the first and most general of the sciences"
([10]353). But even in the Nouveaux essais of c.1704, he approves of
Aristotle's assertion that "the other sciences depend on metaphysics as
the most general" ([12] 6.6, 431); and in his third letter to Clarke of
1716 ([II] 7, 363), he says that metaphysical principles concern
notions which are more general than those of mathematics.
This identification of metaphysics with generality gives rise to some
uses of the term "metaphysical" which might otherwise be puzzling. For
example, in the last years of his life he wrote a paper with the title The
Metaphysical Foundations of Mathematics ([11] 7, 17); yet it contains
no hint of any metaphysics. In fact it is an attempt to bring together the
basic concepts of analysis situs, analytical geometry, and algebra into a
single system. It is "metaphysical" only in the sense of being a relatively
general and abstract science serving as the foundation for more specific
sciences. Similarly, in his letter to de VoIder of 9 January 1700 ([11]2,
205), he distinguishes two ways of arguing for his distinction between
vis viva and momentum: the one "more metaphysical", the other "more
physical". But the difference is not that he introduces metaphysical
considerations into the former (e.g., by deriving it from the vis primitiva
of monads), but that the argument is one of pure mechanics as opposed
to physics (d. [13] 3, 742; [10] 545, 569; [11] 2, 219). Elsewhere,
however, he contrasts pure mechanics with metaphysics in its normal
sense. For example, in an earlier, undated letter to de VoIder, he had
written:
If only I could clearly expound, or had properly worked out, my metaphysical
meditations on the nature of substance as well as I have the mathematical part of
dynamics ([ 11 [ 2, 162).
3.4. Metaphysics as First Philosophy
Despite Leibniz's predilection for the term "first", as in "first truths"

(e.g., [10] 518ff), he hardly ever describes metaphysics as "first philosophy". One instance is in an undated piece, where he says:
The science of size and shape has made excellent progress, but the inner nature of
motion is still obscure, due to the neglect of first philosophy, which is where it should
be sought. For it belongs to metaphysics to deal with change, time, and continuity in
general ([ 11 [ 7, 325).
144
By contrast, in his fifth reply to Clarke, of 1716, he says of the

principle of the identity of indiscernibles:
It must be admitted that this great principle, however well recognized, has not been
used enough. And this is largely the reason why first philosophy has hitherto been so
unproductive, and so undemonstrative ([II] 7,393).
Here there seems to be the implication that "first philosophy" is an

appropriate title for scholastic metaphysics rather than for his own.
Similarly, in a 1694 paper De emendatione primae philosophiae ([ 11] 4,
468ff), he is talking about what needs correcting in the metaphysics of
others.
This interpretation is borne out by his persistent failure to accept
Descartes' or Hobbes' descriptions of their own works as "first philosophy". In his note on Stahl, quoted above, he does at least acknowledge
that Hobbes himself entitled one of the parts of De corpore ([8]) "First
Philosophy"; 14 but he immediately glosses it as on metaphysics. This
would have been anathema to Hobbes, who clearly saw "first philosophy" as a much more neutral expression than "metaphysics", with its
connotation of the scholastic philosophy to which he was so vehemently
opposed. It was for the same reason that Hobbes insisted on using the
term corpus instead of substantia, even though his De corpore was in
effect a treatise on the nature of substance. In the case of Descartes,
Leibniz repeatedly describes his Meditations on First Philosophy as his
Metaphysical Meditations (e.g., [10]192; [11] 4, 469); and in a description of Descartes' Nachlass written in about 1676, he heightens the
effect by describing his Traite de la lumiere as his Physical Meditations:
There is also a Treatise on Light. That is its title. But the treatise itself is what Descartes
calls his World, or Physical Meditations, written, like the Metaphysical ones, in a
familiar style - even though in substance they contain only what is in his Philosophical
Principles ([12] 6.3, 387).
3.5. Metaphysics as Wisdom
The one theme which Leibniz does not seem to take up from Stahl's
definition is the idea of metaphysics as wisdom. There is a very early
writing ([12] 6.2, 119) in which he refers to the "politico-metaphysician"
as a wise man; but presumably his idea is that metaphysics needs to be
combined with political understanding to make a man wise. In another
early work, the De collegiis of 1665, he is more etymologically correct
145
in associating wisdom with philosophy rather than with metaphysics;

and it is clear that by "philosophy" and "wisdom" he means the totality
of knowledge:
The school of the mind of the Italian sort is an academy of science and knowledge
[cognitionis]. The latter covers both PHILOSOPHY, the goal of which is wisdom
[sapientia], and PHILOLOGY, the goal of which is erudition [eruditio]. Wisdom
consists in the three superior so-called faculties [Theology, Medicine, Law], and
philosophy in the strict sense. Philosophy in its turn has many parts, which is why some
colleges are pansophic, and others are of the particular parts of philosophy: logic,
physics, mathematics, politics, etc. Some also are mixtures of a part of philosophy and
the whole of one of the superior faculties, for example, physico-medical, politicojuridical, metaphysico-theological, etc. ([12) 6.2, 5).
Like the majority of his contemporaries, Leibniz seems to have

assumed that theoretical knowledge led directly to virtue in practice,
and that there was no need for any intermediate concept of wisdom.
Thus in the Nouveaux essais of c.1704, he writes:
It should also be understood that true morality is related to metaphysics as practice to
theory, because on the knowledge of substances in general depends the knowledge of
spirits, and in particular of God and the soul, which gives proper scope to justice and
virtue ([12]6.6, 431-2).
3.6. Metaphysics as Theology

Much more significant is the identification between metaphysics and
theology, which Leibniz accepted throughout his life. This overlaps to a
considerable extent with the idea of metaphysics as concerned with
immaterial things, since God is the immaterial being par excellence.
Thus, in the Preface to Nizolius of 1670, he talks of "metaphysics or
theology" ([12] 6.2, 429), and "theology or metaphysics" ([12] 6.2, 439);
and in the De usu geometriae of about 1676 ([12]6.3,449) we find the
phrase "metaphysics or divine contemplations". And in the Nouveaux
essais of c. 1704 he writes:
Natural theology, consisting of two parts, the theoretical and the practical, contains at
the same time real metaphysics, and the most perfect morality ([12]6.6, 432).
Other passages make it clear that his identification of metaphysics

with theology is not merely formulaic, but arises from his deep
conviction that a full understanding of nature requires not merely the
basic principles of reason, but also a knowledge of how everything
146
depends on God. For example, from an undated paper of the mid

1680s: 15
I showed that the laws of mechanics themselves flow not from geometrical but from
metaphysical principles, and that if everything were not governed by mind, it would be
very different from what we experience ([14] 29).
Again, from the Antibarbarus physicus of 1687:

I have at last shown that everything in nature does indeed happen mechanically, but that
the principles of mechanism are metaphysical, and that the laws of motion and of nature
were established not by any absolute necessity, but by the will of the wise cause ([II] 7,
343-4).
This raises the question of the demarcation between the territory of

the metaphysician and that of the theologian. As Leibniz himself
noted: 16
We will philosophize most safely by abstaining from abstractions, especially so as
to avoid impinging on theology through misusing metaphysical speculations. Most
philosophico-theological controversies are conducted unfruitfully because names are
not properly defined. There is a need for definitions such as mine, namely tangible
ones, reduced to something perceptible by means of symbols ([14] 548).
Here he seems to be saying that indulgence in "metaphysical speculation" is liable to lead to trespassing on theological territory, and that his
universal characteristic rather than his metaphysics can be of service to
the theologians. More typical is the following passage from an undated
paper, where he claims that many issues in natural theology can be
settled by means of his metaphysical system:
But many important and certain things can be said about the nature of conatus and of
the principle of that which acts, or of substantial form as the scholastics called it. From
this a great light is thrown even on natural theology, and the darkness spread over the
mysteries of faith by the objections of philosophers is dispelled ([II] 7,326-7).
As examples, he cites immortality, the existence of God, the embodiedness of ange~s, and transubstantiation.
It is abundantly clear that Leibniz regarded himself as a philosopher
and metaphysician, and not as a theologian. But this may say more
about his role in society than about any perception of limits to his
competence on his own part. He was not in holy orders, and he did not
occupy a chair of theology at a university. Apart from that, he was as
147
learned as anyone in theology; he was active in interdenominational

negotiations; and his Theodicy, which was his only full-scale publication
apart from archival material, was heavily theological.
Leibniz's most extended treatment of the question of the demarcation between metaphysics and theology is in the preliminary Discourse
on the Conformity of Faith with Reason prefaced to the Theodicy
([11] 6, 49-101). Interestingly, in this context he usually refers to
"philosophy" rather than to "metaphysics", presumably because of the
traditional identification of metaphysics with theology. Most of the
space is taken up by a historical account of the various positions on the
issue, and a point-by-point refutation of Bayle's claim that faith and
reason are in conflict. In a nutshell, Leibniz's position is that true faith
can never be in conflict with reason; that reason alone can yield a
significant proportion of theological truth (i.e., the whole of natural
theology); but that revealed theology is also necessary for certain truths
that are above, though not against, reason. He is not, however, very
forthcoming about precisely where the line is to be drawn. In section 23
he says:
A truth is above reason when our spirit (or even every created spirit) would not be able
to comprehend it. In my opinion, examples are: the Holy Trinity; miracles reserved for
God alone, such as the creation; and the choice of the order of the universe, which
depends on the universal harmony and the distinct knowledge of an infinity of things at
the same time ([ 11 J 6, 64).
Even this list is suspect, since he has already said in section 2:

Truths of reason are of two sorts: one sort are called eternal truths, which are
absolutely necessary, in that ,heir opposite implies a contradiction .... There are
others which one can call positive, since they are the laws which it has pleased God to
give to nature, or which depend on them ([11 J 6,50).
In other words, the "choice of the order of the universe" seems here to
be included within the scope of reason, rather than to be above it. In
general, Leibniz leaves very little as exclusive to faith; and in his On the
True Mystical Theology of about 1696,17 he goes as far as to say:
Faith without knowledge comes not of the spirit of God but of the dead letter of the
empty echo. Faith without light awakens no love but only fear or hope and is not living
([lSJ 369).
This impression is reinforced by Leibniz's marked (and for his time
148
quite extraordinary) reluctance to appeal to Biblical authority on

theological questions. Even in the Theodicy there are more quotations
from Virgil and Ovid than from the Bible; and in the rest of his
writings, Biblical references are virtually non-existent. 18 The contrast
with someone like Hobbes, whose philosophy was far less God-centred,
is very striking.
In short, Leibniz felt himself competent to address almost any
theological issue using the tools of reason alone, and he was singularly
unconcerned as to whether or when he was crossing the borderline into
the exclusive territory of the theologians. Even if metaphysics was not
exactly coextensive with theology, it at least included most of it.
3.7. Metaphysics as the Science of Being in General
In his early days, Leibniz concurred quite happily with the definition of
metaphysics as the science of being in general. Thus in his 1664 thesis
Specimen quaestionum philosophicarum, he writes:
Now at last it is time to move to metaphysics, where the first question will be one which
touches the very essence of being ([ 1 2] 6.1, 87).
In his De arte combinatoria of 1666:

Metaphysics, to start from the very top, deals both with Being, and with the modes of
Being ... ([12]6.1,170).
In his Demonstrationum Catholicarum conspectus of the late 1660s:

The prolegomena will cover the Elements of Philosophy. That is, the first principles of
Metaphysics (on Being), of Logic (on mind), of Mathesis (on space), of Physics (on
matter), and of Practical Philosophy (on the state) ([12] 6.1,494).
And in his Dialogus inter theologum and misosophum of about the

same time:
But logical and metaphysical principles are common to divine and human concerns,
because they deal with truth and with Being in general, which is common to God and
his creatures ([ 14] 20).
In his later writings, Leibniz became cautious about defining metaphysics as the science of being, presumably because this definition was
too closely associated with the sterile verbal distinctions of scholastic
metaphysics. As he wrote in the Nouveaux essais:
DEMARCA TION IN THE THOUGHT OF LEIBNIZ
149
It is true that the compendia of metaphysics, and other such works of the same stamp
which are generally read, teach of nothing but words. To say, for example, that
metaphysics is the science of Being in general, which explains its principles and the
modes which emanate from it; that the principles of Being are Essence and Existence;
and that its modes are either primitive (i.e., the one, the true, the good), or derivative
(i.e, the same and the different, the simple and the composed, etc.); and in talking of
each of these terms, to give only vague notions, and verbal distinctions - this is
certainly to abuse the name of science ([12J 6.6,430-1).
Having said this, he goes on to praise Suarez, and then himself, for
dealing with metaphysical issues of general importance. But he says
nothing to detract from his criticism of the vacuity of defining metaphysics as the science of being in general.
Notwithstanding Leibniz's later criticism of this definition of metaphysics, it has an important bearing on the distinction between metaphysics and logic, and hence on the question of whether the foundations of Leibniz's philosophy are metaphysical or logical. In his early
writings, at least, Leibniz made a sharp distinction between these two
most general of sciences, the former of which is concerned with Being,
that latter with predication. Thus in a passage quoted above from the
Specimen quaestionum philosophicarum, he continues:
"Can two contradictories both be false, or whether there is a middle between being and
not-being", not in respect of participation, which is the case when they are both true,
but of negation. But in case anyone might think that this is merely a logical question,
having nothing to do with metaphysics, note should be taken of what Henr. Gebhard
says in his Princ. jur. conc. 12. n. 8. pag. 146, that those most common attributes of
Being are often common to metaphysics and logic, but to the former in the mode of
being, and to the latter in the mode of predication ([12J 6.1,87).19
Similarly, in his De principio individui of 1663, he writes:

And as for the individual, while universal, it is at the same time either logical in relation
to predication, or metaphysical in relation to Being ([12J 6.1, 11).
However, with the dropping of the definition of metaphysics as the

science of being, it became difficult for Leibniz to make any sharp
distinction between the two universal sciences of reason: metaphysics
and logic. This difficulty was exacerbated by his use of the expression
"metaphysical necessity" as a synonym for "logical necessity". Thus he
revealingly runs the two sciences together in a remark contained in a
list of definitions of legal terms, dating from the mid 1690s:
Physics is about things, ethico-politics is about persons and acts; logico-metaphysics is
150
about all of these (logic is a sort of practical metaphysics, or metaphysical art)

([14J 792-3).
Russell [26], Couturat [6], and others have maintained that Leibniz's
metaphysics was derived from his logic. I do not wish here to discuss
the substance of the claim; 20 but it is relevant to consider how far the
distinction has any meaning for Leibniz. He certainly frequently
appealed to basic principles of reason, such as those of contradiction,
sufficient reason, the best, and the identity of indiscernibles. He also
claimed that they were essential for ending disputes in metaphysics for example in his Sciemia media of 1677 ([10]25), the Principles of
Nature and of Grace of 1714 ([11]6, 603), and the correspondence
with Clarke of 1715-16 ([11]7, 363). But as far as I am aware, he
never stated whether the principles themselves were logical or metaphysical. Typical is his First Truths of the early 1680s ([10]518-23),
where he starts with principles that we should describe as logical, such
as the principle of contradiction; but then brings in quite specific
metaphysical theorems, such as that there is no vacuum. In fact Leibniz
does not seem to have been interested in drawing the line between logic
and metaphysics, since they are both equally the province of reason.
4. LISTS OF METAPHYSICAL TOPICS
So far I have considered only Leibniz's more explicit pronouncements

about the nature of metaphysics. A distinct source of evidence consists
in various lists he gave of examples of metaphysical topics. The picture
gained from the material in the previous section is that Leibniz was
operating with a wide range of concepts of metaphysics, which serve
only to obscure its distinction from other sciences, such as logic and
theology. However, it is possible to reduce the range to two distinct
basic conceptions. On the one hand there is the idea of metaphysics as
a system of general principles and concepts which serve as a foundation
for all the special sciences. On the other hand there is the idea of
metaphysics as the science of immaterial things, including God. It is in
this latter context that Leibniz often uses the expression "real metaphysics". It may be that his conscious intention was to distance his own
"real" metaphysics from the merely verbal metaphysics of the scholastics; but the expression serves equally to distinguish it from metaphysics
as concerned with concepts.
151
If we now consider Leibniz's lists of examples, we shall find that they

almost all fall into two distinct classes, corresponding exactly to his
"conceptual" and "real" ideas of metaphysics. To take the conceptual
idea first, in the Preface to Nizolius of 1670, he specifies:
The one and the many, whole and part, the same and the different, the necessary and
the contingent, cause and effect, change and duration, and other metaphysical topics
([ 121 6.2,408).
Again, in the Elementa rationis of about 1686:

Yet in the primary laws of mechanics themselves, besides geometry and numbers, there
is something metaphysical concerning cause, effect, power and resistance, change and
time, similitude and determination, through which a transition is provided from
mathematical things to real substances ([101341-2).
And in the De prima philosophiae emendatione of 1694 ([11] 7, 468),

he says that ordinary language includes metaphysical terms such as
"substance, cause, action, relation, similitude", and then goes on to
outline his own work on the concept of force. In a letter to Clarke of
1716 ([11] 7, 363) he mentions cause and effect; and in an undated
note ([11] 7, 325): change, time, and continuity. In short, metaphysics is
concerned with general concepts which are required for the special
sciences.
Elsewhere, however, the topics listed are substantive items in
Leibniz's immaterialist system. For instance, in his De religione
magnorum virorum of about 1690 ([14] 38-40), he gives a catalogue
of metaphysical errors, including: materialism, God as the only
substance, occasional causes, the eternity of the world, the locality of
souls, perpetual creation, that God created all possibilities, the vacuum
of forms, the existence of a creature between God and the angels,
angelic bodies, that things are in God, and miracles. Again, in his
Response to Bayle of 1702 ([11] 4, 569), he distinguishes between a
mathematical treatment of points, indivisibles, and infinitesimals, and
metaphysical questions about their real existence. And in a letter to
Remond of 10 January 1714 ([11] 3, 606), he mentions entelechies,
forms, monads, and phenomenalism.
But the clearest evidence of his using the term "metaphysics" in this
sense is when he applies it to all or part of his own system. Thus he
wrote to Landgraf Ernst von Hessen-Rheinfels on 1 February 1686:
152
I recently composed ... a little discourse on metaphysics, on which I would like to have
Mr. Arnauld's opinion. For I think I have treated the question of grace, the cooperation of God with his creatures, the nature of miracles, the cause of sin and the
origin of evil, the immortality of the soul, ideas, etc., in a way which seems to give new
openings calculated to clarify very great difficulties ([ 11 J 2, 11).
And in a letter to Johann Bernoulli of late 1698, he says: "I now come
to the metaphysical part of your latest letter" ([13] 3, 541), and then
discusses infinite divisibility, primary and secondary matter, incomplete
being, aggregates of substances, monads, etc. Again, in the Nouveaux
essais, he mentions with approval some parts of Suarez's metaphysics:
... on the continuum, on the infinite, on contingency, on the reality of abstract things,
on the principle of individuation, on the origin and the vacuum of forms, on the soul,
and on its faculties, on the cooperation of God with his creatures, etc. ([ 12J 6.6, 431).
He then goes on to claim that in "real metaphysics" he has discovered

important truths about substances in general, the soul, and spirits.
There is, however, one rogue passage, in the Principles of Nature and
of Grace of 1714 ([11] 6, 603), where he divides his discussion into a
"physical" and a "metaphysical" part. The former covers the nature of
substance, monads, perception, organism, causality, consciousness,
reasoning, preformation, and death. The metaphysical part begins when
he starts discussing God, and his relation to the created world. This is
clearly an instance of his adopting the conventional equation of
metaphysics with theology, as contrasted with physics as the study of
created things. But as we have just seen, he normally classes all these
topics together as metaphysical.
5. METAPHYSICS AND THE ESOTERIC
More needs to be said about Leibniz's idea of metaphysics as the

science of immaterial things. I have already mentioned his reluctance to
use "metaphysical" in a pejorative sense. This reluctance is no doubt
connected with the nature of his own metaphysical system. Professedly
anti-metaphysical philosophers criticised metaphysicians both for using
meaningless terminology, and for indulging in non-empirical speculations about the inner or occult natures of things. As we have seen,
Leibniz shared the first criticism of scholastic metaphysics; but he was
hardly in a position to criticise anyone on the second ground. His
153
metaphysics was centred on a distinction between the realm of appearance and underlying reality; and his only reservation could be that his
own system was not speculative, but grounded in reason, or at least in
reasonable hypotheses.
Leibniz himself distinguished between two types of philosophy: the
esoteric or "acroamatic", and the exoteric. 21 He characterized his own
philosophy as esoteric, along with that of his mentor Plato, as contrasted with the exoteric nature of that of Aristotle, Locke, and others.22
Thus, in his Principium quoddam generale of 1687, he writes:
The more one is versed in esoteric [interiore] philosophy, the more readily one will
recognize this ([13] 6, 134).
In his letter to Hansch of 25 July 1707:

So Plato's notions, which he disguised with the term "reminiscence", are far preferable
to the tabula rasa of Aristotle, Locke, and other moderns, who philosophize exoterically ([ 17] 446).
And in the Nouveaux essais:

In fact, although the author of the Essay says a thousand good things which I applaud,
our systems are very different. His has more relation to Aristotle and mine to Plato,
even though we both distance ourselves on many points from the teachings of these two
ancient philosophers. He is more popular, and I am sometimes forced to be rather
more acroamatic and more abstract ([12]6.6, 47-8).
He explains further:
Relevant here is the ancient distinction between two ways of writing: the exoteric, that is
to say popular, and the acroamatic, which is for those who are concerned to discover
the truth. And if someone wants to write as a mathematician in metaphysics or in
ethics, there is nothing preventing them from doing so with rigour; some people have
professed to do this, and have promised us mathematical demonstrations outside
mathematics, but rarely with any success ([12] 6.6, 260-1).
More specifically, there are passages where Leibniz makes it clear

that metaphysics is concerned with what is inner and hidden. In his
dialogue Pacidius Philalethi of 1676, he commends a speech by the
character Theophilus, who ends as follows:
When [Cartesians] have meditated properly, they will see that motion itself is not
subject to the imagination and that it contains certain metaphysical mysteries pro-
154
ceeding from its spiritual nature; and also that an arcane force is inwardly present to us,
which the soul can take advantage of, kindled by love and charity, and elevated by
attentive meditation ([lOJ 627).
Again, in the Elementa rationis of about 1686:

However, the more inner aspects of mechanics [mechanices interioraJ and its primary
laws ... cannot be advanced without metaphysical principles .... Without them, it is
obvious that general physics is entirely incomplete, and that the arcane principles of
things cannot be known ([lOJ 432).
Despite the obviously esoteric character of his philosophy, Leibniz

wanted to distance himself from the wilder fantasies of occultists,
neoplatonists, and mystics; and it is revealing that on at least one
occasion he does so by using the term "hypermetaphysics". In a letter to
Johann Bernoulli of 16 May 1699, he writes:
What [Dethlev CliiverJ says about the world being structured from twelve ranks of
formative intelligences, and about the concourse and coalition of celestial rays, is
hypermetaphysics, which I do not understand ([13J 3,586).
In other words, Cluver's fault is not that he is metaphysical, but merely

that he is too metaphysical.
It is also significant that Leibniz prefers the term "acroamatic" to
"esoteric". Etymologically the former term suggests an inner circle of
hearers. It is certainly the case that, at least until he started publishing
the central tenets of his metaphysical system in the mid 1690s, he was
very reluctant to reveal them to anyone outside his inner circle of
friends and correspondents. Even as late as 1704, he suggests to Queen
Sophie Charlotte ([111 3, 348) that she is one of the few capable of
appreciating his deepest thoughts (cf. [23]).
The esoteric character of Leibniz's system serves to heighten the
contrast between his two main ideas of metaphysics - as the science of
general concepts, and as the science of hidden, immaterial reality. Yet
paradoxically it is the very esoteric and even mystical side of his
thought which provides the clue to how he in fact saw them both as
merely different aspects of a single science.
The passages I have quoted in illustration of Leibniz's idea of
metaphysics as the science of general concepts may have made it seem
a somewhat haphazard subject - as if there just happened to be certain
155
concepts which transcend the borders between the special sciences, and
which it is the task of metaphysics to analyse. However, these passages
have to be understood in the light of Leibniz's project for a "universal
characteristic". He believed that ultimately it must be possible to define
all concepts in terms of a relatively small set of highly general, simple
concepts, analogous to Aristotle's categories. Establishing such a
hierarchy is one of the tasks of metaphysics. He further believed that
the task would be made easier if he could devise a notation which
would make manifest the simple concepts out of which each complex
concept was constructed, as is already the case to a limited extent with
the compound words of natural languages. To this end, he experimented with a number of schemes involving arithmetical notations.
Quite independently of this, Leibniz discovered binary arithmetic,
which he believed to constitute a more fundamental notation for
arithmetic than the decadic system. Putting the two considerations
together, we obtain the result that it ought to be possible to designate
all concepts by binary numbers. This would mean that the ultimate
simple concepts were those denoted by the symbols 1 and O. As far as
is known, Leibniz did not experiment with using binary notation for his
universal characteristic. He did, however, speculate that 1 and 0 might
correspond to the ultimately simple concepts of being and nothingness,
and that just as all numbers, and hence all concepts, can be generated
out of 1 and 0, so the real world is generated out of being and
nothingness. Thus in his De organa sive arte magna cogitandi, probably
of the early 1670s, he writes:
Perhaps only one thing is conceived independently, namely God himself - and also
nothing, or absence of being. This can be made clear by a superb analogy.... [He then
outlines the binary system, and continues:J I shall not here go into the immense
usefulness of this system; it would be enough to note how wonderfully all numbers are
thus expressed by means of Unity and Nothing. But although there is no hope in this
life of people being able to arrive at the secret ordering of things which would make it
evident how everything arises from pure being and nothingness, yet it is enough for the
analysis of ideas to be continued as far as is necessary for the demonstration of truths
([lOJ 430-1).
In other words, there is a perfect one-to-one correspondence between

the hierarchy of concepts, and the components of the real world; and a
metaphysics of concepts will ultimately coincide with a "real" metaphysics of the underlying world of immaterial being (cf. [20]101-2).
156
But for all practical purposes, such a unification of metaphysics was

just a pipe dream. In the meantime there remains the question of how
metaphysics, in whichever sense, is related to physics, or what we
would now call "science".
6. THE RELATION BETWEEN METAPHYSICS AND PHYSICS
One thing that needs to b' made clear from the start is that Leibniz did
not avail himself of the distinction between the a priori and the
experimental in order to demarcate metaphysics from physics. He was
certainly as aware as anyone of the difference between truths of reason
and truths of fact - it lay at the very heart of his logic. But it is equally
certain that he saw the distinction as applying within physics, rather
than between physics and other sciences such as mathematics, logic,
and metaphysics. Indeed, as we have already seen, he even allowed a
role for experience in metaphysics. He is notoriously reticent as to the
precise relationship between experiential or experimental evidence and
a priori structures in physics. In practice, virtually all of Leibniz's own
work on physics was utterly a priori, whether in the form of mathematical reasoning and conceptual analysis, or in the form of speculative
theorizing (as in his Hypothesis physica nova, for example). He did
conduct some experiments in the technological sphere, but only in the
sense of making different attempts to get machines or chemical processes to work. As far as I am aware, he never conducted a controlled
experiment in order to test a scientific theory. At best he conducted
thought-experiments, for example to establish the difference between
momentum and kinetic energy; 23 but he was so confident of his
predictions that he never bothered to carry out the experiments in
practice.
Leibniz always maintained a sharp distinction between theories or
hypotheses, and phenomena: the former being the product of reason,
the latter of sense experience. But the role of the phenomenon seems to
be confined to that of providing an explicandum, to be accounted for
by a theory. Leibniz hardly concerned himself with the modern problem of the consistency between theory and phenomena. 24 He took it for
granted that any number of hypotheses might be consistent with the
phenomena, so that the most pressing problem was that of choosing
between theories on grounds other than their conformity with the facts.
Thus in his Phoranomus of after 1696, he writes:
157
The hypothesis to be chosen is the one which is more intelligible; the truth of a
hypothesis consists in nothing other than its intelligibility ([10] 591).
I have already had occasion to mention Leibniz's Platonism. Here

too he reveals a strongly Platonic attitude to the relation between
experience and physical theory: the senses serve merely to prompt the
soul into an intellectual inquiry, which is then carried out more or less
on its own terms. We might say that the value of sense experience is
heuristic: it poses the problems, and, as Gerd Buchdahl has ably shown,
it supplies us with a fund of models which can be applied by analogy to
physical, and indeed to metaphysical theories (cf. [5], and [4]388469). But physics itself is concerned with the general rational structure
which God imposed upon his creation, rather than with the particular
details of the semi-real world of experience.
Moreover, Leibniz, along with most of his contemporaries, conceived the structure of the universe in mathematical terms. The material
world was a plenum of pieces of matter of different shapes, sizes, and
masses, moving in interaction with each other. The ultimate science of
matter would therefore consist of geometry enriched by purely quantitative measures of time, and active and passive forces. Although the
resultant theory had of course to be consistent with the observed
phenomena, the main preoccupation of the physicist was with the construction of a deductive system on the model of Euclidean geometry.
Now, if physics is basically an a priori science, it will clearly be
difficult to demarcate it from metaphysics. Indeed, if metaphysics is
taken in the conceptual sense, the difference can only be one of
generality. Strictly, Leibniz should say that a conceptual issue is
metaphysical only when it involves a concept common to a number of
special sciences - substance, or cause and effect, for example.
However, for Leibniz himself the question of precisely when he was
moving from metaphysics to physics was of no particular significance,
since he regarded himself as equally competent at both. An issue arises
only when metaphysics and physics are distinct specialisms, with
separate personnel, institutions, and intellectual traditions. This is much
more the case now than it was in Leibniz's day; and it is a development
which he would have deplored, precisely because it creates an artificial
discontinuity between the special laws of physics, and the more general
concepts and principles in which they are rooted. He was already aware
of the trend towards compartmentalization, and he repeatedly criticised
158
physicists for ignoring metaphysics - and indeed metaphysicians for

ignoring physics. As he wrote to de VoIder in 1699:
I have always considered [the quantification of vis] the most appropriate gateway to
true metaphysics. It enables the spirit to be freed gradually from the false notions of the
scholastics, and even of the Cartesians, about matter, and motion, and physical
substance. The spirit learns that the laws of vis and of actions cannot be derived from
such notions, and that one must then either have recourse to a deus ex machina, or
come to understand that there is something deeper in physical substance. But if the
unprepared mind is led into these innermost sanctuaries, where the utterly unexpected
nature of substance and matter lies open to view, right from its very origins, then it is to
be feared that its mental darkness will be overwhelmed by an excess of illumination
([11]2, 195).25
So, on the conceptual interpretation of metaphysics, there is a

smooth continuum between metaphysics and physics, with no definite
line of demarcation between the two. But if metaphysics is the "real"
science of God and immaterial things, then the position is very
different.
Leibniz held that there was a perfect harmony between the two
distinct realms of matter, governed by efficient causes in the form of the
laws of mechanics; and the realm of immaterial substances, governed by
final causes in the form of appetition towards the good. The two realms
are so different that it is a category mistake 26 to apply concepts drawn
from the one to the other. As he wrote to des Bosses on 24 April 1709:
Many years ago, when my philosophy was not yet sufficiently mature, I located souls in
points. .. . But after closer deliberation, I realized that this not only landed us with
innumerable difficulties, but that there was also here a certain category mistake, so to
speak. Souls are not to be assigned spatial predicates, and their unity or multiplicity is
not to be subsumed under the category of quantity, but of substance - that is not from
points, but from primitive power of acting. But the activity proper to the soul is
perception, and the unity of the perceiver depends on the interconnection of
perceptions, whereby later perceptions are derived from preceding ones ([II] 2,372).
In the macrocosm, the harmony is between the totality of monads,

and the phenomenal world of material objects. In the microcosm, there
is a similar harmony between the soul and the body. In the Monadology, section 17, Leibniz gives a graphic account of the categorial
distinction between the two:
Imagine there was a machine constructed in such a way as to make it think, feel, and
perceive. One could suppose it enlarged with the same proportions preserved, so that
one could enter it, as into a mill. On this assumption, if one went inside, one would find
159
nothing but components moving each other, and never anything which would explain a
perception ([II] 6,609).
As he put it later in the Monadology, sections 78-81:

These principles have given me a way of providing a natural explanation for the union,
or rather the conformity of the soul and the organic body. The soul follows its own
laws, and the body likewise its own laws; and they come into contact with each other by
virtue of the pre-established harmony between all substances, since they are all
representations of the same universe.
Souls act in accordance with the laws of final causes, through appetitions, ends, and
means. Bodies act in accordance with the laws of efficient causes, i.e., the laws of
motion. And the two realms, that of efficient causes and that of final causes, are in
harmony with each other. ...
This system means that bodies act as if (per impossibile) there were no souls, and
that souls act as if there were no bodies; and that both act as if the one influenced the
other ([ 11] 6, 620-1).
This categorial distinction between souls, of which the science is

metaphysics, and matter, of which the science is physics, would seem to
imply that it is perfectly possible to function as a physicist without any
reference to metaphysics. Indeed, Leibniz often warns against importing
divine intervention or final causes into the explanation either of
particular events, or even of general features of the material world. It is
on these grounds that he rejects Malebranche's theory of occasional
causes, and Newton's suggestion that God adds new force to the
universe to prevent it from running down. Thus he writes in On Nature
ItselJof 1698:
To attribute [the power of acting which things have] to a decree of God, which, once
made, had no further effect on things and left no traces, is to be rejected out of hand.
Far from making the matter more explicable, it would mean abandoning the role of the
philosopher, and cutting the Gordian knot with a sword ([11]4, 508).
But despite these warnings against importing metaphysics into

physics, Leibniz is far from contradicting his thesis that the physicist
cannot dispense with metaphysics. The issue is simply one of when and
how metaphysics should be brought to bear. I have already mentioned
the heuristic value of analogies between the parallel realms of mind and
matter, which can provide useful clues for the physicist - for example,
the principle of least action, framed in terms of final causes, suggests a
number of fundamental axioms in optics and mechanics. But more
160
importantly, Leibniz insists that the laws of mechanics themselves

cannot be explained within the science of mechanics. If they are to be
explained at all, they must be derived from non-mechanical, immaterial
principles - i.e., from metaphysics. This is why he regards metaphysics
as "the gateway to true metaphysics", even though metaphysics is
logically prior, as being the science of what really exists, rather than of
the semi-real world of matter. As he writes to Remond on 10 January
1714:
I flatter myself that I have penetrated the harmony of the different realms, and that I
have seen that the two sides are both right, in so far as they are not in direct conflict at
all: that everything happens mechanically and metaphysically at the same time in the
phenomena of nature, but that the source of mechanics is in metaphysics. It was not
easy to uncover this mystery, because few people take the trouble to bring together
these two sorts of study ([ II] 3, 607).
It is this mystery that lies at the heart of his esoteric or "real" metaphysics. As he puts it in his Principium quoddam generate of 1687:
From this it is now obvious (rather more so than in the normal run of explanations)
how true physics is to be derived from its source in the divine perfections. For God is
the ultimate reason of things, and knowledge of God is no less the principle of the
sciences than his essence and will are the principles of things. The more one is versed in
esoteric philosophy, the more readily one will recognize this ([13]6,134).
In conclusion, Leibniz had two radically distinct conceptions of

metaphysics: the one as the science of general principles and concepts;
the other as the science of immaterial reality. On both conceptions
physics was grounded in metaphysics, but in very different ways: in the
former case through the dependence of specific principles and concepts
on more general ones; in the latter case through the dependence of
matter on the immaterial realm of God and monads. Only through
Leibniz's mystical belief in a structural parallel between the hierarchy of
concepts, and the universe as generated out of being and nothingness,
do the two conceptions fuse together into a unitary science of metaphysics.
NOTES
I
I discuss the history of the distinction between science and philosophy in [22].
In fact he deleted the first occurrence of the word "metaphysical" in the MS. This
161
supports the interpretation that the word is being used in the same sense in each
occurrence.
3 The misspelling "Methaphysico" is not in the MS.
4 Cf. [14) 396. In [14) 315, Leibniz seems to be using the expression metaphysice
loquendo in this sense. He writes: "For metaphysically speaking it would be possible for
there to be infinitely many worlds or universes in infinite time and space ...".
Presumably he means that it is a possibility which can be conceived, even though it is
not a hypothesis that can be maintained in practice.
5 Similarly, in [14) 550, he refers to "good ... in the metaphysical sense".
6 For example in an undated letter to de VoIder (11 J 2, 192), he implicitly equates
per/ectior with having plus realitatis.
7 For example Gabriel Wagner, under the pseudonym "ReaIis of Vienna" ([12) 1. 7,
676), Tentzel ([12)1.9, 361, 485), Nizolius ([12)6.2, 408, 429), and Locke ([12)6.6,
430-2).
8 For example to Locke ([12) 6.6,430-2).
9 However, there may be no pejorative implication here he may simply mean
"metaphysically possible" in the sense mentioned above.
10 Boehm ([1] passim), Rescher ([25]114-16), and Mates ([24]198) take the
vinculum substantiale seriously. Others such as Latta ([9]118-20), Russell ([26) 152),
and Broad ([2) 124-9) dismiss it as an optional extra. For a recent detailed treatment
of the whole topic, cf. [7) passim.
II [27]3. The relevant part of the text is reproduced in [12]6.1, 21-2.
12 Brown ([3) passim) argues that Leibniz abandoned the deductivist approach in
favour of a hypothetico-deductivist one. See note 24, below.
IJ Significantly, he originally wrote "theology" instead of "metaphysics". See section
3.6, below.
14 He was, however, incorrect in saying that De corpore had two parts: prima
philosophia and physica. In fact it has four parts, of which these are Parts II and IV
respectively. Part I is logica, and Part III is de rationibus motuum, et magnitudinum.
15 See also his remark in the De originatione radicali of 1697 ([11) 7, 304): "From this
it is now wonderfully comprehensible how in the very beginning of things there was
brought to playa certain divine mathesis, or metaphysical mechanism".
16 Grua gives the date "1688?"; but the strongly Hobbesian sentiments suggest an
earlier date.
17 Forthe dating of this, see [19)132, n. 31.
18 One exception is his letter to Henriette Charlotte von Pollnitz of 14 June 1700
([16)10,62-70), which contains three Biblical quotations.
19 In the printed text these are transposed. I do not know whether the slip is Leibniz's
or the editor's.
20 I discuss this in [18).
21 Wohrmann [28) argues that the esoteric approach involves demonstrating everything
rigorously, whereas the exoteric approach is looser and more dogmatic. He rests his
case largely on a passage from the Preface to Nizolius of 1670, where Leibniz does
indeed write as follows: "There is, however, a great difference between two ways of
philosophizing. One is, so to speak, acroamatic, the other exoteric. The acroamatic way
is that in which everything is demonstrated; the exoteric in which certain things are said
without demonstration, though they are confirmed by certain similes and probable
162
reasonings, or even demonstrative reasonings based on probable assumptions .... Such

a manner of speaking is certainly dogmatic or philosophical, but not acroamatic, that is,
not absolutely rigorous or precise" ([12)6.2, 416). I argue against the generalization of
this position in (23).
22 I go into further detail in (19).
2J For example the Discourse on Metaphysics, section 17 ([11]4,442-4).
24 Brown ([3) passim) interprets Leibniz as a hypothetico-deductivist. For my reservations on this interpretation, see ]21].
25 Cf. Leibniz's letter to Johann Bernoulli of I September 1699 ([ 13] 3, 610).
26 fA,ETaf3aau; elf: liMo ytVOf:. Cf. [11]6, 60; (12]6.1, 230.
REFERENCES
1. Boehm, A., Le "vinculum substantiale" chez Leibniz (2nd ed. Paris: Vrin, 1962).
2. Broad, C. D., Leibniz: An Introduction (Cambridge: Cambridge Univ. Press,
1975).
3. Brown, S., Leibniz (Brighton: Harvester, 1984).
4. Buchdahl, G., Metaphysics and the Philosophy of Science: The Classical Origins,
5. Buchdahl, G., 'The interaction between science, philosophy and theology in the
thOUght of Leibniz', Studia Leibnitiana, Sonderheft 9 (1981),74-83.
6. Couturat, L., La Logique de Leibniz d'apres documents inedits (Paris: Alcan,
1901 ).
7. Fremont, c., L 'etre et la relation (Paris: Vrin, 1981).
8. Hobbes, T., De corpore (London, 1655).
9. Latta, R., Leibniz: The Monadology and other Philosophical Writings (London,
1898).
10. Leibniz, G. W., Opuscu/es et fragments inedits, ed. C. Couturat (Paris: Presses
Univs. de France, 1903).
11. Leibniz, G. W., Die philosophischen Schriften, 7 vols, ed. C. I. Gerhardt (Berlin
and Leipzig, 1875-90).
12. Leibniz, G. W., Samtfiche Schriften und Briefe, incomplete. Ed. Preussische
Akademie der Wissenschaften (Darmstadt and Leipzig: Akademie Verlag, 1923-).
13. Leibniz, G. W., Mathematische Schriften, 7 vols, ed. C. I. Gerhardt (Berlin and
Halle, 1849-60).
14. Leibniz, G. W., Textes inedits, ed. G. Grua (Paris: Presses Univs. de France, 1948).
15. Leibniz, G. W., Philosophical Papers and Letters, tr. and ed. L. E. Loemker (2nd
ed. Dordrecht: Reidel, 1969).
16. Leibniz, G. W., Die Werke von Leibniz, 11 vols, ed. O. Klopp (Hanover, 186584).
17. Leibniz, G. W., Opera philosophica quae extant, ed. J. E. Erdmann (Berlin, 1840).
18. MacDonald Ross, G., 'Logic and ontology in Leibniz', Studia Leibnitiana,
Sonderheft 9 (1981),20-6.
19. MacDonald Ross, G., 'Leibniz and renaissance neoplatonism', Studia Leibnitiana,
Supplementa 23 (1983),125-34.
20. MacDonald Ross, G., Leibniz (Oxford: Oxford Univ. Press, 1984).
21. MacDonald Ross, G., Review of [3], The Leibniz Newsletter 2 (1985), 23-5.
DEMARCA TION IN THE THOUGHT OF LEIBNIZ
163
22. MacDonald Ross, G., 'Science and Philosophy', in Companion to the History of
Modern Science, eds. G. N. Cantor, J. R. R. Christie, M. J. S. Hodge, and R. C.
Olby (Beckenham: Croom Helm, forthcoming).
23. MacDonald Ross, G., 'Leibniz's exposition of his system to Queen Sophie
Charlotte and other ladies', Studia Leibnitiana, SonderheJt, forthcoming.
24. Mates, B., The Philosophy of Leibniz (New York and Oxford: Oxford Univ. Press,
1986).
25. Rescher, N., Leibniz: An Introduction to his Philosophy (Oxford: Blackwell,
1979).
26. Russell, B., A Critical Exposition of the Philosophy of Leibniz (Cambridge:
Cambridge Univ. Press, 1900).
27. Stahl, D., Compendium metaphysicae (Jena, 1655).
28. Wohrmann, K.-R., 'Die Unterscheidung von Exoterik und Esoterik bei Leibniz',
Studia Leibnitiana, Supplementa 21 (1980),72-82.
ROGER S. WOOLHOUSE
LEIBNIZ AND OCCASIONALISM
Abstract. Towards the end of his life Leibniz wrote that "the transition from preestablished harmony to occasional causes doesn't seem very difficult". Nevertheless he
was at pains to answer suggestions that his pre-established harmony is not essentially
different from occasionalism. Two differences he points to are that occasionalism
involves miracles, and that - like Cartesian interactionism - it involves physical
impossibilities and causal correlations between body and mind. Just as the second
difference marks him and Spinoza off from both Malebranche and Descartes, so the
first marks him off from both Spinoza and Malebranche.
1. INTRODUCTION
On more than one occasion it was suggested to Leibniz that his system
of pre-established harmony between body and mind was really not any
improvement over, or essentially different from, the occasionalism of
Cartesians such as Malebranche. It seemed to Arnauld in 1687 that
Leibniz was "saying the same thing in other words" as the occasionalists; 1 it seemed to Foucher in 1695 that Leibniz's system was
"scarcely more advantageous than that of the Cartesians"; 2 and it
seemed to Jaquelot in 1704 that, apart from its being unable to cope
with free will, it was not significantly different from occasionalism. 3
Leibniz's own way of talking perhaps encouraged such judgements.
Towards the end of his life he wrote to Raymond de Montmort that his
views were not very far distant from Malebranche's. "The transition
from occasional Causes to pre-established Harmony", he said, "doesn't
seem very difficult".4 That Leibniz does talk in this way has something
to do with his having arrived at the pre-established harmony via
Malebranche's occasionalism. He saw his views, not as denials of
Malebranche's, but as advances from, or developments of them.
But, all of this notwithstanding, the differences between his views
and Malebranche's were important to Leibniz. He was at pains to
answer suggestions that his pre-established harmony is not essentially
different from occasionalism. And, when it suited his purposes, he
could be very critical of occasionalism. This essay explores "the
165
R. S. Woo/house (ed.), Metaphysics and Philosophy of Science in the Seventeenth and
1988 by K/uwer Academic Publishers.
166
ROGER S. WOOLHOUSE
transition from occasional Causes to pre-established Harmony", and

the differences, as Leibniz saw them, between the two.
Malebranche's occasionalism was, of course, formulated at least
partly in response to Descartes' doctrine of causal interaction between
body and mind. Early in 1679 Leibniz wrote to Malebranche that he
agreed with the criticisms of Descartes' metaphysics which Malebranche
had expressed in The Search after Truth (1674) and Christian Conversations (1677). "I am", he said, "entirely of your opinion concerning the
impossibility of conceiving that a substance which has nothing but
extension, without thought, can act upon a substance which has nothing
but thought, without extension".5 A few months later he wrote to
Malebranche again, saying that "I approve most heartily ... [this
proposition] which you advance ... that strictly speaking, bodies do not
act upon US".6 Indeed, in the early 1680s Leibniz can be found not
merely going along with Malebranche's denial of causal interaction
between mind and body, but also asserting what sounds like positive
Malebrancheanism: "~]t is only through the medium of God our ideas
represent to us what passes in the world; for on no other supposition
can it be conceived how the body can act on the soul" ([19]73).
In his letters to him in 1679 Leibniz was, however, not entirely
happy with Malebranche's dealings with the question of the union
between mind and body. For one thing, Malebranche had not, in
Leibniz's view, traced things back to first principles. Malebranche had
not appreciated that the denial of causal interaction between body and
mind follows from "certain axioms" which Leibniz "do[es] not as yet see
used anywhere" (including by Malebranche) ([20]1,330: [14]210). For
another thing, in making that denial, Malebranche had "gone only
halfway" ([20]1, 328: [14]209). He had failed to see that the "important reasons" there are for denying interaction between body and mind
show also that "matter is something different from mere extension"
([20]1,328,330: [14]209, 210).
2. THE OBJECTION FROM MIRACLES
Leibniz made similar comments just a few years later, in 1686, in the
Discourse on Metaphysics. The greater detail of these comments begins
to show what he had in mind. It was ~vlnething which was to develop,
in the course of time, into one of his major objections to occasionalism,
an objection which turns on his view of substances as active and as
167
containing in their own natures the principle of their changes. Thus

we learn here that the "axioms" and "important reasons" which
Malebranche had neither appreciated nor made full use of are that
"every substance has a perfect sponeneity, [and] ... that everything
which happens to it is the result of its idea or its being" ([20] 4, 458:
[14]324). These facts about substances, Leibniz said, unexpectedly
provide "a clear insight into the great mystery of the union of body and
soul" ([20]4, 458: [14]324). A similar passage that same year, in an
explanation of the Discourse drafted for Arnauld, makes the same
connection between the union of mind and body and Leibniz's conception of individual substances: ''The hypothesis of concomitance"
between body and mind (as Leibniz then called what was later termed
the system of pre-established harmony) "is a consequence of the
concept I have of substance", namely that the "concept, notion, essence
or nature" of an individual substance "must include everything that is to
happen to it".7
Put briefly, Leibniz's "insight into the great mystery" goes like this in
the Discourse on Metaphysics. Since "everything that happens to the
soul and to each substance is a consequence of its notion" it follows
that "all its appearances or perceptions must be born from its own
nature" - from its nature, and not because the body has any "[causal]
influence" on it. The relation between what happens in the mind and
"what happens in the body which is assigned to it" is simply one of
"correspondence" ([20]4, 458: (14]325). Though Leibniz continued to
agree here with Malebranche's anti-interactionist conclusion that "there
is no way in which we can conceive of an influence of the [soul] ... on
the [body]" ([20]4, 458: (14]324) his own positive account of their
connection is different from Malebranche's, which he explicitly rejected.
In explaining the union of body and mind, he said, "it is unreasonable
simply to have recourse to the extraordinary operation of the universal
cause in so ordinary and particular a thing" ([20) 4,458: [14) 324).
Other than saying it is "unreasonable" Leibniz did not say in the
Discourse exactly what is wrong with occasionalism. But when the topic
cropped up in the ensuing discussion with Arnauld it began to emerge
that the unreasonableness depends on the idea that occasionalism
"introduces a sort of continual miracle" and is not ''worthy of God". It
has God "constantly changing the laws of bodies, on the occasion of the
thoughts of minds, or changing the regular course of the thoughts of the
soul by arousing in it other thoughts, on the occasion of the movements
168
ROGER S. WOOLHOUSE
of bodies". In short, it demands more activity of God than merely the

simple maintenance in existence of a substance "in its [own] course of
action and in the laws established for it".8
At first Arnauld could see no difference between Leibniz's "hypothesis of the concomitance and harmony between substances", and "that
of occasional causes". Leibniz's idea that "God created the soul in such
a way that ... what happens to ... [it] is born to it in its own depths,
without its having to adapt itself subsequently to the body, any more
than the body to the soul"9 seemed to Arnauld to be "saying the same
thing in other words as those who claim that my will is the occasional
cause of the movement of my arm and that God is the real cause of
it".l0 The occasionalists, he said, "do not claim that God ... [causes my
arm to move] through a new act of will which he exercises each time I
wish to raise my arm; but by that single act of the eternal will, whereby
he has wished to do everything which he has forseen that it would be
necessary to do, in order that the universe might be what he deemed it
was to be"Y
In answering this, Leibniz accepted that, for the occasionalist, God is
not continually making fresh decisions. But he reiterated his claim
about miracles. The occasionalists "introduce a miracle which is no less
one for being continual". Even if God acts "only according to a general
rule" it does not follow that his actions are not miraculous. "(T]he
concept of the miracle does not consist of rarity .... [A] miracle differs
intrinsically and through the substance of the act from a common
action, and not by an external accident of frequent repetition. . ..
[S]trictly speaking God performs a miracle whenever he does something
that exceeds the forces which he has given to creatures and maintains in
them".!2
Neither the Discourse on Metaphysics, nor the related correspondence with Arnauld, was published during Leibniz's lifetime. A few years
later, however, when, in "A new system of the nature and the communication of substances, as well as the union between the soul and the
body" (1695), he came to publish an account of the views he had been
working out in those writings, he had the objection from miracles ready
to hand and was quick to level it. Malebranche is quite correct that
"there is no real influence of one created substance upon another and
that all things, with all their reality, are continually produced by the
power of God". But his appeal to God as "a general cause ... without
offering any other explanation drawn from the order of secondary
causes is, properly speaking, to have recourse to miracle" .13
169
The publication of the "New system" led to a number of disputes,

one of the most famous of which was that with Pierre Bayle. In the
article "Rorarius" in the first edition of his Dictionnaire historique et
critique (1697) Bayle suggested that Leibniz's objections to occasionalism "involve a false supposition". In effect repeating Arnauld's point,
he argued that occasionalism just does not "require ... that the action
of God occur miraculously". As God's involvement in the world's
happenings is "only according to general laws, it is not a question of his
acting extraordinarily" ([1 ] 238).
Leibniz's reply to Bayle was published the next year in 1698, in
Histoire des ouvrages des savants, as a letter to its editor Basnage de
Beauval. Just as he had with Arnauld, Leibniz agreed with Bayle that
occasionalism holds that God acts only through general laws, but then
added that "in my opinion that does not suffice to remove the miracles".
No matter how general and regular God's actions are "they would not
cease being miracles, if we take this term, not in the popular sense of a
rare and wonderful thing, but in the philosophical sense of that which
exceeds the powers of created beings".14
Bayle replied to Leibniz five years later in a footnote added to the
second, the 1702, edition of the Dictionary ([1] 245-6). He is, he said,
"as much convinced as ever" that occasionalism does not involve
miracles: "for an action to be miraculous God would have to produce it
as an exception to the general laws; and everything of which he is the
author, in accordance with these laws, is distinct from a miracle,
properly so called". He agreed, however, to drop the matter, and turned
his attention from defence of occasionalism to criticism of pre-established harmony (as it was called by then). For Leibniz's own view, he
agreed, is "the surest means of removing all miraculous conditions . . .
[by its supposition] that created substances are actively the immediate
causes of the effects of nature". The change of tactic on Bayle's part
accounts for Leibniz's not mentioning miracles when he replied to
Bayle in that same year. 15 But he reverted to the objection, after Bayle's
death, in the Theodicy of 1710. 16
The Theodicy gives what is perhaps the clearest statement of what
lies behind this objection to occasionalism. If the ordinary regular
course of things is not to be miraculous it is, Leibniz argued, not
sufficient simply that it be the ordinary regular course of things. How it
comes about is relevant too. It must come about not because of God's
action on the things he has created. Things must do as they do not
because God makes them do so, but because they are active substances
170
ROGERS.WOOLHOUSE
in whose nature it is to do as they do, and which do so of their own

accord. "The distinguishing mark of miracles . . . is that they cannot be
accounted for by the natures of created things" ([20]6, 241: [15]257).
So unless a general law "serve[s] to explain ... [events] through the
nature of things, it can only be put into execution by a miracle" ([20] 6,
326: [15]338).
By way of dwelling longer on Leibniz's view of the world as one
whose events result, not miraculously, but from the expression of their
own natures by active substances, we may note a certain unclarity in
Bayle's first reply to Leibniz. Bayle suggested that as God's involvement
in the world's happenings is "only according to general laws, it is not a
question of his acting extraordinarily" ([1] 238). One thing Bayle meant
here is that occasionalism does not say that God is forever making up
his mind. It does not say he decides anew, on each fresh occasion, that
this hand will rise because its owner wants it to. What it does say is
rather that God makes up his mind once and for all that hands will rise
when their owners desire. But grant that occasionalism does not have
God continually making fresh decisions about each particular case.
Grant it says that events are in accord with antecedently-made general
decisions. How, nevertheless, does it stand on the question whether
God needs to act on each new occasion? Even though God has decided
once for all, and as a perfectly general matter, that arms shall rise when
their owners desire might he not, on each particular occasion of desire,
still need to do something about the particular rising of an arm?
Nothing Bayle said provides a very clear answer to this, and for one we
need to return to Arnauld. He was evidently of the view that occasionalism holds that in order for a particular arm to rise on a particular
occasion God need do (or, rather, need have done) nothing - nothing
other than already have made a general decision about occasions of
that sort. The occasionalists, Arnauld said, "claim that God ... [causes
my arm to move] by that single act of the eternal will, whereby he has
wished to do everything which he has forseen that it would be necessary to do, in order that the universe might be what he deemed it was
to be"P
Now Leibniz's reaction to an occasionalism which held that any
particular event in the world results from God's present activity
(whether in accordance with antecedently-made general decisions or
not), and hence as a miracle, would have been that though things could
happen like this it is inconsistent with God's wisdom and dignity that
171
they do. But his reaction to an occasionalism of the kind described by

Arnauld is that it is unintelligible.
This reaction is not manifest in his discussion with Arnauld but is so
in his 1698 and 1702 replies to Bayle. There it is clear that it is not
enough simply that God antecedently make general decisions: for the
mere making of a decision does not of itself ensure its being carried
out. There must be some means by which it is carried out. Either God
must be "the executor of his own laws" ([20]4, 563: [14]580) and must
do it, when there is occasion; or angels "charged expressly with this
responsibility" ([20]4, 521: [14]494) must do it; or substances must
have "instruments for such execution" ([20]4, 563: [14]580) and there
must be set up "a natural means of carrying it out, that is, all that
happens must also be explained through the nature which God gives to
things" ([20]4,520: [14]494).
The same reaction is manifest in comments Leibniz made on certain
ideas of Christian Sturm who had written to him in 1695 following the
publication of the "New system". As Leibniz describes it, Sturm held a
view of the kind alluded to in outline by both Arnauld and Bayle, that
events "now taking place result by virtue of an eternal law once
established by God, which law he then calls a volition and command,
and that no new command or new volition of God is then necessary".
Apparently Sturm further held that nothing more than this initial
volition is necessary for it to be later followed by events which satisfy it
- no "new conatus or some laborious effect" is needed. Events happen
now simply because of this prior decision, according to Sturm; he
rejected the idea that "God moves things as a wood chopper moves his
ax". Leibniz, however, found this unintelligible. A "command in the past
no longer exists at present" and if it "bestowed upon things only an
extrinsic denomination", and did not leave "some subsistent effect
behind which has lasted and operated until now" then it "can accomplish nothing". What is necessary is that "this divine law ... conferred
upon [things] some created impression which endures within them, ...
an internal law from which their actions and passions follow". It is, said
Leibniz, "not enough ... to say that in creating things in the beginning,
God willed that they should observe a certain law in their progression".
It is necessary also that "the law set up by God does in fact leave some
vestige of him expressed in things .... ~t must be that] things have
been so formed by the command that they are made capable of
fulfilling the will of him who commanded them".18
172
ROGER S. WOOLHOUSE
3. THE OBJECTION FROM PHYSICAL IMPOSSIBILITY
The "New system" made public the ideas on pre-established harmony

Leibniz had been developing earlier in the Discourse on Metaphysics
and correspondence with Arnauld. Because of the controversy it
aroused he was led to publish various "Explanations". An importance
of the first of these is that it contains the first clear statement of an
objection which, though its initial target is Cartesian interactionism, is
explicitly aimed at occasional ism too. As with much else in the "New
system" and its various "Explanations" the material for this objection is
present in the correspondence with Arnauld ([20J 2, 94: [17J 117-18).
Before we turn to it some explanation of its background is necessary.
As is well known, one of the laws of Cartesian physics is the law of
conservation of "motion". As Descartes meant it, the quantity of
"motion" possessed by a body is the product of its "size" and its
"speed". "We must", he said, "reckon the quantity of motion in two
pieces of matter as equal if one moves twice as fast as the other, and
this in its turn is twice as big as the first". According to the Cartesian
conservation law, when two bodies collide any decrease of "motion" in
one must be matched by an increase in the other for, in general, the
total quantity of motion in the universe is a constant ([8J pt. 2, prop. 36;
see also [9J 65-9).
The "law" is false. Some twenty years after Descartes' death and
towards the end of the 1660s it began to emerge (though not necessarily into general acceptance) from work done by Huygens, Wallis, and
Wren, that closed physical systems do not conserve Cartesian motion.
Two further things were emerging. One was that, unlike Cartesian
motion, momentum (which takes into account the direction of a body's
movement and so is measured by the product of "size", or "mass",
together with the speed of the body considered as a vector) is
conserved in a closed physical system. The other was that, at least in
elastic collisions, the product of the size, or mass, with the square of the
speed is also conserved. Leibniz knew of the falsity of the Cartesian law
at any rate by 1680. 19 He had knowledge of the other two, true,
conservation laws by about the same date. 20 He sometimes claimed that
he discovered the conservation of momentum himself,21 but at other
times he spoke as though he learnt of it from Huygens. 22
In the 1680s Leibniz began to criticise Descartes for a certain
"error",23 This was the error of supposing, first, that there is conserva-
173
tion of "force", and of then identifying ''force'' with "motion".24 Despite

the fact that Leibniz knew that Cartesian motion is not conserved it was
not on that ground that He rejected the identity he attributed to
Descartes. It was because Cartesian motion seemed not to be identifiable with force for other reasons. What Leibniz himself identified with
force was vis viva, the product of mass or size and the square of speed.
From the conservation of vis viva Leibniz deduced that momentum is
conserved (see [12]109).
By 1685, then, Leibniz knew, as a matter of physics, that Cartesian
motion is not conserved and that momentum is. And it is about this
time that he began to use these facts in his metaphysics as the basis for
an objection to both Cartesian interactionism and occasionalism. The
objection was somewhat undeveloped and not completely clear in the
correspondence with Arnauld ([20] 2, 94: [17]117); and, like so much
else in that correspondence, it appeared more clearly, and in public, ten
years later at the time of the "New system" and its various explanations. 25 Thus, in 1696, in the first "Explanation of the new system",
Leibniz gave his readers "a further reflection, which seems to me
helpful in making the reality and use of my system [of pre-established
harmony] better understood" ([20]4, 497: [16]327). The reflection
proceeded initially by way of criticism of Descartes.
It was part of Descartes' view of the union of body and mind that, as
in voluntary action, a desire in the soul could cause a movement of an
arm. The soul, he says, has "power to move the body", "the mind ...
can set the body in motion" ([7]138, 235). At least at first sight there is
an inconsistency between this and the claimed conservation of motion.
When a collision between your arm and mine causes mine to move
the change in Cartesian motion involved in my arm's moving must,
according to that claim, be counterbalanced by a corresponding change
of motion in yours. So how could my desire to move my arm cause it to
move? That would involve a change of motion in my arm which was
not counterbalanced by a corresponding change, for any calculation
about the Cartesian motion of my immaterial mind makes no sense.
According to Leibniz's first "Explanation of the new system"
Descartes was aware of this prima facie inconsistency and sought to
overcome it. Descartes, he said, "was perplexed by the changes which
take place in the body in consequence of the modifications of the soul,
because they seemed to break this law [of conservation of motion] ...
[and] he thought he had found a way out of it (which is certainly
174
ROGER S. WOOLHOUSE
ingenious)" ([20]4, 497: [16]327). According to Leibniz, Descartes'

''way out of it" was to distinguish between motion and direction of
motion, and to suggest that' the action of the mind on body does not
change the quantity of motion of the body or its parts, but simply its
direction.
Other than Leibniz's statement there is no evidence that Descartes
ever was "perplexed" in the way Leibniz suggests and, consequently,
none to suppose that he took this "way out".26 Apart from the
possibility of such evidence having been lost one explanation of this is
that Descartes simply failed to see there was a prima facie case to be
answered about the consistency of his interactionism and conservation
law. Another, which is suggested and argued for by Daniel Garber, is
that while believing that ([12]115) all causal interactions within the
realm of extended substance must conserve motion Descartes also
believed that the realm of extension is not causally closed. According to
this suggestion Descartes' position may have been that, while all causal
interaction between bodies must obey this law of conservation of
motion, causal interaction between mind and body need not do so.
But even if Leibniz is wrong about Descartes himself his reference to
the distinction between motion and its direction as a "way out" of this
problem of Cartesian interactionism is not purely fanciful. 27 Samuel
Clarke, in his Demonstration of the Being and Attributes of God
(1705), refers to "some ... who denying men the power of Beginning
motion, would yet seem in some manner to account for their actions, by
allowing them a power of determining motion" ([5]178). At about the
same time and somewhat along the same lines Benjamin Bayley is
strikingly cautious, in talking about free will, whether the mind gives
"motion" itself, or merely "a different determination of motion" to the
body or its parts ([2]11-12).
But, whether Descartes took it or not, this "ingenious" way out of the
problem will, Leibniz pointed out, get us nowhere. In avoiding conflict
with the supposrd "law" of conservation of Cartesian motion it runs up
against the true law of conservation of momentum. If your arm,
colliding with mine, causes mine to move in a different direction the
change in momentum involved in that change must be counterbalanced
by a corresponding change of momentum in yours. So how could an
interactionist be right that a desire could cause an arm to move? Even
if this involved only a change in the direction of motion of the arm or
parts of it this would involve a change of momentum which was not
LEIBNIZ AND OCCASION ALISM
175
counterbalanced by a corresponding change elsewhere; for, of course,

calculation about the momentum of an immaterial mind makes no
sense.
Leibniz repeated this argument over the years, both in private and in
public. 28 He made it plain from the outset that his own doctrine of
pre-established harmony does not face the same objection. But it was
only later that he made a rather stronger claim, the claim that if
Descartes had known of the law of conservation of momentum he
would have been led direct to Leibniz's doctrine of pre-established
harmony. If the law of conservation of momentum "had been known in
Descartes' day ... he would undoubtedly have been led to my system
of pre-established harmony, for he would have recognized that it is just
as reasonable to say that the soul does not change the quantity of the
direction of the body as it is to deny to the soul the power of changing
the quantity of its force [i.e., Cartesian motion], both being equally
contrary to the order of things and the laws of nature, since both are
equally inexplicable". 29
Bertrand Russell ([23]81) found this claim far too bold. Why, he
asked, could Descartes not as easily have been led to occasionalism or
to Spinoza's view of the union between mind and body as to Leibniz's
pre-established harmony? In fact when the claim is first made, in
"Considerations on vital principles" there is at least the hint of an
answer to this; namely that the argument against Cartesian interactionism applies equally to occasionalism. However, Leibniz did not here
offer any explanation why the argument applies equally to occasionalism, and for one we must tum to the Theodicy of five years later.
There Leibniz repeated that if Descartes had been aware that "direction
is ... conserved", if, that is, he had been aware that momentum is
conserved in changes of bodily movement, he would have "recognized
that without a complete derangement of the laws of Nature the soul
could not act physically upon the body". And if he had recognized that
he would have been "led direct to the Hypothesis of Pre-established
Harmony". But he also gave an explanation of this step, an explanation
why the same objection applies to occasionalism. "[O]ne could not", he
added, as if in direct answer to Russell, "here listen to philosophers ...
who produce a God ... to bring about the final solution of the piece,
maintaining that God exerts himself deliberately to move bodies as the
soul pleases, and to give perceptions to the soul as the body requires".
And why not? Because "this system ... of occasional causes ... does
176
ROGER S. WOOLHOUSE
not obviate the derangement of the natural laws obtaining in each of

these same substances, which ... their mutual influence would cause"
([20) 6, 136: [15) 156-7).
Reading this hastily one might suppose that it is, in effect, the
Objection from Miracles which was discussed in section 2. One might
suppose that it is because occasionalism denies "real" causation in the
created world, because it denies that things do as they do as active
expressions of their nature, that it is charged with involving "the
derangement of ... natural laws". But the objection here is not that
natural laws are deranged if the causation in the movement of colliding
bodies is divine and not "real" or natural; it is not that it is inconsistent
with natural law for God to move a body on the occasion of some
change in another body. The objection is that it is inconsistent with
natural law for God to move a body on the occasion of some change
in a mind. It is that occasionalism is no improvement on Cartesian
interactionism for it shares with it the crucial supposition that the
mental and the physical are not causally isolated from each other, each
with their own sacrosanct laws. It is that, like Descartes, Malebranche
envisages that, be it "real" or "occasional", the cause of a change in
body will sometimes be in a mind, and not always in body.
What lies behind Leibniz's objection to interactionism, then, is not
simply that it is inconsistent with known physical laws, but that it
involves the idea that the causes of bodily changes are not always to be
found in body. And it is because it too involves this idea that he
objected to occasionalism. Though it was not then set so firmly in the
physical context of the conservation of momentum this general metaphysical point came out clearly in the correspondence with Arnauld.
Arnauld was never quick to understand the theory of pre-established
harmony, despite Leibniz's careful attempts to get it across. It seemed
to him to be "saying the same thing in other words" ([20) 2, 84:
[17) 105) as occasionalism. For one thing, he could not, as we have
already seen, understand why the one should be thought to avoid
miracles and the other not. As the correspondence developed Leibniz
offered more and more by way of explanation. In the end came the
point that according to him everything that a body does is "by virtue of
its own laws" whereas "according to the authors of occasional causes
God changes the laws regarding bodies on the occasion of the soul". It
is there, Leibniz concluded, "[t)here lies the essential difference in our
views" ([20) 2,93: [17) 117; see also [20) 2,57-8: [17) 65).
177
It is, then, a feature of pre-established harmony that the causes of

changes in body are always other changes in body and never changes in
mind. But, essentially different from this, it is a feature of occasionalism

that sometimes the cause of a change in body will lie in mind. And this
is why, for Leibniz, occasionalism constituted no real improvement
over Cartesian interactionism. They both envisage that sometimes the
laws of physics may be interrupted by mental events. As Leibniz went
on to explain to Arnauld, "[o]ne must not be worried ... as to how the
soul can impart some movement or new determination to animal spirits,
since in fact it never does; ... the same problem is to be found with the
hypothesis of occasional causes as with the [Cartesian] hypothesis of a
real influence" ([20]2, 93-4: [17]117). Though interactionists and
occasionalists disagree about the ultimate metaphysics of causation,
they agree that, be it "real", or be it "secondary" and "occasional", there
is causation between mind and body. They both allow that the material
extended world is not causally closed and that, from time to time,
bodily movements are not subject to otherwise true conservation laws.
It is, however, an essential part of pre-established harmony that the
causes of changes in body are always other changes in body and never
changes in mind.
It is of course true that Leibniz did not deny that there are correlations between body and mind. After all, his pre-established harmony is
a doctrine about the union of body and mind, and not a straightforward
denial of it. He did not hold that a desire to move an arm and the arm's
movement are quite unrelated. In The Discourse on Metaphysics he
explicitly said that "the passions and actions of the ... [mind] are
accompanied by ... corresponding phenomena, in the [body]" ([20] 4,
458: [14]324). He was quite prepared to pair pin-pricks with pains. 3u
But since this is so can the difference between him and both occasionalist and interactionist really be that he asserts that the causes of
changes in body (or, mind) are always changes in body (or, mind) and
never changes in mind (or, body)? Can it really be that "[t]here lies the
essential difference in our views"?
It is not satisfactory to repeat that Leibniz denies that correlations
between body and mind, which he acknowledges, are causal, and that
this is why pre-established harmony is different from both interactionism and occasionalism. For, as it stands, this is a mere denial. What
needs to be realized is that that denial follows from something else
Leibniz says about those correlations. It follows from an as-yet unmen-
178
ROGER S. WOOLHOUSE
tioned difference between pre-established harmony and the views

Leibniz rejects. It is not his view, as it is Descartes' and Malebranche's,
that merely some physical (o'r, mental) events are related to mental (or,
physical) events, but that they all are. This came out most clearly in
Leibniz's 1702 reply to Bayle's later remarks on pre-established
harmony. "[C]ertain movements, rightly called involuntary, have been
ascribed to the body in such a way that nothing is believed to
correspond to them in the soul; and reciprocally, it is believed that
certain abstract thoughts are not represented at all in the body. But
there is an error in both of these views" ([20]4, 563: [14]580). In this
same reply Leibniz similarly claimed that "To all the movements of our
body there correspond certain more or less confused perceptions or
thoughts of our soul" and that "the soul never makes any resolutions to
which the movements of the body do not correspond" ([20]4, 559:
[14]577; also Discourse on Metaphysics ([20]4, 458-9: [14]325.
The view that every bodily (or, mental) event has a corresponding
mental (or, bodily) event, combined with the view that bodily (or,
mental) events always have bodily (or, mental) causes is inconsistent
with the view that the correlations between mind and body are causal.
If they were causal then bodily events would be over-determined and
sometimes have both a mental and bodily cause. 3l
4. CONCLUSION
One thing that has come out in the previous section is a reply to the
first part of Russell's comment about the over-boldness of Leibniz's
claim that had Descartes known of the law of conservation of momentum he would have been led to the doctrine of pre-established
harmony. It has come out that the objection Leibniz had at this point to
Descartes is one that applies, and was applied by Leibniz, to occasionalism too. What then of the second part of Russell's comment that
Spinoza's view of the union between mind and body, was surely another
possible alternative? There is something right, something almost too
right about this.
It is because of certain of its differences from them that Leibniz's
doctrine of pre-established harmony escapes the objection from physical impossibility which he has to occasionalism and interactionism.
Those theories hold, first, that bodily (or, mental) events sometimes
have bodily (or, mental) causes and sometimes mental (or, bodily)
179
causes; and, second, that the correlations between body and mind are
causal. According to the doctrine of pre-established harmony, however,
bodily (or, mental) events 'always have bodily (or, mental) causes;
second, for every bodily (or, mental) event there is a correlated mental
(or, bodily) event; and, third, the correlation between body and mind is
non-causal. Now precisely these three, associated, features of preestablished harmony are also features of Spinoza's view of the union
between body and mind; and in this sense Russell is almost too right
that Spinoza's view would have provided a possibility for Descartes
additional to Leibniz's. According to Spinoza, bodily (or, mental)
events always have bodily (or, mental) causes, never mental (or, bodily)
ones: "The modes of each attribute [thought, or extension] have God for
their cause only insofar as he is considered under the attribute of which
they are modes, and not insofar as he is considered under any other
attribute" ([26] pt. 2, prop. 6); or, more explicitly, "Body cannot
determine the Mind to thinking, and the Mind cannot determine the
Body to motion, to rest or to anything else" ([26] pt. 3, prop. 2; also pt.
5, pref.). Also according to Spinoza, every bodily (or, mental) event has
a corresponding correlated mental (or, bodily) event: "whether we
conceive nature under the attribute of Extension, or under the attribute
of Thought ... we shall find one and the same order" ([26] pt. 2, prop.
7 schol.). Finally, according to Spinoza, the correlation between mind
and body is non-causal. The union of mind and body is in representational terms of idea and object: "the Body is the object of the Mind"
([26] pt. 2, prop. 21 ).
But there is also something wrong, at least from Leibniz's point of
view, with the second part of Russell's comment. Consideration of this
brings us back to the second section of this paper and to Leibniz's
complaint against occasionalism that it involves miracles. At the bottom
of this objection was a conception of the created world as composed of
active substances which contain in their own natures the principle of
their changes. Now this conception is decidedly not something which
Leibniz and Spinoza share. Indeed, in his complaints about this aspect
of occasionalism, Leibniz more than once included Spinoza in its scope.
"The doctrine of occasional causes", he said in his 1698 article "On
nature itself", "is fraught with dangerous consequences .... So far is
this doctrine from increasing the glory of God by removing the idol of
nature ... it seems rather, like Spinoza, to make out of God the nature
of the world itself, by causing created things to disappear into mere
180
ROGERS.WOOLHOUSE
modifications of the one divine substance, since that which does not
act, which lacks active force, and which is despoiled of all distinctiveness and even of all reason 'and ground for subsistence can in no way
be a substance" ([20] 4, 515: (14] 506-7). This connection between
occasionalism and Spinoza is made elsewhere in the same paper. In the
course of arguing (as we saw in section 2) that God's commands for the
world must leave a permanent impression in the natures of created
things, he said that if this were not so there would be no created
substances: "everything would reduce to certain evanescent and flowing
modifications or phantasms, so to speak, of the one permanent divine
substance. And, what reduces to the same thing, God would be the
nature and substance of all things - a doctrine of most evil repute,
which a writer who was subtle indeed but irreligious, in recent years
imposed upon the world" ([20] 4, 508-9: [14] 502). Then, again, four
years later, in his 1702 reply to Bayle, Leibniz argued that unless one
rejects occasionalism and supposes that things do as they do because of
their own natures, one is committed to supposing that "there would be
no substances beyond his [God's] own - a view which would lead us
back into all the absurdities of Spinoza's God .... Spinoza's error
comes entirely from his having pushed too far the consequences of the
[occasionalist] doctrine which denies force and action to creatures"
([20]4,568: (14]583).
NOTES
Arnauld to Leibniz, 4 March 1687 (12012, 84: 1171105).
Foucher to Leibniz, Journal des savants, September 1695 (12011, 425: tr. R. Niall D.
Martin in 141 98).
.' Jaquelot to Leibniz, 2 February 1704 (1201 3,463).
4 Leibniz to Raymond de Montmort, 26 August 1714 (1181704).
5 Leibniz to Malebranche, 13 January 1679 (1201/, 328: 1141209).
h Leibniz to Malebranche, 22 June 1679 (1201/, 330: 1141 210).
7 Draft of a letter from Leibniz to Arnauld (1201 2, 68-9: 1171 84).
x All quotations in this paragraph are from Leibniz to Arnauld, 4/14 July 1686 (1201 2,
57-8: j1 7165).
~ Arnauld to Leibniz, 28 September 1686, quoting Leibniz to Arnauld, 4/14 July
1686 (12012, 64: 1171 78).
10 Arnauld to Leibniz, 4 March 1687 (120] 2,84: 11711(5).
II Arnauld to Leibniz, 4 March 1687 (120]2, 84: 1171105-6).
I~ All quotes in this paragraph are from Leibniz to Arnauld, 30 April 1687 (12012,
92-3: j171116).
1] All quotes in this paragraph are from "A new system", Journal des savants, 27 June
1695 (120]4,483: j14]457).
I
LEIBNIZ AND OCCASION ALISM
181
14 "Clarification of the difficulties which Mr Bayle has found in the new system of the
union of soul and body", Histoire des ouvrages des savants, July 1698 (1201 4, 520:
[141 494).
"Reply to the thoughts on the system of preestablished harmony contained in the
second edition of Mr Bayle's Critical Dictionary, article 'Rorarius''', Histoire critique de
la republique des lettres, 1702 (1201 4,554-71: 1141 574-85).
16 He also remakes it in "Considerations on vital principles and plastic natures, by the
author of the system of preestablished harmony", Histoire des ouvrages des savants,
May 1705 (1201 6,541: 1141 587).
17 Arnauld to Leibniz, 4 March 1687 (120J 2,84: [171105-6).
IN "On nature itself, or on the inherent force and actions of created things", Acta
emditomm, September 1698 (120J 4,507: J14J 500-1).
IY Leibniz to Fillipi, January 1680 (1201 4,286).
20 For example in "A brief demonstration of a notable error of Descartes and others
concerning a natural law, according to which God is said always to conserve the same
quantity of motion; a law which they also misuse in mechanics", Acta emditomm,
March 1686 (121] 6, 117-23: J14J 296-301).
21 "Explanation of the new system of the communication between substances, by way
of reply to what is said about it in the Journal of 12 September 1695", Journal des
savants, April 1696 ([20J 4, 497: J16J 327); Leibniz to Bernoulli, 1696 (121J 3, 243:
Jl6J 328, n. 30).
" Specimen dynamicum (1695) (121 J 6, 240: J14J 439); letter to Nicolas Remond, 10
January 1714 (l20J 4,607: Jl4J 655).
2.1 "Brief demonstration" (see note 20); Discourse on Metaphysics(l20J 4, 441-4:
J141314-15).
24 Descartes does not make exactly this identity: "force to continue motion in a straight
line ... does not depend solely on the quantity of matter that is in each body, but also
on the extent of its surface" (191 113).
25 It also appears clearly in the letter to Bernoulli, 1696 (1211 3, 243: Jl61 328, n. 30).
Discussion of some such objection as Leibniz's can be found down to the present day
(l3J 103-9; 16J 253-8; 110J 88-9; 113J 64-5; 122J 86; 125J 256; 127J 126-43;
1281291-2,298-9).
26 Presumably modern accounts of how Descartes took this way out are blindly
following Leibniz, cf. j111205; 123J 81; 124J 583; 127J 129. In his 1702 reply to Bayle
(see note 15) Leibniz attributes this way out to "the Cartesians" rather than to
Descartes himself (l20J 4, 559: 114J 577).
27 See 1121130, n. 35 for references earlier than Leibniz's.
2N "Considerations on vital principles and plastic natures, by the author of the system
of pre-established harmony", Histoire des ouvrages des savants, May 1705 ([20J 6, 540:
1141587); Monadology (1714) (12016, 620-1; Jl61263-4); Leibniz to Remond, 10
January 1714 (12013, 607: Jl41655).
29 This, from "Considerations on vital principles" (see note 28) (1201 6, 540: J14J 587),
is the earliest instance I know of this stronger claim .
.10 See "Reply to the thoughts ... in the second edition of Mr Bayle" (see note 15)
(120J 4,559: [14J 577; [20J 2, 70, 111-12, 113-14: Jl7186, 143, 146).
.'1 C. A. Strong is the only relatively recent writer I know of who is quite clear that
there is a connection between the three features of Leibniz's view (the features that, (1)
bodily (or, mental) events have only bodily (or, mental) causes, (2) there is a thorough-
I,
182
ROGERS.WOOLHOUSE
going correlation between body and mind, and (3) the relation between mind and body
is non-causal. He explicitly points out that "[t]he conception of the physical world as a
closed circle would imply an exact. 'psychophysical representation' for all mental states,
to account for their apparent action" ([27]74). Later he makes clear that "[t]he
assumption of ... any dependence of ... [mind (or, body)] on [body (or, mind)], not
only contradicts its [psychophysical parallelism's] anti-causal essence, but is inconsistent
with the thorough-going correspondence it involves" ([27]79).
On the other hand, C. D. Broad misses the connection completely. Speaking of "two
sides" to psychophysical parallelism (a "negative" one, that mind and body do not
causally interact, and a "positive" one, that there is a one-one correlation between
mental and physical events), he says that the second might be accepted by someone
who rejected the first and that the second is compatible with interactionism ([3]121,
133).
REFERENCES
1. Bayle, Pierre, Historical and Critical Dictionary: Selections, tr. R. H. Popkin
(Bobbs-Merrill: Indianapolis, 1965).
2. Bayley, Benjamin, 'Of the immateriality of the soul, and its distinction from the
body', in 'An appendix containing some pieces found among Mr. Toland's papers',
in A Collection of Some Pieces of Mr. John Toland, 2 vols (London, 1726;
reprinted New York: Garland, 1977) 2,3-28.
3. Broad, C. D., The Mind and Its Place in Nature (London: Routledge and Kegan
Paul, 1925).
4. Brown, Stuart, Leibniz (Milton Keynes: Open Univ. Press, 1983).
5. Clarke, Samuel, A Demonstration of the Being and Attributes of God (London,
1705; reprinted Stuttgart: Frommann, 1964).
6. Cornmann, James W. and Keith Lehrer, Philosophical Problems and Arguments:
An Introduction (2nd ed., New York: Macmillan, 1974).
7. Descartes, Rene, Descartes: Philosophical Letters, tr. and ed. A. Kenny (Oxford:
Blackwell, 1970).
8. Descartes, Rene, Principles of Philosophy (1644), in volume 1, The Philosophical
Writings of Descartes, 2 vols, tr. J. Cottingham, R. Stoothoff, and D. Murdoch
9. Descartes, Rene, Le Monde, ou traite de la lumiere (1664), Eng. tr. M. S.
Mahoney (New York: Abaris, 1979).
10. Ducasse, c., 'In defense of dualism', in Dimensions of Mind, ed. S. Hook (London:
Collier-Macmillan, 1960),85-9.
II. Gale, G., 'Leibniz's dynamical metaphysics and the origin of the vis viva controversy', Systematics 11(1973), 184-207.
12. Garber, Daniel, 'Mind, body and the laws of nature in Descartes and Leibniz',
Midwest Studies in Philosophy 8 (1983),105-33.
13. Hoffding, H., Outlines of Psychology (London, 1891).
14. Leibniz, G. W., Leibniz: Philosophical Papers and Letters, tr. and ed. L. E.
Loemker (2nd ed., Dordrecht: Reidel, 1969).
183
15. Leibniz, G. W., Theodicy, by G. W. Leibniz (1710), tr. E. M. Huggard (London:

Routledge and Kegan Paul, 1951).
16. Leibniz, G. W., Leibniz: The Monadology and other Philosophical Writings, tr. R.
Latta (London, 1898).
17. Leibniz, G. W., The Leibniz-Arnauld Correspondence, tr. and ed. H. T. Mason
(Manchester: Manchester Univ. Press, 1967).
18. Leibniz, G. W., Opera philosophica quae extant, ed. J. E. Erdmann (Berlin, 1840).
19. Leibniz, G. W., System of Theology, tr. C. W. Russell (London, 1850).
20. Leibniz, G. W., Die philosophischen Schriften, 7 vols, ed. C. I. Gerhardt (Berlin,
1875-90).
Halle, 1849-60).
22. Paulsen, F., Introduction to Philosophy, tr. F. Thilly (New York: Holt, 1912).
23. Russell, B., The Philosophy of Leibniz (London: Allen and Unwin, 1900).
24. Russell, B., History of Western Philosophy, (London: Allen and Unwin, 1946).
25. Sellars, R. W., The Essentials of Philosophy (Macmillan: New York, 1917).
26. Spinoza, B., Ethics, in The Collected Works of Spinoza, volume 1, ed. and tr. E.
Curley (Princeton NJ: Princeton Univ. Press, 1985).
27. Strong, C. A., Why the Mind Has a Body (New York: Macmillan, 1903).
28. Watts Cunningham, G., Problems of Philosophy (London: Harrap, 1925).
MARY B. HESSE
VICO'S HEROIC METAPHOR
To Gerd Buchdahl,
who taught us the art of hermeneutics
in history of philosophy of science.
Abstract. Vieo's "new science of humanity" is best known as an attempt to supplement
Baconian methodology with a study of human history based on a method of poetic
imagination. Vieo took the key to this study to be a theory of the poetic origins of
language, which he analysed into three "ages", parallel to the cyclic repetitions of
history: the "divine" language of signs and gestures, the "heroie" language of emblems
and metaphors, and the "rational" language of science and logic. Two questions arising
from Vieo's theory are of particular importance for modern discussions of language.
First, it is argued in this essay that for Vieo language is primarily conventional, and
must therefore be studied as a social institution rather than a natural phenomenon, and
that it incorporates social value judgments. Second, the nature of "metaphor" and its
dependence on the imaginative faculty is discussed, and it is concluded that Vico meant
metaphor not only to be typical of the first two ages of language, but to have an
essential cognitive function also in the third, rational age.
1. THE WISDOM OF THE ANCIENTS
In 1710 Giambattista Vieo 1 published a work entitled De antiquissima

Italorum sapientia, 2 as a conscious but exaggerated echo of Francis
Bacon's De sapientia veterum of 1609. Bacon had adopted the then
popular device of interpreting the ancient Greek myths as allegories of
cosmological and philosophical theories, for example, "Cupid's nakedness" is an allegory of primary matter and its fated motions, that is, of
atomism. 3 Vico, however, rejects all such speculations about the
meaning of myth, and shifts attention to its underlying metaphorical
language, setting out to derive the "most ancient Italian wisdom"
through an etymology and semantics of the latin vocabulary ([45] 45f,
[10]). He later repudiated this attempt also, but the nature of metaphor
remains fundamental in his thought. In this respect his approach to
language is so different from that of the main philosophical traditions
from the seventeenth to the twentieth centuries that it is worth trying to
185
186
MARY B. HESSE
reconstruct it, especially now that debate about the metaphoric nature
of language has revived among philosophers as well as literary critics.4
Vico's admiration for Bacon is tempered by his view that concentration on "physics and geometry" in Baconian science has led to neglect
of ethics and the study of man and history,5 and he rejects Bacon's
scientific ideal of an "Alphabet of Nature".6 Bacon had followed the
tradition that the ideal language was given by God to Adam, who
named the animals, and that this language was lost at the flood. Bacon
saw his task as being to restore it, and with it the knowledge of nature
which man was meant to have in his pristine state. The scientific
language will be an alphabet, in which letters correspond to the natures
of things, and words and sentences will be made up according to rules
that correspond to the causal relations among things, that is, to the laws
of nature. The notion that this language will be a return to a lost
original was largely abandoned by the end of the seventeenth century,
but the vision of its future establishment by the accumulation of
knowledge was not. It is maintained through early Royal Society
attempts at a "Real Character", and in Leibniz's "Characteristica Universalis",7 and in twentieth-century attempts at a formalization of
language based on logic and science rather than natural speech.
We may take Hobbes's theory of "naming" as typical of the way the
Adamic tradition was received by Vico.8 Hobbes rejects the idea of the
original Adamic language, noting that it is not complete: the "names of
animals" do not include other ingredients of language such as numbers,
qualities, and abstract names. Mere names of things can be private
marks, but genuine language is a social phenomenon: "to show to others
that knowledge which we have attained, which is, to counsel and teach
one another" ([27) ch. 4, p. 19). Hobbes does, however, share the
aspiration for an ideal language which develops with man's rationality
and is distinct from lower-level animal communication. We use arbitrary signs for conceptions of universals, but there are no universals in
the world: there is "nothing in the world universal but names; for the
things named are every one of them undivided and singular" ([27) ch. 4,
p. 19; [26) 1, ch. 2.4). The same universal name is imposed on many
things by means of their similarity, and is used to recall to mind any
one of these singular things. Although words for universal names are
conventional, the proper definition and use of the names is not.
Hobbes' ideal is geometry: here the right conceptions are given proper
definitions, and analytic truths are known concerning their causal
187
relations. In general, the causes of universal things must be known

before the causes of singular things, although ''universal things are
contained in the nature of singular things" ([26] J, ch. 6.4); that is, it is
only in singular things that the causal relations have real existence.
These causal relations are therefore known by applying analytic truths
to particulars by the synthetic method.
Thus, for Hobbes, ideal language is both conventional and natural. It
is conventional in that words expressing universals are chosen arbitrarily, but it is natural in that there is a proper way of eliciting
universal terms from the similarities of particulars, and this way will
enable analytic truths to be known which are at the same time the
implicit causal relations between things. The conception of science is
very like that of Galileo: there is an idealization of phenomena, for
example the abstraction of "constant acceleration" from the behaviour
of falling bodies, and the analytic derivation of simple mathematical
relations between the variables of motion, which are then taken to
express the true laws of motion of particulars.9
This is not the place to comment further on Hobbes' unjustified faith
in the analytic derivation of causal relations in science. But it is this that
provides motivation for his attitude to linguistic equivocation and
metaphor; an attitude from which Vico departs more decisively than
from other features of seventeenth-century linguistic theory. For
Hobbes reasoning is essentially computation with universal names,
analytic relations and syllogisms ([27] ch. 5, p. 25; [26]1, ch. 1.2). It
follows that "inconstant names", which are applied differently by
different people, cannot be "true grounds of any ratiocination. No more
can metaphors, and tropes of speech: but these are less dangerous,
because they profess their inconstancy: which the other do not" ([27]
ch. 4, p. 25; [26]1, ch. 2.12). But "in reckoning and seeking of truth,
such speeches are not to be admitted" ([27] ch. 5, p. 28). Metaphors,
indeed, are abuses of speech by use of words ''in other sense than that
they are ordained for; and thereby deceive others" ([27] ch. 4, p. 19).
Vico's relation to the Bacon-Hobbes tradition can be summarized in
four points. First, he objects to Bacon's presentation of ancient wisdom
as hidden in myths and allegories, and needing to be brought to light by
the open inductive methods of science. For Vico the first language is
not esoteric science, but poetic imagery which was openly expressed
and generally understood. It does, however, need interpretation for our
day, not as the unravelling of a cryptogram into our rational pre-
188
MARY B. HESSE
suppositions, nor as a product of inductive laws of nature, but as a

hermeneutic exercise. Vico does not suppose that interpretation is a
direct recovery of the "savage mind", that is of the mythopoeic view of
the world, in the later fashion of attempts to reconstruct "actor's
meaning", or of a Collingwoodian empathy. "With our civilized natures",
he says, "we cannot at all imagine and can understand only by great toil
the poetic nature of these first men" ([52134; see [351 166f; [6196;
[71 80). Interpretation is possible for us only because we share in
general the imaginative faculty of human beings, and can therefore
describe and so far understand what their mythological beliefs were,
without being able to share them.
Secondly, Vico agrees with Hobbes against the individualism implicit
in the Adamic tradition, and in subsequent Cartesian philosophy of
reason. Language is not a matter of private reference to mental images,
but is essentially social. For Vico its study must be historical, and must
show how it constitutes the basis of social order in the different ages of
social development. There are three such ages, and correspondingly
three types of language. In the first "divine" age, language consists of
gestures, signs and physical objects having natural relations to the ideas
expressed. This language is "mute", having neither speech nor letters. In
the second "heroic" age, language consists of "emblems, or similitudes,
comparisons, images, metaphors and natural descriptions" (32). It is a
language of symbols, typified by the heraldic devices of the "heroes".
The third age is called variously "human", "vulgar", or "rational".
Language is then articulate, alphabetical and conventional, and meanings "are agreed upon by the people" (32, 52, 173,432,928-31). This
language is sometimes called simply "rational", although it must be
remembered that for Vico there are also three kinds of "reason"
appropriate to the three ages (947f).
It is important to recognize that the "ages" are not stages in a single
linear development of history, but are repeated in modified forms by all
nations, including contemporary societies. The stages are spiral-like
rather than purely cyclic, for societies traverse them and return to
barbarism, but at a different and potentially worse level than the
primitive (1106). Vico calls this the corsi and ricorsi of "ideal eternal
history". It does, however, have an eventual historical terminus, for,
presumably in line with Christian doctrine, Vico rejects the idea that
there are infinite worlds "born from time to time throughout eternity"
(1096, 348). Rome is the major example of the ricorsi (e.g., 114), but
189
there is also mention of the subsequent decline of (Latin) language into

barbarism in Spain, France and Italy (485), and the rise of France again
in the twelfth century (159). Even the ancient Egyptians are said to
have identified the three ages prior to their own time (52, 173, 432,
915). Only the Hebrews and Christianity escape the regular corsi, since
the Hebrews are said to have had no primitive age, and it was
Christianity that prevented total return to barbarism after the fall of
Rome (1047).10
Thirdly, the question arises whether Vieo's theory of the origin of
language is "naturalist" or "conventionalist". The question is important
because it relates to the kind of knowledge we can have of language
according to Vieo: is language like the natural world, whieh God
created, or like human society, whieh is created by men? The answer is
not clear-cut, any more than it is in the case of Hobbes. It will be
discussed in the next section.
Fourthly, Vieo agrees with Hobbes, against the Adamie tradition,
that the first languages were not God-given categorizations "in accordance with the nature of the things it dealt with ... the giving of names
to things according to the nature of each" (401). On the contrary,
according to Vieo, they were languages of poetie imagery, largely drawn
from personifications and comparisons with the human body, to
express what later became abstract classifications. Vieo calls the poetic
origin of language the "master key of the New Science" (34). It is a
principle that conflicts with the theory of a natural, ideal Adamie
language "in the beginning", but still leaves open the question whether
the "pre-rational" peoples had a poetic language only because they had
no fully developed rational categories, or whether metaphor and other
linguistic tropes have an essential place in all languages. It also leaves
open the the related question of whether Vieo agrees with Hobbes that
an ideal, literal language, corresponding to the nature of things, is the
proper and attainable goal of a true science. The discussion of these
questions will take up most of this paper from sections 3 to 5.
2. LANGUAGE, NATURAL OR CONVENTIONAL?
As in the case of Hobbes, Vieo's theory of language is neither purely

conventional nor purely natural. Most interpreters take him to have
decisively rejected mere conventionalism, certainly for the first two ages
of language, and probably also for the third (see [10] 286f; [11]583;
190
MARY B. HESSE
[30) 365). He stresses the "natural significations" of the first hieroglyphic

characters and the heroic symbols, and explicitly contrasts these with
the conventions "agreed upon by the peoples, a language of which they
are absolute lords" (32) in the third, democratic, age. Before the age of
human reason, men, like children, were unable "to form intelligible
class concepts of things, and had a natural need to create poetic
characters" (209) which had natural resemblances to the things signified. On the other hand, when powers of abstraction develop in the
third age, Vico describes the choice of signs as "conventional", but not
as wholly conventional, for he objects to the philologians who "have
accepted with an excess of good faith the view that in the vulgar
languages meanings were fixed by convention. On the contrary, because
of their natural origins, they must have had natural significations" (444).
The conception seems to be that, although most alphabetic signs of
third-age language, are, as is obvious, conventional, nevertheless their
origins in natural significations, metaphors of resemblance, onomatopoeia, etc., remain pervasively present. Gardiner concludes that, for
Vico, "language is not an artificial medium which men have deliberately
constructed to give expression to pre-existent ideas, but has evolved
naturally, the course of its development being inseparable from that of
the human mind itself" ([16) 249).
There are, however, difficulties about this conclusion. The first
follows from passages in Scienza Nuova where Vico states that men
themselves created the images of the gods and the first languages, as
well as all other social institutions:
We shall show clearly and distinctly how the founders of gentile humanity by mellilS of
their natural theology (or metaphysics) imagined the gods, how by means of their logic
they invented languages; by morals, created heroes; by economics, founded families,
and by politics, cities; by their physics, established the beginnings of things as all divine;
by the particular physics of man, in a certain sense created themselves (367).
And introducing the question of how the three languages and their
characters were formed he says:
We must establish this principle: that as gods, heroes, and men began at the same time
(for they were, after all, men who imagined the gods and believed their own heroic
nature to be a mixture of the divine and human natures), so these three languages began
at the same time, each having its letters, which developed along with it (446, cf. 692,
916).
191
These two passages seem to mean that the first languages, along with all
social institutions, were created by men and were therefore in that
sense conventional. It also appears that there was never an entirely
"bestial" age such as Vico describes elsewhere as the state of the ''first
men". There was no historic time in which humans had only a gestural
or symbolic language, but all three types of language are coeval.
There is a second, more serious, difficulty in supposing that Vico's
language has a wholly natural character. This concerns his view of the
nature of knowledge itself, and the possibility of knowing the history of
man and society. The primary aim of Scienza Nuova is to develop a
new science of the human to complement Baconian science of the
natural world, and the best known feature of Vico's work is his claim
that we can know the truth only of what we ourselves have made. This
is the verum-factum principle, expressed in Scienza Nuova as "the
eternal and never failing light of a truth beyond all question: that the
world of civil society has certainly been made by men, and that its
principles are therefore to be found within the modifications of our
own human mind" (331). The principle is taken for granted and little
discussed in Scienza Nuova, but Vico discussed it at length in earlier
works, where his theory of natural knowledge is developed and never in
essence repudiated. I I
Nature is made by God, hence Vico believes we cannot completely
know it, except insofar as we construct experiments and applications in
terms of our own categories. In his description of natural science Vico
echoes Bacon's emphasis on experimental "operations", and Lockean
doubts as to whether we can ever know the corpuscular essences of
things. Physics is not capable of logical demonstration - it is "mere
probability", and in spite of its great successes, we should not assume
that physics is "identical with nature itself",i2 Again, when the study of
chemistry was "directed towards something quite useless", that is
towards natural truths that were "denied it by nature", nevertheless it
"exceeded the end it had set itself and produced pharmacology, a
working art of great use to the human race":
Human knowledge arises, therefore, from a defect of our mind, i.e., from its extremely
limited character, as a result of which, being external to everything and not containing
what it strives to know, it does not produce the truths which are its aim. The most
certain things are those which, redressing the defects of their origin, resemble divine
knowledge in their operation, inasmuch as in them the true is convertible with what is
made. 13
192
MARY B. HESSE
It seems to follow that insofar as human language is a natural

phenomenon, in the sense that the subject-matter of physics is natural,
then it cannot be completely known, and its study does not share the
"certainty" of the knowledge of human institutions. If we conclude from
Vico's arguments about the natural character of the first languages that
he has in mind a purely animal-like origin of significant gestures and
speech, then the history of language would be like the history of
biological evolution, and share its nature as a "probable science", a
science whose subject-matter God alone made. Such a conclusion
seems quite inconsistent with Vico's claim that the history of the origin
of languages is the master-key for the whole history of human institutions, which were made by men, and which are the subject-matter of the
"new", non-Baconian science.
This dilemma can be resolved by distinguishing between "natural" as
applied to the God-created subject-matter of the natural sciences, and
"natural" as applied to the development of human society, and by
noting that in Vico's view the second sense applies not only to the first
languages, but also to the origins of other human institutions. The full
title of Scienza Nuova in its first (1725) edition is Principles of a New
Science concerning the Nature of the Nations, by which are found the
Principles of Another System of the Natural Law of the Gentes. 14 In
their Introduction to the translation of the third edition, Bergin and

Fisch note that by "nature" Vico means simply "birth" or "origins" (SN
p. xxii, cf. (147-8, and that, like Aristotle, he tends to elide the
standard distinction between "natural" man on the one hand and the
conventional institutions of the polis on the other:
Vico here agrees with Aristotle. When he calls the world of nations the world of men,
he means simply that what were beasts in the world of nature become men in the world
of nations, and it is by the becoming of the world of nations that they become men. Or,
as he puts it otherwise, in a sense they make the world of nations, and in the same sense
they make themselves by making it (367, 520, 692; SN p. xxiv).
The "natural law of the gentes" is then to be understood as the laws of

human development in the three historic ages (SN p. xxvii).
This conclusion is confirmed by the way Vico continually uses the
term "nature" to mean both "origins" and "essence" in reference to all
social institutions: religion, language, marriage, law, morals, politics
(507, 584, 916, 1008). And he implicitly distinguishes his view of the
first men from the naturalistic view of Hobbes, when he interprets the
193
"so-called state of nature" as the "family state", which was already a

social institution in the first age (955).
How then does Vico discriminate between what is created by God
and hence yields only probable knowledge, and what is created by men
and can hence be known with certainty? It seems that there is no sharp
distinction. One clue is that it is providence that ordains a special place
for human beings: "Hence providence ordained that the first peoples,
ferocious by nature, should be persuaded by this their religion to
acquiesce naturally in force" (923).15 Providence has a distinctive place
for the human race within created nature. That place is specific, in that
there is a common human nature across all historic ages and places,
and Vico does not discuss further how the "break" came between the
truly "animal" and the "bestial" forms of primitive man. In particular he
does not believe in the diffusion of human culture and society from one
geographical place to all others: he does not encourage us to think of a
sudden unique creation of man. Nevertheless, human nature being what
it is, there is a commonality of "modifications of the mind" which
means that we can in some sense understand the human as we cannot
understand nature, not directly by "empathy", but by imaginative
descriptions of alien institutions and beliefs as responses to social
problems that we all share (see [34) 16lf; [35) 166f; [6) 96; [7) 80).
3. VICO'S THEORY OF IMAGINATIVE GENERA
Vico emphasises that the "master key" which "has cost us the persistent
research of almost all our literary life" (34), is the principle that the
primitives had by "necessity of nature" to speak in poetic characters
because of the "poverty" of their language and their inability to form
intelligible class concepts (209, 384). Before language men were like
Proteus looking at his own reflection in the water: "their minds were so
limited to particulars that they regarded every change of facial expression as a new face, ... and for every new passion they imagined a new
heart, a new breast, a new spirit" (700, cf. 688). The poetic characters
of the first language are "imaginative genera" to which primitives
"reduced all the species or all the particulars appertaining to each
genus" (34). To understand the concept of imaginative genera, it is
useful to look first at Vico's account of "genera or ideas" in chapter 2 of
De sapientia where he discusses "metaphysical forms". Vico later
repudiated much of the metaphysics of De sapientia ([47) 153), but it is
194
MARY B. HESSE
clear that the metaphysical forms are the forerunners of the imaginative
genera, although they are more dependent on Vico's account of the
physical science than is the case in Scienze Nuova.
In De sapientia Vico identifies forms with the Latin "genera", and
asserts that
the ancient philosophers of Italy believed the genera to be forms which were infinite,
not in extension but in perfection. . . . On the other hand, they held that species, or
particular things, were representations modelled on these forms (152]60).
Metaphysical forms are like those used by a modeller: they are "made
in the mind", and in the case of nature they are made by God. Physical
forms, on the other hand, are like seeds, which "become more complete
as the seed develops each day" and approaches the perfection of its
metaphysical form. Geometry and the human arts, such as painting,
sculpture modelling and architecture, participate in the divine modelling process, because they are concerned with "prototypes which the
human mind contains within itself" ([52]61). Arts such as oratory,
politics and medicine do not approach perfection, because their forms
are not present in the human mind. In particular, Aristotelian physics
"confuses the forms", because it claims that the physical forms are
universal, whereas "all particular forms are imperfect",16 and only
Platonic forms attain perfection. It is a mistake to rely on universal
terms, which are in fact homonyms or ambiguities, because they lead to
neglect of the differences between particulars: ''universal genera [in the
Aristotelian sense] do not provide for what is new, wonderful and
unexpected" ([52]62, cf. 73). Vico goes on to draw a conclusion about
methods of argument, and to favour induction over Aristotle's syllogistic or Zeno's sorites:
It is a likely conjecture that the ancient philosophers of Italy approved neither the
syllogism nor the sorites, but used induction based upon resemblance in their arguments. . .. ]T]he most ancient dialectic of all was induction and the comparison of
resemblances, which Socrates was the last to employ (152174).
In other words, the validity of formal logic always depends on identities

of the forms of particular things, but there are no such identities in the
world, therefore inference about particulars must proceed by induction
from resemblances. I 7
Metaphysical forms connect with the physical world by means of the
"elements" God made, which Vico calls "metaphysical points".18 These
195
are modelled on the notion of geometrical points, which mediate

between man's God-like "making" of the geometrical truths, and their
realization in the physical world. Fisch, Verene and other commentators have pointed out that the conception is similar to Galileo's
theoretical ideals realized in experimental operations, which we have
seen also to be influential in the case of Hobbes. By the time Vico came
to write Scienza Nuova, however, he had abandoned his view that
geometry is the model for human science, and had also rejected the
distinction between metaphysical forms as ideal identities, and universal
words as homonyms. In Scienza Nuova, as we shall see, it is possible to
use signs that stand for univocal ideals.
Metaphysical genera are introduced in Scienza Nuova in terms of
the poetic characters that necessarily constituted the first languages.
Egyptian hieroglyphics are the best-known examples, but more illuminating is the story of the Scythian king who sent "five real words" to
Darius with whom he was at war (435, ct. 48). The "words" were a frog,
a mouse, a bird, a ploughshare and a bow, denoting respectively that
the king was born of Scythian earth, made his home there, had taken
the auspices, had tamed his land by ploughing, and had the duty to
defend it. These were clear and open meanings, and so were all primitive signs, contrary to the scholars who regarded Egyptian hieroglyphics
as a secret, priestly language. It is only we, in the "rational" age, who
need an articulate commentary.
In general, myths and fables describe divine or human personalities
("characters" in a non-linguistic sense) which constitute both signs and
ideals for some aspect of human experience. For example, Hermes is
the Egyptian poetic character for one "wise in vulgar wisdom" (68),
Orpheus is the image of the theological poet (81), and similarly for
Homer's gods and heroes (809). About such characters Vico says that
they "were true fables or myths, and their allegories are found to
contain meanings not analogical but univocal, not philosophical but
historical, of the peoples of Greece of those times" (34, ct. 210, 403).
This insistence on "univocal" truth in connection with metaphor and
myth may appear surprising. Part of Vico's meaning may be discerned
from the example of Godfrey the Crusader as described by Tasso
(205). Godfrey is the "true war chief ... and all the chiefs who do not
conform throughout to Godfrey are not true chiefs of war". "True" here
clearly means ideal, proper, perfect, and this sense explains how Vico
can say just before giving the example that "poetic truth is metaphysical
196
MARY B. HESSE
truth, and physical truth which is not in conformity with it should be

considered false" (205). The characters are "ideal types", and once
enshrined in myth, they have social force: "peoples who have first
created heroic characters for themselves will afterward apprehend
human customs only in terms of characters made famous by luminous
examples" (809). It is only later in historical development that men fall
from the ideal in their actions, and then the fables become corrupted
and their original meanings lost, so that they appear, as Bacon held, to
be part of a hidden lore (81 ).
Not only does Vico identify poetic characters as ideal types in every
early human language, but he also states and claims to solve the
problem of "incommensurability" between languages: ''With our civilized
natures we cannot at all imagine and can understand only by great toil
the poetic nature of these first men" (34). But a "great principle" of
Vico's New Science is that "uniform ideas originating among entire
peoples unknown to each other must have a common ground of truth"
(144). In speaking of a common ground of truth, he seems to have in
mind the same sense of ideal truth that I have elucidated above, for he
says later that "there must in the nature of human institutions be a
mental language common to all nations, which uniformly grasps the
substance of things feasible in human social life" (161). In other words,
what is true in the "common sense of the human race" (145) is what has
a normative law-like relation to the conditions of human existence, and
forms what he calls the "natural law of the gentes" which it is the aim of
the new science to discover. There are three elements of the sensus
communis which form the fundamental principles of the science: "(1)
divine providence, (2) marriage and therewith moderation of the
passions, and (3) burial and therewith immortality of human souls....
And let him who would transgress them beware lest he transgress all
humanity" (360, d.130, 333).19
From these common notions a "mental dictionary" can be formed in
terms of which the origins of all the diverse languages can be assigned,
and the histories of all nations can be developed. There is a pattern or
"ideal eternal history" of the common notions which become embodied
in the institutions of language, religion, statehood, law, science and
morals, all of which develop through their appropriate three stages in
each human society (35,145).
Since imaginative genera are crucial to our understanding of Vico's
claims for the essential function of metaphor, we must go a bit deeper
197
into the way he conceives the relation between these genera and the
intelligible class concepts of the rational ages of history. Parallel with
the three ages, he says, there is first feeling, then perceiving "with a
troubled and agitated spirit", and then reflection "with a clear mind"
(218); "poets were the sense and philosophers the intellect of human
wisdom" (779). Poetic sentences, he goes on,
are formed by feelings of passion and emotion, whereas philosophic sentences are
formed by reflection and reasoning. The more the latter rise toward universals, the
closer they approach the truth; the more the former descend to particulars, the more
certain they become (219, cf. 821).
This contrast between "truth" and "certainty" seems to sit uncomfortably with the sense of poetic fable as "true" which we have
discerned above. Pompa and McMullin ([341, [31]) interpret the
contrast here as similar to the hypothetico-deductive method, where the
philosophical knowledge of universals is a realistic interpretation of
scientific theory, and the "poetic certainty" of particulars is the empirical check provided by mutual reinforcement of theory and evidence.
Indeed it is clear in his discussion of method that Vico does oppose
both Cartesian a priorism and simple Baconian induction. But without
further elaboration the alternative hypothetico-deductive interpretation
neglects the special and original character of Vico's theory of representation, which is based on the ideal as well as the factual, on imagination
as well as cognition, and on rhetorical as well as literal and argumentative uses of language. In developing this idea I shall largely follow
Verene's account of imaginative genera, which is more adequate than
the hypothetico-deductive account in enabling us to see how Vico's
theory of metaphor is of fundamental importance. 2o
Verene takes the distinction between the "imaginative universal" and
the "intelligible universal" ([45171f) to belong to the distinctive forms
of language of the first two and the third ages respectively. Primitive
language is essentially metaphorical, but in a special sense of "metaphor". Basic to all human perception and representation is the ability to
impose order on the flux of experience by imagining unities and
identities (Vico's word for imagination is fantasia). This is quite
different from the process of abstraction that yields intellectual universals as the common content of diverse particulars, which is the basis of
Aristotelian class logic. Such abstraction requires there to be factual
identities between properties if Aristotle's real forms are to result. 21
198
MARY B. HESSE
Abstraction relies upon similarities of particulars, and is necessarily a

later stage of representation, because it depends on the identities of
particulars already being established by their own imaginative universals. Vico understands not only that particulars have to be named by
universal terms to make language possible at all, but also that this
depends on a prior, inarticulate (Le., language-less) perception of
particulars as already imaginative classes. These primitive classes
cannot indefinitely be analysed into diverse particulars and their
abstract properties, on pain of infinite regress. Properties are only
known as realized in particulars, so some particulars must be primary.
In this sense the imaginative concepts or fables are "true" and "univocal" because there is nothing else in the first language for them to be.
So, according to Verene, metaphor is a "primal perception of identity".
"Metaphor" is not used in the sense that meanings are consciously
transferred or constructed by abstracting similarities of particulars:
"Metaphor is that by which identity is originally achieved in perception.
It is the form perception most immediately takes. Metaphor is the first
of the tropes and the first of the operations of mind in the act of
knowing" ([451 79-80, my italics).
Verene's analysis of imaginative universals steers a judicious course
between an inappropriate rationalization of Vico on the one hand, and
romanticization of him on the other: "Primitives have often been seen
as naive or confused or as simply giving 'metaphorical' accounts of
ordinary empirical events, but this is not true in Vico's view" ([45177).
There is only one point I would wish to add that Verene does not
emphasise. This is the ideal or normative character of imaginative
universals. Metaphors are "ideal portraits": "the first men ... had a
natural need to create poetic characters; that is, imaginative class
concepts or universals, to which, as to certain models or ideal portraits,
to reduce all the particular species which resembled them" (209).22
Without this feature of idealization it is difficult to understand how
Vico could claim "truth" (almost by definition) for the first imaginative
class concepts drawn pre-linguistically from experience.
We are now familiar with the notion that perception and observation
are "theory-laden", and also with the problem that, since theories
change, observation statements expressed in the idiom of one theory
do not necessarily preserve truth or meaning in another. And yet Vico
can speak of a "degeneration" of meanings and from truth to falsity in
the history of particular fables. These apparently absolute judgments
199
about truth only make sense if we assume Vico to have had a strong
belief in the appropriateness of the "original" imaginative classes.
Backing for such a belief comes from his theory of the "mental
dictionary", which represents those concepts which are appropriate and
normative for all human societies as such. The original fables are
necessarily true and "ideal", not because they belong to any particular
historical events, but because they belong to the very conditions of life,
although they later become corrupted in the cycle of history. In this
belief Vico may be said to reveal his own version of the Adamic myth.
It is not that Adam had uncorrupted knowledge of the nature of
physical things, but that in the first languages human beings were
directly in touch with the true springs of social reality, and after that the
Fall (d. 401-2).
In this interpretation of the imaginative genera Vico's problem is
taken to be the process of concept formation before articulate language;
a process that involves for the primitive mind implicit normative
judgments as well as pre-linguistic perceptions. A post-Kantian philosophical tradition would probably dismiss this problem as merely
psychological. But Vico is explicitly concerned with the interface of
what he calls "philosophy" and "philology", where by "philology" he
means the study of the origins of language and more generally with the
historical conditions of apprehension of particulars. When, however, he
contrasts the ascent of philosophy towards universals and "truth", with
the descent of poetry towards particulars and "certainty", he has in
mind a sense of "philosophical truth" which is different from the
original imaginative truth of fables. The philosophical sense is appropriate to the analytic "rational" stage of human thinking - the stage
where the principles of the New Science itself get established, and
abstract intelligible class concepts are defined (209, 460). "Falsity" does
not even exist until the reflective stage of thinking is reached: "since the
first men of the gentile world had the simplicity of children, who are
truthful by nature, the first fables could not feign anything false; they
must therefore have been ... true narrations" (408). "Reflection ... is
the mother of falsehood" (817).
Rational philosophy, however, with its logical concepts of truth and
falsity, is not enough for understanding, because "poetic certainty" is
also required in the application of logic to the world. The certainty of
the human sciences results from our ability as human beings to reconstruct the original imaginative conditions under which human concepts
200
MARY B. HESSE
get established, both in primitive language and in the developing

language of every child. It is not only the primitive who constructs his
world by imagination, but also the philosopher-philologian. As Verene
puts it:
Vico's work is a training in fantasia . ... The grasp of a meaning through fantasia, upon
which [rationalJ metaphysical criticism depends, occurs suddenly, like original speech
itself. The image opens the world .... We must reconstruct the human world not
through concepts and criteria but as something we can practically see ([45J 157-8).
4. METAPHOR AND OTHER TROPES
There are two senses of "metaphor" in Vico. The sense that is in

conformity with normal usage is the transference of linguistic meaning
via conceptual similarities. But for Vico the primary sense relates to the
formation of imaginative genera by the pre-linguistic gathering of a
number of particulars into their ideal unity. This occurs before the
development of articulate language, in perception and in the acts of
signing and gesturing. Two questions remain to be answered. First, why
does Vico use the term "metaphor" for this fundamental process? And
second, how, if at all, does it relate to the conscious use of linguistic
tropes as these have been understood in the traditions of grammar and
rhetoric? These questions take us further into Vico's theory of language, and lead to a consideration of the main problem of this paper,
namely how far does he conceive metaphor as essential to all language
stages from primitive through heroic to rational?
In attempting to revive the "ancient wisdom" in face of "modern"
Newtonian science, Vico stands in the tradition of classical rhetoric (see
particularly [49], [12], [18], [17]). In that tradition rhetoric is logic of
argument in the realm of the probable; it lives in law court and political
arena, and hence it involves the art of persuasion as well as argument. It
exists at the interface of truth and its application to human affairs, and
it is this latter aspect that Vico thinks has been lost in scientific
concentration on merely critical or demonstrative reason. The theory/
practice interface is, however, also where Vico locates the imaginative
genera - it is they that mediate between language and the world, and
make possible the application of linguistic terms to things. The certum
that underlies all reasoning about the human world might be called a
limiting case of rhetorical judgment bridging theory and practice. This
is not a limit in the sense of tending towards certainty in the realm of
VI CO'S HEROIC METAPHOR
201
the natural as against probability in the realm of the human. It is rather

the opposite, since for Vico we can know the natural less certainly than
the human. The certainty we can have in human science is a limit in
that, as I have noted above, the principles of this science are fixed by
the "common sense" of the human race as necessary conceptions for
social survival.
So rhetoric in this sense underlies all argument that has reference to
the world. The logic of Descartes and Port-Royal, which denigrates
rhetoric, is in fact parasitic upon it. Rhetoric, in the classical view,
proceeds by the "art of topics", which Aristotle defines as the way of
reasoning "from opinions that are generally accepted" (Topics, 100a18,
101 b 13), and of finding all the relevant premisses for a given problem
(SN 497). The topoi are common-places (we retain the metaphor of
place) where are located generally accepted beliefs to which appeal has
to be made in the conduct of practical reason and persuasion.23 It is
significant for our argument about the generality of rhetoric as underpinning critical logic, that Vico describes the imaginative genera as
"sensory topoi" and as "the primary operation of the human mind"
(495-7,699; see [45]167-73), for the topoi in rhetoric are analogous
to the accepted premisses of deductive argument.
In the sixteenth and seventeenth centuries rhetoric acquired a
somewhat pejorative image as "mere ornamentation" of speech by
tropes and figures, which should be separated from logic and rational
argument (see [6]104; [2] 285f; [29] ch. 4; [33] ch. 1). This was the
burden of the influential textbook of Peter Ramus, published in
multiple versions between 1543 and the mid-seventeenth century. Here
rhetoric becomes a matter of elocution and pronunciation, and its
classical functions of discovery (invenio) - the finding of premisses
and arguments, are assimilated to syllogistic logic.
In spite of the higher rational value he gives to rhetoric, Vico takes
from Ramus the four-fold distinction of linguistic tropes into metaphor,
metonymy, synecdoche and irony ([32]204). In Scienza Nuova metaphor is described as the trope that is "the most necessary and frequent",
and had its origin in the most primitive use of "likenesses taken from
bodies to signify the operations of abstract minds" (404), when the
poverty of language prevented direct description of generals and
abstracts (34, 237, 407, 816). Metonymy is the trope that puts agent
for act, subject for form and accident, and cause for effect, effect for
cause. This was also necessary for the "first poets" because of the lack
202
MARY B. HESSE
of names for actions, and the inability to abstract qualities from subjects
(406). Vico gives the example of "Hermes" to stand for civil wisdom
(209), and a sad and ugly woman to stand for poverty, age and death.
Synecdoche is the putting of whole for part or part for whole, and is the
way general terms were first named from their particulars or parts, for
example individuals ("Godfrey") to stand for classes ("knighthood"),
"heads" for "men", "harvests" for ''years'' (407). All these tropes were
necessitated by the early poverty of language. Irony, on the other hand,
must be later than the other tropes, since it "could not have begun until
the period of reflection, because it is fashioned of falsehood by dint of a
reflection which wears the mark of truth" (408); it is a deception,
whereas the early fables and their expressions in metaphor, metonymy
and synecdoche were necessarily "true narrations".
Vico goes on to describe how the tropes became figurative: "these
expressions of the first nations later became figurative when, with the
further development of the human mind, words were invented which
signified abstract forms or genera comprising their species or relating
parts with their wholes" (409). Poetic speech becomes prose speech "by
contracting into a single word as into a genus, the parts which poetic
speech had associated" (460). For example, "the blood boils in my
heart" becomes simply stomachos, ira, col/era, anger. 24 Metaphors
become explicit likenesses instead of identities (404). Personifications
of mental faculties become poetic devices instead of the identifying
names of these faculties (402-3). Heraldic symbols have "to be
explained by mottoes, for their meanings are analogical; whereas the
natural heroic emblems were such from lack of mottoes, and spoken
forth in their very muteness" (484).
In these paragraphs the discussion of tropes is almost painfully
orthodox, and one might draw the conclusion that Vico ascribes
metaphor and the rest mainly to the childhood of the race, after which
rational abstraction and reflection render them unnecessary for developed thought. He places a high value on poetic speech as proper and
necessary in its time, but its time is only that of the primitive. It is easy
to mount arguments for this interpretation. There is no doubt that
Vico's theory of history is predominently progressive, not cyclic or
relativist. Poetic language belongs to the earliest ages, and is not
repeated in the same form in the ricorsi after the cycle has once
brought a society through the age of reflection. Metaphors of the body
are not recovered unselfconsciously, and indeed are strictly unimagi-
203
nab Ie, once the power of abstract thought has been developed. The
primitive nature of metaphor in Vico is emphasised, for example, by
Dorfles, who admits its function of conveying meaning, but concludes
that this simply implies that "much of our cognitive power still remains
entangled in the coils of a form of discourse which is predominently
irrational" ([11] 586).
Another argument against the pervasiveness of Vico's metaphor in
all ages of language might be drawn from the theory of the three ages
itself. In "The tropics of history: the deep structure of the New Science",
Hayden White suggests that there is a transition from metaphor to
metonymy in the religious age, from metonymy to synecdoche in the
heroic age, and from synecdoche to irony in the rational or human age
and its ricorsi into decadence and barbarism ([53] 72f). This would
place metaphor, metonymy and synecdoche in the early ages, and irony
along with literal speech in the rational age. But further consideration
indicates that White is not here arguing the primacy of the linguistic
tropes as the style of language in the respective ages. Rather, his
argument is a subtle comparison of the tropes with corresponding
social structures: the rule of the gods goes along with their personification in the divine age, and this is metaphoric; in the heroic or
aristocratic age replacement of "heroes" for their characteristics (subjects for qualities) is metonymic; in the human, democratic, age, the
elevation of individual against class (part for whole) is synecdochic. The
types of language in White's analysis remain the "mute", the "heraldic",
and the "articulate" ([53J 77) as they do consistently in Vico himself.
Vico was not consistent in the way he coupled tropes with the three
ages, and this reinforces the suggestion that he did not place great
weight on any evolution from trope to trope to non-tropical language.
In Scienza Nuova metaphor is not by any means restricted to the poetic
age. Not only is it said "to make up the great body of the language
among all nations" (444), but it is also associated in particular with the
heroic language: "the signs in which the heroes wrote .. . must have
been metaphors, images, similitudes or comparisons, which having
passed into articulate speech, supplied all the resources of poetic
expression" (438). And synecdoche, far from superseding metaphor
and metonymy, is said to develop "into metaphor as particulars were
elevated into universals or parts united ... [to] make up their wholes"
(407). And in the first edition of Scienza Nuova metonymy, synecdoche
and metaphor are mentioned in that order as forming third-age
204
MARY B. HESSE
language ([46] para. 366). It must be concluded that, in Vico's view,

none of the tropes are exclusively associated with particular linguistic
phases.
Vico does, however, believe that in the pre-articulate ages, language
is necessarily wholly metaphorical or tropical. The possibility of such a
language existing at all has often been questioned on the grounds that
tropes depend for their understanding on previously established literal
meanings, since tropes are generally defined in terms of transfer of
meaning from its "proper", literal use to an "improper" context. How
then can "all language be metaphorical" (d. [4] ch. 8) without infinite
regress? The objection is not, however, cogent, since if all language is
indeed metaphorical, new metaphors may be redefined as transfers of
meaning from old metaphors to new contexts, creating new metaphoric
meanings. That this occurs would generally be agreed in the sense that
"literal" language is admitted to be full of "dead" metaphors which have
become essentially literal. To use one of Vico's examples, the term
"character" first meant a "name" (433), then a "definition", then by
extension a particular personality, as in a fable or drama. The question
is, does such a sequence of etymologically related metaphors have to
have a first term which is strictly literal?
Vico's reply to this question would surely be in terms of his primary
imaginative genera. This conception may be compared to that of
Wittgenstein's family resemblances, except that Wittgenstein's "genera"
shift arbitrarily with context and convention, whereas Vico's were
originally fixed by the necessities of human existence. But in both cases
languages are held to arise from the perception of particulars, which are
found to bear varying degrees of similarity or difference to each other.
Bundles of sufficiently similar particulars are classified together and
given a class name by the family resemblances uniting them and
discriminating them from those sufficiently different. There is no
abstract class or universal perceived to be really present in particulars
as Aristotle held, but only a socially accepted judgment that some
particulars belong to classes whose names get established in natural
language.
If some such Wittgensteinian account of the origins of names and
abstract terms is correct, we can reply to the objection that the
metaphorical must presuppose the literal. What is normally called the
"literal" is not something given in the nature of words and their
application to the world, but is a partly conventional, socially estab-
205
lished, assignment of terms to particulars classified by their immediately

perceived resemblances. The mechanism is the same as that of conscious metaphor coined anew by noticing resemblances, but it emerges
from the pre-linguistic human capacity to recognize similarity that can
be shared and issue in an intersubjectively workable language. There is
in this sense no first literal term in a sequence of related metaphoric
meanings - their establishment and subsequent shifts are continuous
from animal-like perceptions and "noises" to the first human use of
signs and speech. What we now call the "literal" and distinguish
pragmatically from the metaphorical, is just that part of language which
is generally received in our society, full of "dead" metaphors as it is, or
better, full of the primary recognitions of imaginative genera which
continue to pervade all language. The difference between Vico's view
and that of a modern linguistic relativist is that for Vico the "true
genera" were initially established by providential ordinance, and subsequent history is a constant process of degeneration and recovery of
these poetic meanings. For the modern relativist, as for Bacon and
Hobbes, if an ideal language is contemplated at all, it is the accumulation of scientific knowledge that restores or discovers the true genera.
As Vico noted with dismay at the end of Scienza Nuova, the epistemology of his contemporaries was putting science in the place of
providence (1111-12).
5. "THE GREAT BODY OF THE LANGUAGE
AMONG ALL NATIONS"
The analysis of imaginative genera is undoubtedly the most fundamental argument for the nature and importance of metaphor in the
poetic and heroic ages. But if we seek to argue that in Vico's view
metaphor is pervasive, necessary and normative for all language, the
objection might still be made that he regards third-age metaphors as
being superseded by explicit figurative speech, and therefore reducible
to the merely persuasive and ornamental. In this concluding section I
shall summarize the arguments Vi co himself gives for the metaphoric
character of early language, in order to consider how far he intended
these arguments to imply the metaphoric character of all language.
Vico has four kinds of argument for metaphor in the first languages.
First, the "poverty of language and the need to explain and be understood" (34, cf. 456) requires "poetic locution" to express and extend
206
MARY B. HESSE
language. Second, there is in particular an absence of names for

qualities and abstract ideas, and hence metaphors from the human body
and the names of human persons are used for these. Thirdly, the way in
which these metaphors are assigned is a pre-linguistic construction of
identity classes - Achilles is valour, boiling blood is anger. Hence arise
the imaginative genera, which are cross-cultural in virtue of the
commonality of the human condition. Fourthly, imaginative genera
carry normative force, and thus combine in early language the functions
of logic and rhetoric in the rational age. Fables, heraldic devices,
metaphors, etc. are descriptive and persuasive, but the persuasive
element is not just ornamental and subjective, it is an integral part of
the "truth" of the common presuppositions of human society.
These arguments about origins do not of course dictate the character
of later language. And yet it seems that Vico did intend to use three of
the four arguments to assert the perennially tropical character of
language as such. Only the second argument from the need for
metaphors of body fails to apply to later language, even in periods of
the ricorsi into barbarism. In looking for arguments about later
language in Scienza Nuova, one finds very little analysis of the third age
as compared with the lengthy accounts of the first two ages. 25 Vico
does, however, comment on the essential poverty of all language: "the
language of men [third-age], almost entirely articulate and only very
slightly mute, there being no vulgar language so copious that there are
not more things than it has words for" (446, ct. 581). But the argument
from the poverty of language as compared with the multiplicity of
things is more explicit in earlier works. In the Institutiones orationiae of
1711, "metaphoric language ... is ordinary language inasmuch as it is
born from the poverty of speech" and ''what are called synonyms do
not exist in any language" (quoted in [17J 35). In describing the theory
of tropes in the first edition of Scienza Nuova, Cantelli comments
"Allegory and metaphor are not bold ends in themselves; they are,
rather, the means which, through continual transposition of images,
permit man's mind to adjust to the infinite multiplicity of meanings in a
living universe" ([8J 62). And Mooney relates Vico's view to that of his
mentor Cicero: "tropes are necessary, since things are more numerous
than words" ([32] 202. Cf. [19]170, 180f).
The argument is that language needs metaphor because it always falls
short of the multiplicity of things, and needs to be extended to new
situations. Unless we regard language as a purely static and formal
207
system, which Vico certainly did not, this is an argument that applies to
all natural language as such. The extensibility of language also implies
the use of what Vico calls "imagination" in the grasp of new classifications of things and the naming of new concepts. So the mechanism of
imaginative genera is also a common feature of all living language. For
Vico it is the way all children learn language (186, 209, 376), and it is
pervasive in etymological development. Even more important is the fact
that Vico's theory of induction, which certainly applies to the third age,
emphasises particulars against Aristotelian universals. If we cannot rely
on universals being present identically in different particulars for
purposes of inference then we cannot rely on particulars being named
by univocal general terms. But this is just where Vico's imaginative use
of metaphor enters his theory of language.
The argument for the existence of tropes alongside logic in rational
language is part of the more general argument for the continuing
rational function of rhetoric in Vico's philosophy of the third age. On
this there is some disagreement among the commentators, the "rationalists" arguing that topics and tropes are superseded by logic, and the
"imaginists" arguing that rhetoric retains a rational, normative function,
which is complementary to critical reason. We have seen an example of
this debate in the exchanges between Pompa and Verene.
Support for the continuing rational function of rhetoric comes,
however, from the very style and structure of Scienza Nuova itself. As a
"science" it must be "universal and eternal", as Vico states that every
science must be (163, 332), but does this imply that it must eschew
metaphor and rhetoric? 26 Vico does not explicitly pursue this question,
but there are reasons for thinking that the New Science cannot be
contained in a purely "third-age" concept of reason. He can quite
consistently claim that the principles of the New Science are "universal
and eternal", without implying that such universal truth is identical with
what the Baconians and Cartesians take to be the product of critical
and empirical reason. Indeed the whole thrust of Vico's corpus suggests
that to get universal truth about human affairs, we must adopt a
philological and historical, and therefore rhetorical, method as well as a
Baconian one. In other words, it is not incorrect to regard Vico as the
father of a hermeneutic method that defines a new sort of cognitive
reason.
The style of Scienza Nuova is full of repetitions, and is certainly far
from being clear and distinct, a fact about which commentators have
208
MARY B. HESSE
frequently complained. But this may not simply be due to poverty of

argument. It has been suggested by Frankel that the whole structure of
Scienza Nuova as a philosophical work, is an intentional reversion from
Enlightenment rationality to earlier Renaissance rhetoric: "He may have
composed his work in the form of an enlarging spiral of reiterations
designed to be a model in writing of his theory of cyclic history and of
the separate but analogous developments of single nations, each one
growing through the same stages" ([13]49-50). Frankel argues for this
conclusion by means of a detailed analysis of the frontispiece of Scienza
Nuova (the "Dipintura"), which acts as a "mute" symbol for the three
ages of history with their specific symbols and types of reason and
language, and the relations between these.
If this interpretation is correct, Vico's view is deeply self-reflexive
and contains within its own organization the principle that symbol and
metaphor are necessarily, and ought consciously to be, pervasive in all
stages of language. As Vico puts it: "poetic speech ... continued for a
long time into the historical period, much as great and rapid rivers continue far into the sea, keeping sweet the waters borne on by the force of
their flow" (412). The "rational" age leads to the over-rationalization
of argument, which is what motivates Vico himself to revive rhetoric
and its tropes, in order to restore the balance between practical
wisdom, and logic and natural science.27 Scienza Nuova itself is an
exercise in "sweetening the waters" within the ricorsi into the new
barbarism that Vico believes to be threatening in his own time. In the
light of such pervasive concern with rhetoric it cannot be supposed that
Vico regarded the perennial use of metaphor, as the vehicle of rhetoric,
as merely ornamentative or logically improper.
NOTES
I English translations of Vieo's principal works appear in ]47], ]49], ]50], ]52]. ]50] is a
definitive edition and translation by T. G. Bergin and M. H. Fisch of the third edition
(1744) of the Scienza Nuova. References to this translation in the text will be given to
paragraph numbers in brackets, preceded where necessary by the abbreviation SN. The
principal recent commentaries on Vieo in English are ]6], ]7], ]34], ]41], ]42], ]43], ]44],
]45].
An English translation of part of this work appears in ]52]. It will be abbreviated in

the text as De Sapientia, and page numbers will be to this translation except where
noted otherwise .
.1 For discussions of Bacon's theory of the myths, see ]37] and ]21]. On Bacon and
Vico on myth, see ]9].
2
209
For a recent philosophical debate with references, see [36J and [25J.
See SN(163), and De nostri temporis studiorum ratione (1709), pp. 33,41. This
work is translated in [49J, and, in part, in [52J. Page references will be to the latter
edition unless noted otherwise. Vico is most outspoken about his contemporary science
in his Autobiografia (1728), translated in [47J. For example: Boyle's experimental
physics is profitable for "medicine and spagyric ... [butJ contributed nothing to the
philosophy of man" (p. 128); Descartes "craftily feigned" his studies "to exalt his own
philosophy and mathematics and degrade all the other studies included in divine and
human erudition" (p. 113); and Bacon's De sapientia veterum is "more ingenious and
learned than true" (p. 148).
6 He knew of the Alphabet from Bacon's De augmentis scientiarum (see [47J 139). In
the Baconian tradition "language" is used in an extended sense to mean "naming" or
"representation", but with little attention to syntax (see [30J 359).
7 For accounts of this movement see 11 L[22J, [391, 1401.
S For Vico and Hobbes see particularly 151, 1141, 1301, [381.
9 For Hobbes' debt to Galileo see the Introduction by I. C. Hungerland and G. R. Vick
to [28125.
10 For Vico on Hebrew and Christian religion see, for example, [451185.
liOn the verum-factum principle and its relation to "certainty" (certum) see De
sapientia, [52]50-6; 134] ch. 7; and ]451 ch. 2. Vico does repudiate his earlier view of
geometry, which ceases to be the model of a "man-made" institution like those of
society. and becomes mere abstraction. having "less reality" than human institutions
(SN 349).
11 De nostri temp oris, 152]40.
I)
De nostri temp oris, 152]55.
14 This first edition of Scienza Nuova was almost totally rewritten in the 1744 edition.
Extracts are translated in 152] 81-156.
For Vi co's theory of providence see, for example, 17] 64f., and [341 ch. 5.
1(, 152163. Compare Vico's Platonic "ladder of forms" (l52J 62) with Bacon's "ladder of
axioms" described, for example, in ]211146.
17 Cf. Scienza Nuova (424): "the order of human ideas is to observe the similarities of
things first to express oneself and later for purposes of proof. Proof, in tum, is first by
example, for which a single likeness suffices, and finally by induction, for which more is
required. Socrates ... introduced by induction the dialectic which Aristotle later
perfected with the syllogism, which cannot proceed without a universal".
On the other hand Aristotle "deduces particulars from their universals rather than
uniting particulars to obtain universals ... land this did not yieldJ anything more
notable to the advantage of the human race. Hence with great reason Bacon ...
proposes, commends, and illustrates the inductive method in his Organum, and is still
followed by the English with great profit in experimental philosophy" (499).
IX De sapientia, ch. 4.2, in 148] 269; 112] 410; 1451 49f.
I" Winch borrows the thrcc principles in his "Understanding a primitive society"
(154J 107). It is the loss of the sensus communis by the time of Dilthey and his
successors that makes the epistemology of what is "made by man" more problematic
than the epistemology of natural science, cf.115J 196f.
20 Comparing his view with Pf':,lpa's in 1341 and 1351, Verene says: "The central
difference, and I believe it to P', c' very important one, is that his view is not grounded
I,
210
MARY B. HESSE
in Vico's notion of fantasia. Pompa does not explore the sense of fantasia present in the
origin of human mentality, upon which the New Science rests .... The difference
remains that Pompa's view approaches Vieo's thought through cognition and the
philosophical argument and the present study comes to Vieo through the basis of his
thought in rhetoric and the image" ([45] 155n).
21 Contrary to Vico's remarks on induction, above p. 194. I have discussed these issues
in [23], [24], and [4] ch. 8.
22 See [52]60f., and 145]74. The socially normative character of Vieo's use of "truth"
is emphasised by [3] ch. 12, [15]2lf, and 120] 75f. In [5]614, Barnouw objects to their
identification of Vico's rhetoric with classical practieal reason, or phronesis, where this
is assumed to lead to a dualism of reason in the "rational" age, arguing that Vieo has
rhetorie developing into universal rational ideas in the third age. How far Vieo retains a
dualism of logic and rhetoric in the third age is a question I return to in section 5.
23 Compare the account in [55] chs. 2, 10.
24 "Mi bolle il sangue nel cuore": the humours seem to have got somewhat mixed in
this idiom.
25 Just as there is little analysis of the verum-factum principle or of methodology as
compared with his earlier works.
26 This question is given somewhat different answers by McMullin (131] 60) and
Mooney 132] 208.
27 Vico complains about the "overfine" intelligence of immature intellects when they
pass prematurely from the sciences to metaphysics (159). And in an Oration of 1732
he exhorts the students of Naples to retain the "heroie mind" (] 51 D.
REFERENCES
1. Aarsleff, H., From Locke to Saussure (London: Athlone Press, 1982).
2. Abrams, M. H., The Mirror and the Lamp (Oxford: Oxford Univ. Press, 1953).
3. Apel, K-O., Die Idee der Sprache in der Tradition des Humanismus von Dante bis
Vico (Archiv for Begriffsgeschichte, 8) (Bonn: Bouvier, 1963).
4. Arbib, M. and M. Hesse, The Construction of Reality (Cambridge: Cambridge
Univ. Press, 1987).
5. Barnouw, J., 'Vieo and the continuity of science, the relation of his epistemology to
Bacon and Hobbes', Isis 71 (1980),609-20.
6. Berlin, I., Vico and Herder (London: Hogarth Press, 1976).
7. Burke, P., Vico (Oxford: Oxford Univ. Press, 1985).
8. Cantelli, G., 'Myth and language in Vico', in 142]47-63.
9. De Mas, E., 'On the new method of a new science; A study of Giambattista Vico',
Journal of the History of Ideas 32 (1971), 85-94.
10. De Mauro, T., 'Giambattista Vieo: From rhetorie to linguistie historicism', in
141 ]279-95.
11. Dorfies, G., 'Myth and metaphor in Vico and in contemporary aestheties', in
141]577-90.
12. Fisch, M. R, 'Vieo and pragmatism', in [41]401-24.
13. Frankel, M., 'The "Dipintura" and the structure of Vieo's New Science as a mirror
of the world', in 144]43-51.
VI CO'S HEROIC METAPHOR
211
14. Funkenstein, A., 'Natural science and social theory: Hobbes, Spinoza and Vico', in
142]187-212.
15. Gadamer, H.-G., Truth and Method (London: Sheed and Ward, 1975).
16. Gardiner, P., 'Vico', in Encyclopedia of Philosophy, 8 vols, ed. P. Edwards
(London: Collier-Macmillan 1967) 8, 247-51.
17. Giuliani, A., 'Vico's rhetorical philosophy and the new rhetoric', in 142] 31-46.
18. Grassi, E., 'Critical philosophy or topical philosophyT, in 1411 39-50.
19. Grassi, E., The priority of common sense and imagination: Vi co's philosophical
relevance today', in 14311,163-90.
20. Habermas, J., Theory and Practice (London: Heinemann, 1974).
21. Hesse, M., 'Francis Bacon', in A Critical History of Western Philosophy, ed. D. 1.
O'Connor (London: Collier-Macmillan, 1964), 141-52.
22. Hesse, M., 'Hooke's philosophical algebra', Isis 57 (1966), 67-83.
23. Hesse, M., The Structure of Scientific Inference (London: Macmillan, 1974).
24. Hesse, M., The cognitive claims of metaphor', in Metaphor and Religion, 2 vols,
ed. J. P. van Noppen (Brussels: van Noppen, 1984) 2, 27-45.
25. Hesse, M., Tropical talk', Proceedings of the Aristotelian Society Supplementary
Volume 61 (1987),297-311.
26. Hobbes, T., English Works, 11 vols, ed. W. Molesworth (London, 1839).
27. Hobbes, T., Leviathan, ed. M. Oakeshott (Oxford: Blackwell, 1946).
28. Hobbes, T., Computatio sive Logica, eds. I. C. Hungerland and G. R. Vick (New
York: Abaris Books, 1981).
29. Howell, W. S., Logic and Rhetoric in England, 1500-1700 (Princeton: Princeton
Univ. Press, 1956).
30. Land, S. K., The account of language in Vico's Scienza Nuova', Philological
Quarterly 55 (1976), 354-72.
31. McMullin, E., 'Vico's theory of science', in 143]1, 60-90.
32. Mooney, M., The primacy of language in Vico', in 14311,191-210.
33. Padley, G. A., Grammatical Theory in Western Europe, 1500-1700 (Cambridge:
34. Pompa, L., Vico, A Study of the "New Science" (Cambridge: Cambridge Univ.
Press, 1975).
35. Pompa, L., 'Imagination in Vico', in 144]162-70.
36. Rorty, R., 'Unfamiliar noises: Hesse and Davidson on metaphor', Proceedings of
the Aristotelian Society Supplementary Volume 61 (1987), 283-96.
37. Rossi, P., Francis Bacon: From Magic to Science (London: Routledge and Kegan
Paul, 1968).
38. Rossi, P., The Dark Abyss of Time: The History of the Earth and the History of
Nations from Hooke to Vico (Chicago: Chicago Univ. Press, 1984).
39. Schapiro, B. J., John Wilkins, 1614-/672 (Berkeley and Los Angeles: Univ. of
California Press, 1969).
40. Slaughter, M. M., Universal Languages and Scientific Taxonomy (Cambridge:
41. Tagliacozzo, G. and H. V. White (eds.), Giambattista Vico, An International
Symposium (Baltimore: Johns Hopkins Press, 1969).
42. Tagliacozzo, G. and D. P. Verene (eds.), Giambattista Vico's Science of Humanity
(Baltimore: Johns Hopkins Press, 1976).
212
MARY B. HESSE
43. Tagliacozzo, G., M. Mooney, and D. P. Verene (eds.), Vico and Contemporary
Thought, 2 vols (Atlantic Highlands NJ: Humanities Press, 1979).
44. Tagliacozzo, G. (ed.), Vico Past and Present (Atlantic Highlands NJ: Humanities
Press, 1981).
45. Verene, D. P., Vico's Science of Imagination (Ithaca: Cornell Univ. Press, 1981).
46. Vico, G., La Scienza Nuova Prima, ed. F. Nicolini (Bari: Gius. Laterza, 1931).
47. Vico, G., The Autobiography of Giambattista Vico, eds. M. H. Fisch and T. G.
Bergin (Ithaca: Cornell Univ. Press, 1944).
48. Vico, G., Opere, ed. F. Nicolini (Milano: Ricciardi, 1953).
49. Vico, G., On the Study Methods of our Time, ed. E. Gianturco (Indianapolis:
Bobbs-Merrill, 1965).
50. Vico, G., The New Science of Giambattista Vico, revised translation of the third
edition (1744), T. G. Bergin and M. H. Fisch (Ithaca: Cornell Univ. Press, 1968).
51. Vico, G., 'On the heroic mind', in 1431 2,228-45.
52. Vico, G., Vico Selected Writings, ed. L. Pompa (Cambridge: Cambridge Univ.
Press, 1982).
53. White, H. V., The tropics of history: The deep structure of the New Science', in
142165-85.
54. Winch, P., 'Understanding a primitive society', in Rationality, ed. B. R. Wilson
(Oxford: Blackwell, 1970), 78-111.
55. Yates, F. A., The Art of Memory (London: Routledge and Kegan Paul, 1966).
PETER M. HARMAN
DYNAMICS AND INTELLIGIBILITY:

BERNOULLI AND MACLAURIN
Abstract. This essay discusses the role of metaphysical argument in the vis viva
controversy. The debate between Johann Bernoulli and Colin MacLaurin over the vis
viva concept and Leibnizian dynamics is reviewed. Special emphasis is placed on
Bernoulli's appeal to the Leibnizian metaphysical principles of causality and continuity
in supporting his theory of motion, and to the special status he accords the principle of
the conservation of vis viva in dynamics. MacLaurin's critique of Bernoulli is reviewed,
with special emphasis on his discussion of the law of continuity, and his defence of
Newton's concept of fluxions and his disparagement of the status of infinitesimal
quantities in Leibniz's mathematics and Bernoulli's dynamics. The physical and metaphysical arguments he deploys in criticising the principle of the conservation of vis viva
are discussed, his approach being essentially Newtonian in inspiration.
1. INTRODUCTION
All natural philosophers who wished to proceed mathematically in their work had
therefore always (though unknown to themselves) made use of metaphysical principles
and were obliged to make use of them, even though they otherwise solemnly protested
against any claim of metaphysics on their science.... (Kant, Metaphysical Foundations
of Natural Science ([13[ 4,472.
The discussion of the metaphysical components of scientific theories

has been a guiding theme of Gerd Buchdahl's work. In particular, he
has discussed the role of regulative maxims such as the law of causality
and criteria of analogy and continuity in the articulation of Newton's
concept of gravitation, and analysed the attempts to justify the intelligibility of gravity by explications of the meaning of the concepts of
matter and force ([41, [5]). BuchdaW has emphasised that these metaphysical arguments were intended to supplement the physical content
of the theory of gravity; by emphasising the metaphysical components
of scientific theories he has stressed the interaction between the
physical and metaphysical components in the articulation of scientific
theories. BuchdaW's studies of the debates over the conceptual status of
concepts of force and gravity, from Newton to Kant, have amply
illuminated his treatment of this philosophical theme, which is deeply
Kantian in inspiration.
213
214
PETER M. HARMAN
Following Buchdahl, I have myself argued that reference to metaphysical foundations is constitutive of the conceptual development of
classical physics ([7); [10)); and in particular that debates over the status
of the concepts of force and inertia were integral to reception of
Newton's physics ([7) 8-80; [8); [9)). In this essay I will further
illustrate this theme by reviewing the debate between Johann Bernoulli
and Colin MacLaurin over the vis viva concept and the Leibnizian law
of continuity (see also [11)).
The correspondence between Leibniz and Samuel Clarke in 171516 had given public expression to the disagreement between Newton
and Leibniz about the status of the concept of gravitational attraction,
and also reawakened Bernoulli's interest in Leibnizian dynamics. The
publication of Leibniz's "Specimen dynamicum" (1695) had led to a
flourishing correspondence with Bernoulli, who became deeply involved
in Leibniz's priority quarrel with Newton over the invention of the
calculus, to which the correspondence between Leibniz and Clarke was
a tailpiece (6)). In 1716 Leibniz drew Bernoulli's attention to his
debate with Clarke and Newton, whom he considered to be his real
adversary lurking behind the scenes. Leibniz criticised Newton's doctrine
of the "spontaneous diminution of active forces and final cessation in
the world", pointing out to Bernoulli that by the Leibnizian principle of
the conservation of vis viva ("living force") the "same quantity of forces
is always preserved" (14) 3, 964). Leibniz thus sees the principle of the
conservation of vis viva as refuting Newton's claim, in Query 23 of the
Latin edition of the Opticks (1706), that because motion is constantly
being dissipated, for example in the collision of inelastic or partially
elastic bodies, it was apparent that "some other principle is necessary
for conserving the motion": "the variety of motion which we find in the
world is always decreasing, (and hence) there is a necessity of conserving and recruiting it by active principles" ([18) 397-401). While
active principles were laws of nature, they were also conceived as the
mediating agents by which God conserved motion and gravity in the
cosmos. Clarke therefore rejected Leibniz's concept of the conservation
of vis viva on theological grounds, maintaining that to deny (as did
Leibniz) that "every action is the giving of a new force to the thing
acted upon" would be to suppose that God is "quite excluded from
the government of the natural world". Clarke affirms the Newtonian
doctrine that "there must be a continual increase and decrease of the
whole quantity of motion in the universe" which had the consequence
DYNAMICS AND INTELLIGIBILITY
215
that "every thing be not mere absolute mechanism" ([15]110). In his

reply to Leibniz, Bernoulli however expresses agreement with the
Leibnizian view of the sufficiency (for purposes of the scientific
explanation of phenomena) of mechanical laws, arguing that "no force
is destroyed, without giving rise to an equivalent effect"; the "effect is
nothing other than the transformed force", and hence "it is necessary
that the same quantity of force be preserved" ([14]3, 966). Leibniz and
Bernoulli reject any claim that there are certain natural phenomena gravity or the collision of inelastic bodies - that "cannot be explained
mechanically" .1
These physical and metaphysical arguments are fully developed in
Bernoulli's memoir "Discours sur les loix de la communication du
mouvement" (1727), submitted for the 1724 prize of the Paris
Academie Royale des Sciences on the laws of the collision of hard
bodies. Bernoulli's entry was disqualified on the grounds that he had
denied that hardness or inflexibility was a fundamental property of
bodies, and the prize was won by the young Scottish mathematician and
supporter of Newtonian physics and mathematics, Colin MacLaurin.
Bernoulli's memoir was however subsequently published, and collected
with other papers submitted to the Academie, and provided a systematic
account of the law of the conservation of vis viva. 2 MacLaurin consequently took Bernoulli's memoir as a substantive statement of
Leibnizian dynamics in his Treatise of Fluxions (1742) and his Account
of Sir Isaac Newton's Philosophical Discoveries (1748), works which
established his reputation as the leading British defender and populariser
of Newton's physics and mathematics. I will begin with an account of
Bernoulli's "Discours sur ... mouvement", and will then tum to an
examination of MacLaurin's critique of this work, with an especial
emphasis on Bernoulli's appeal to the Leibnizian metaphysical axioms
of causality and continuity, and MacLaurin's response.
2. BERNOULLI'S THEORY OF MOTION
Bernoulli presents his memoir "Discours sur les loix de la communication du mouvement" as a systematic elaboration of the Leibnizian
science of dynamics, remarking that he was the first natural philosopher
for 28 years (since the publication of "Specimen dynamicum" in 1695)
to espouse Leibnizian dynamical principles. Based on Leibniz's metaphysical principles of causality and continuity, the elasticity of matter,
216
PETER M. HARMAN
and the conservation of vis viva, Bernoulli's work provides a detailed

presentation of Leibnizian dynamics, bringing together in a systematic
and coherent fashion themes which Leibniz himself never fully developed. Most fundamentally, Bernoulli adheres to the fundamental
presupposition of Leibnizian dynamics, the law of continuity, that "no
change occurs through a leap". In correspondence with Bernoulli in
1698 Leibniz had stressed that the law of continuity had profound
implications for his dynamics: the mathematical law of continuity had
its counterpart in the physical world (14]3, 544).
In his "Disc ours sur . . . mouvement" Bernoulli follows Leibniz in
supposing that motion is produced in "infinitely small degrees" in
accordance with the law of continuity; therefore vis viva is a "finite and
determinate quantity" and is produced as a result of an infinite number
of infinitesimal impulses of vis mortua ("dead force"), and is "inherent
in a body when it is in uniform motion" ([2] 3, 23, 36, 39). The
Leibnizian law of continuity implies the harmony between the order of
nature and the mathematical relation between infinitesimal and finite
quantities. Bernoulli notes the "perfect conformity which prevails
between the laws of nature and those of geometry; a conformity which
is observed so constantly and in all circumstances, that it appears that
nature has consulted geometry in establishing the laws of motion" ([2] 3,
58). The "law of continuity ... natura non operatur per saltum" is an
"immutable and perpetual" general law of nature, which explains the
generation of motion in terms of the transition of a body from a state of
rest to a state of motion by a transition "through all the insensible
motions which lead from one to the other". Moreover the law of
continuity establishes the "necessary connection" between the states of
motion and rest, rendering the order of nature intelligible ([2] 3, 9-10).
Bernoulli applies this concept of motion to the analysis of the
collision of bodies. He follows Leibniz in maintaining that the assumption of the "absolute hardness" of the elementary corpuscles of matter
by the partisans of the atomic theory of matter (notably Newton) was
"absolutely impossible". It was of course his denial of this assumption,
that led to the disqualification of his essay from the 1724 Paris
competition. Bernoulli however proceeds to give an account of hardness, as a property of matter, in Leibnizian terms, rejecting the
Newtonian view that hardness (along with extension, impenetrability,
mobility and inertia) was to be construed as an essential or fundamental
property of matter. Bernoulli supposes that matter is inherently elastic,
217
and that the hardness of a body is not to be conceived as a fundamental

property of matter but as a relational condition. Matter therefore is
fundamentally elastic and the hardness of bodies is explained in terms
of the relative stiffness and rigidity of inherently elastic matter. He
illustrates this concept of the essential elasticity of matter by representing bodies as elastic springs. On collision motion is transferred
between bodies qua elastic springs by infinitely small degrees. Thus
motion is transferred from one body to another "successively and by
elements", in consonance with the Leibnizian concept of the transference of motion in accordance with the law of continuity, that "no
change occurs through a leap" ([2] 3, 9-10, 15-16).
Bernoulli develops this model of matter as an elastic spring to give a
physical and mathematical representation of the relation between vis
mortua and vis viva. He follows Leibniz in supposing that the force
manifested when a compressed elastic spring begins to expand is an
example of vis mortua, and argues that as a spring expands vis viva is
generated. While vis mortua is proportional to the infinitesimal velocity
as the spring begins to expand, the vis viva is proportional to a finite or
"actual" velocity. Bernoulli develops the argument to demonstrate that
the measure of vis viva is mv 2. Representing the "pressure" or vis
mortua in the spring, corresponding to the "endeavour" of the spring to
expand, by p, this produces an increment of velocity d v in an increment
of time dt by the expansion of the spring through a space dx. From "the
well-known law of acceleration" dv = pdt (a result first stated by
Varignon and subsequently used by Bernoulli himself), he obtains the
relation vd v = pdx, and this "by integration gives 1- vv = f pdx". If the
action of two springs of unequal size pressing against two balls of equal
mass are compared, then the ratio between the vires vivae will depend
on the ratio between the square of the velocities acquired by each ball;
and "if the balls are of unequal mass, it is apparent that their vires vivae
are as the product of the masses by the square of the velocities" ([2] 3,
45-7).
Bernoulli gives metaphysical support for his theory of motion by an
appeal to the Leibnizian law of causality. The concept of vis viva is the
true measure of force in nature because vis viva is "equivalent to that
part of the cause that is consumed in producing it". Because "the whole
efficient cause is equivalent to the full effect which is produced", the
conservation of vis viva follows from "the equality between the effect
and the efficient cause" (12] 3, 38). The principle of the conservation of
218
PETER M. HARMAN
vis viva is a fundamental natural law expressing the order and harmony
of nature: if the quantity of vis viva, the "single source of the continuation of motion in the universe", were not conserved then "all nature
would fall into disorder". He explicitly contrasts this Leibnizian expression of the self-sufficiency of nature, as grounded on the preservation
and conservation of vis viva in mechanical processes, with the Newtonian
doctrine of the diminution of activity in the collisions of bodies ([2J 3,
58). Bernoulli's theory of motion is justified by the appeal to the
metaphysical principles of causality and continuity, and to the harmony
between mathematics and nature as expressed by the generation of vis
viva from vis mortua.
Bernoulli made it plain that his theory of motion and concept of vis
viva was essentially Leibnizian in inspiration. He contrasts his own
statement of the law of the conservation of vis viva with Huygens'
formulation of the conservation of the quantity mv 2 in collisions. For
Huygens the quantity mv 2 was merely a number, without any fundamental dynamical significance. In his Horologium Oscillatorium (1673)
Huygens demonstrates that the distance through which the centre of
gravity of a body or system of bodies will fall under the action of
gravity is equal to the height to which it can ascend as a result of the
velocity acquired in the fall; there is an equality between actual descent
and potential ascent of the centre of gravity of a system of moving
bodies. From Galileo's law of falling bodies the descent of each body is
proportional to the square of its velocity; the quantity mv 2 is conserved
in the descent and ascent of the centre of gravity of a system of bodies
([12J 18, 147,247,255). Bernoulli remarks that Huygens had regarded
his theorem of the conservation of mv 2 as a mathematical proposition,
a mere formula, without realizing its fundamental status as a general
dynamical natural law. Indicating his own espousal of Leibnizian
dynamics, he declares that "without recourse to nature and first
principles, the most important theorems degenerate into simple speculations" ([2J 3, 58).
While Bernoulli presents his "Disc ours sur . . . mouvement" as a
treatise on the motions and collisions of bodies which was essentially
Leibnizian in inspiration, there are important differences between his
argument and Leibniz's own presentation in "Specimen dynamicum".
Bernoulli's discussion is confined to the physics of Leibnizian derivative
forces. The Leibnizian theory of substance is basic to Leibniz's own
science of dynamics; derivative forces are phenomenal analogues of the
219
pnmitIve forces which define the nature of substances, while being

conceived as part of the order of nature and expressing the mechanical
laws of nature. This plays no part in Bernoulli's exposition of the theory
of motion. Bernoulli's differences with Leibniz over the status of
infinitesimals are also important. Writing to Bernoulli in 1698 Leibniz
had declared that "infinitesimals are imaginary [quantities]" and did "not
exist in nature"; the representation of vis viva as generated from
infinitesimal impulses of vis mortua is a mathematical model ([14]3,
499,551). In "Specimen dynamicum" he declares that the infinitesimals
posited in his dynamics are "mathematical entities" which were not
"really found in nature as such" but were "the means of making
accurate calculations" ([14]6, 238). In his correspondence with Leibniz
Bernoulli had diverged from Leibniz's view that infinitesimals are
imaginary, declaring that "the infinitesimal exists" ([14] 3, 563). In a
later essay "De vera notione virium vivarum" (1735) he declares that he
conceives vis viva as a "real and substantial entity", as a fundamental
"absolute" quantity, an existent quantity which is inherent in substances
([2] 3, 239-41). This concept of vis viva contrasts with Leibniz's view
of vis viva as a derivative force which is a phenomenal analogue of the
primitive forces which define substances. In Bernoulli's dynamics it is
implied that both vis mortua and vis viva have the status of real,
existent quantities though their dynamical status differs: while vis
mortua is transient vis viva is conserved and undiminished.
Despite these differences with Leibniz, Bernoulli's work can be seen
as Leibnizian in its physics and in its appeal to metaphysical axioms of
continuity and causality. Bernoulli certainly intended that this should be
so, but of course he was too powerful a personality and too creative a
mathematician to be merely the follower of another's metaphysical
system, a system moreover which he apparently found impenetrably
obscure ([14]3,545,547,556).
3. MACLAURIN'S "NEWTONIAN" CRITIQUE
In his Treatise of Fluxions (1742) MacLaurin develops a presentation

of Newton's concept of fluxions by appealing to the method of first and
last ratios of nascent and vanishing quantities approaching their limits.
The book was written as a response to George Berkeley's The Analyst
(1734), and MacLaurin took the opportunity to subject the concept of
infinitesimals to a characteristically Newtonian critique. MacLaurin's
220
PETER M. HARMAN
An Account of Sir Isaac Newton's Philosophical Discoveries (1748) was

apparently begun after Newton's death in 1727, but only published
posthumously, and is a wide-ranging discussion of Newtonian natural
philosophy, and includes a critique of Leibnizian dynamics. Bernoulli's
theory of motion is discussed in both Fluxions and Newton, where
Bernoulli is taken as Leibniz's main protagonist, "the most learned and
skilful advocate for the new [Leibnizian] doctrine" ([17]118), whose
elements are described in the following terms in Fluxions:
From geometry the infinites and infinitesimals passed into philosophy, carrying with
them the obscurity and perplexity that cannot fail to accompany them. An actual
division, as well as a divisibility of matter in infinitum, is admitted by some. Fluids are
imagined consisting of infinitely small particles, which are composed themselves of
others infinitely less; and this subdivision is supposed to be continued without end.
Vortices are supposed, for solving the phaenomena of nature, of indefinite or infinite
degrees, in imitation of the infinitesimals in geometry; that, when any higher order is
found insufficient for this purpose, or attended with an insuperable difficulty, a lower
order may preserve so favourite a scheme. Nature is confined in her operations to act
by infinitely small steps. Bodies of a perfect hardness are rejected, and the old doctrine
of atoms treated as imaginary, because in their actions and collisions they might pass at
once from motion to rest, or from rest to motion, in violation of this law. Thus the
doctrine of infinites is interwoven with our speCUlations in geometry and nature.
Suppositions, that were proposed at first diffidently, as of use for discovering new
theorems in this science with the greater facility, and were suffered only on that
account, have been indulged, till it has become crowded with objects of an abstruse
nature, which tend to perplex it and the other sciences that have a dependence upon it
([1611,39).
This powerful passage covers the main elements of Leibniz's physics,

planetary theory and mathematics. Here I will omit all discussion of
Leibniz's "Tentamen de motuum coelestium causis" (1689), where he
applied the general principles of his differential calculus to the problem
of planetary motion, where a rotating vortex provides a mechanism for
the necessary forces ([1]125-51). MacLaurin grasps the close link
between Leibnizian dynamics and the Leibnizian metaphysical axiom of
continuity, and in Newton he discusses this point with explicit reference
to Bernoulli's "Discours sur ... mouvement":
We have another instance of the art by which they support their schemes, in the
pretended demonstration they give against the possibility of atoms, or of any perfectly
hard and inflexible bodies. According to what they call the law of continuity, all
changes in nature are produced by insensible and infinitely small degrees; so that no
body can, in any case, pass from motion to rest, or from rest to motion, without passing
221
through all possible intermediate degrees of motion; from which they conclude that
atoms, or any perfectly hard bodies, are impossible; because if two of them should meet
with equal motions, in contrary directions, they would necessarily stop at once, in
violation of the law of continuity ([17]87-8).
In response MacLaurin seeks to turn the notion of "continuity"

against Leibniz and Bernoulli. He observes that his opponents suppose
a distinction between two kinds of "forces": "when the velocity is finite,
how small soever it may be, the force is measured by the square of the
velocity; but when the velocity is infinitely little (as it is, according to
the favourers of the new opinion) in consequence of the first impulse of
the power that generates the motion, the force is simply as the velocity
[thus: as dv]; and we cannot but observe, that this sudden change of
the law does not appear to be consistent with the favourite principle
of continuity, so zealously maintained by the same philosophers"
([17]132). The supposition of two kinds of "forces", which Bernoulli
himself had conceived to be "heterogeneous" ([2]3, 37) is therefore in
contradiction to the concept of continuity; there must be a continuity
and hence identity between the "forces" at the commencement of
motion and when motion is established. Similar critiques of Leibniz and
Bernoulli were made by Euler in his paper "De la force de percussion
et de sa veritable mesure" (1745), and by Kant in his Gedanken von der
wahren Schiitzung der lebendigen Kriifte (1747) ([7] 63-4).
In section 1 of the Principia's first book Newton justifies his
infinitesimal arguments by the method of prime and ultimate ratios of
nascent and vanishing finite quantities approaching their limits. He
claims that he does not suppose indivisible magnitudes but finite limit
ratios: "by the last ratio of vanishing quantities you must understand
not the ratio of quantities before they vanish, not that afterwards, but
that with which they vanish". He claims that the determination of this
limit is a "geometrical problem" ([19]1, 87-8). These claims shape
MacLaurin's defense of Newton and critique of the Leibnizian calculus.
He notes that the proponents of the Leibnizian calculus differed among
themselves about the status of infinitesimals. While Leibniz supposed
infinitesimals to be "no more than fictions", others (such as Johann
Bernoulli) "place them on a level with finite quantities and endeavour
to demonstrate their reality". He maintains that these views are
confused: mathematicians have "involved themselves in the mazes of
infinity". While he concedes that "the computations in (Newton's]
method are the same as in the [Leibnizian] method of infinitesimals", he
222
PETER M. HARMAN
urges that Newton's method is to be preferred as being "founded on

accurate principles agreeable to ancient geometry". Newton's method
"requires the supposition of no quantities but such as are finite and
easily conceived" ([16]1, 38-49). He explains that "Newton considers
the simultaneous increments of the flowing quantities as finite, and then
investigates the ratio which is the limit of the various proportions which
those increments bear to each other, while he supposes them to
decrease together till they vanish" ([16]2, 420). While MacLaurin blurs
the similarities between Newtonian and Leibnizian infinitesimal methods,
his concern was with the justification of these methods, appealing to the
canons of classical geometry ([3]).
MacLaurin contests the Leibnizian theory of matter as adopted by
Bernoulli, which formed the basis of Bernoulli's representation of
collision in terms of the action of elastic springs. He argues that "there
never has been discovered as yet anyone body whose elasticity is
perfect; and when any two bodies meet with equal motions, they
rebound with less motions, and there is always force lost by their
collision". It is MacLaurin's case that Bernoulli's arguments apply to the
special case of "perfectly elastic bodies only ... not one of which has
hitherto been found in nature" ([17] 86). On turning to an examination
of the problem of the collision of bodies, MacLaurin argues that "Mr
Bernoulli has resolved only a very limited case of this problem in his
Essay on motion, Paris 1726; for he supposes the bodies to be
perfectly elastic" ([17]192). Moreover, Bernoulli's treatment of collision in terms of the action of elastic springs is erroneous: "it is the last
spring only, which is in contact with the body, that acts upon it, the rest
serving only for sustaining it in its action; so that any change produced
in the body, by whatever name it be called, oUght to be determined
from the action of this last spring only, and in just reasoning ought to
be computed from it alone" ([17]134-5).
MacLaurin therefore concludes that the Leibnizian principle of the
conservation of vis viva, which Bernoulli had sought to explicate, "is not
general". The principle is of use in the case of the collision of perfectly
elastic bodies: "They who hold this principle to be general confine this
theory too much to one sort of bodies, which for any thing appears
from nature have no prerogative above others". MacLaurin does
however allow that the principle of the conservation of vis viva is
applicable to special cases in the theory of mechanics. He observes that
"the principle which Mr. Huygens calls the conservatio vis ascendentis
223
[the conservation of the quantity mv 2 in the descent and ascent of the

centre of gravity of a system of bodies] ... and which seems to be much
the same with what is called the conservatio vis viva of late, obtains
indeed in many cases besides those he has considered, and may be of
use in several inquiries concerning the motions of bodies that have no
elasticity, as well as those that are perfectly elastic, but is not general"
([16]2, 438n.). Equating the Leibnizian vis viva principle with Huygens'
concept, MacLaurin sought to deny Bernoulli's special claim of the
dynamical and ontological significance of the principle. MacLaurin thus
rejects Bernoulli's metaphysical justification (by the law of continuity)
of Leibnizian dynamics, the theory of matter by which Bernoulli
supports his representation of collision, and Bernoulli's claim that the
principle of the conservation of vis viva is a fundamental law of nature,
not merely a special theorem as Huygens had supposed.
MacLaurin's claim in Newton that the principle of the conservation
of vis viva "is not to be held a general principle or law of motion"
([17]194) provides the basis for his general defence and presentation of
the truth of the Newtonian system of natural philosophy. In rejecting
the Leibnizian principle, espoused by Bernoulli, that "the quantity of
force is for ever the same in the universe", MacLaurin affirms the
Newtonian view, stated by Clarke in his correspondence with Leibniz,
that "it may be better that the Author of the world should act
immediately in it, cherishing and governing his work, and sometimes
changing or renewing it". Leibnizian dynamics, which rests on the
principle of the conservation of vis viva which establishes the selfsufficiency of nature, denies Newton's "opinion that the fabrick of the
universe, and course of nature, could not continue for ever in its
present state, but would require, in process of time, to be re-established
or renewed by the same hand that formed it". Thus "the capital doctrine
of this philosophy that represents the universe as a perfect machine,
such as may continue for ever by mechanical laws in its present state, is,
that the same quantity of force and vigour remains always in it, and
passes from one portion of matter to another, without undergoing any
change in the whole" ([17] 84-6). MacLaurin's critique of Bernoulli
thus forms part of a more general attack on Leibnizian dynamics and
theology, an element in his thoroughgoing defence of Newtonian
principles. This entailed a denial of the possibility of the sufficiency of
mechanical laws in explaining the structure of the universe, and rejection of the appeal by Leibniz (in his correspondence with Clarke) and
224
PETER M. HARMAN
Bernoulli (in his "Disc ours sur ... mouvement") to the principle of the
conservation of vis viva as establishing the self-sufficiency and intelligibility of nature. To Bernoulli's claim that if "the quantity of vis viva, the
single source of the continuation of motion in the universe were not
conserved ... all nature would fall into disorder" ([2]3, 58), MacLaurin
countered with the Newtonian argument that "tho' the course of nature
was to be regular, it was not necessary that it should be governed by
those principles only which arise from the various motions and
modifications of un active matter, by mechanical laws" ([17]84).
MacLaurin's Newton concludes with discussion of the concept of
gravity, where these physico-theological arguments are further deployed.
Following Newton and Clarke he claims that gravity "seems to surpass
mere mechanism" ([17]387). Buchdahl has discussed the complex of
arguments arising from this claim, and writes that for Newton
the argument operates however at two different levels, a physical ... and a metaphysical level. It is physical when Newton argues that the growing irregularities in
astronomical motions, as well as the continual "loss of motion" [in inelastic collisions[,
require a power that will "reform" the parts of the universe at suitable moments in time .
. . . On the other hand, the "metaphysical" argument is quite different. ... He contends
that the existence of gravitational phenomena becomes rational (and thus real) only on
the supposition that they are an expression of divine providence, often described as an
"active principle" ... ([5J 179-80).
As in the case of the intelligibility of dynamics, this metaphysical

argument was intended to supplement and justify the concept of gravity,
a dimension to scientific theorizing highlighted in Buchdahl's studies of
Newtonian themes.
NOTES
Leibniz, 'Specimen dynamicum' (1695) ([14J 6,242).
Bernoulli, Jean, 'Discours sur les loix de la communication du mouvement', in Recueil
des pieces qui a remporte les prix de I'Academie des Sciences, Tome premier (Paris,
1732); reprinted in [2[3,7-107.
I
REFERENCES
I. Aiton, E. J., The Vortex Theory of Planetary Motions (London: Macdonald, 1972).
2. Bernoulli, Jean, Opera omnia, 4 vo1s (Lausanne and Geneva, 1742).
3. Bos, H. J. M., 'Differentials, higher-order differentials and the derivative in the
Leibnizian calculus', Archive for History of Exact Sciences 14 (1974), 1-90.
225
4. Buchdahl, Gerd, 'Gravity and intelligibility: Newton to Kant', in The Methodological

Heritage of Newton, eds. R E. Butts and J. W. Davis (Oxford: Blackwell, 1970), 74102.
5. Buchdahl, Gerd, 'Explanation and gravity', in Changing Perspectives in the History of
Science, eds. M. Teich and R M. Young (London: Heinemann, 1973), 167-203.
6. Hall, A. R, Philosophers at War. The Quarrel between Newton and Leibniz
7. Harman, P. M., Metaphysics and Natural Philosophy. The Problem of Substance in
Classical Physics (Brighton: Harvester Press, 1982).
8. Harman, P. M., 'Force and inertia: Euler and Kant's Metaphysical Foundations of
Natural Science', in Nature Mathematized, ed. W. R Shea (Dordrecht: Reidel,
1983),229-49.
9. Harman, P. M., 'Concepts of inertia: Newton to Kant', in Religion, Science and
Worldview. Essays in Honour of Richard S. Westfall, eds. M. J. Osler and P. L.
Farber (Cambridge and New York: Cambridge Univ. Press, 1985), 119-33.
10. Heimann IHarmanl, P. M., 'Helmholtz and Kant: The metaphysical foundations of
Ober die Erhaltung der Kraft', Studies in History and Philosophy of Science 5
(1974),205-38.
11. Heimann IHarmanl, P. M., '''Geometry and nature": Leibniz and Johann Bernoulli's
theory of motion', Centaurus 2i (1977), 1-26.
12. Huygens, Christiaan, Oeuvres completes, 22 vols (The Hague: Nijhoff, 18881950).
13. Kant, Immanuel, Kants gesammelte Schriften, 29 vols in 32. Ed. Konigliche
Preussische Akademie der Wissenschaften (Berlin: Reimer and de Gruyter, 190283).
Halle, 1849-60).
15. Leibniz, G. W., The Leibniz-Clarke Correspondence, ed. H. G. Alexander
(Manchester: Manchester Univ. Press, 1956).
16. MacLaurin, Colin, A Treatise of Fluxions, 2 vols (Edinburgh, 1742).
17. MacLaurin, Colin, An Account of Sir isaac Newton's Philosophical Discoveries
(Edinburgh,1748).
18. Newton, Isaac, Opticks (4th ed., London, 1730; repro London: Dover, 1952).
19. Newton, Isaac, isaac Newton's Philosophiae Naturalis Principia Mathematica: The
Third Edition (1726) with Variant Readings, 2 vols, eds. A. Koyre and I. B. Cohen
CATHERINE WILSON
SENSIBLE AND INTELLIGIBLE WORLDS IN

LEIBNIZ AND KANT 1
Abstract. Kant is generally held to have demonstrated the false pretence of a system
constructed on the order of Leibniz's Monadology to describe a supersensible or
"intelligible" reality. The Critique is thus conceived as superseding all previous
philosophical systems, which are thereby revealed as "dogmatic" or based on wishful
thinking. This is certainly the way Kant portrays himself in works dealing with the
subject of progress in metaphysics. However, when faced with what he regarded as the
corruptions of the contemporary Leibnizians, Kant could also depict himself as the true
defender of Leibniz and his own critical work as the key to the correct understanding of
the systems of the past. Clearly, then, the relation of Monadology to Critique is not
simply the relation of a conjecture to its refutation. The present essay examines some
aspects of Kant's own fascination with the idea of the supersensible, arguing that the
Critique is an attempt to unmake and then remake the notion in more serviceable form.
The relation between the Leibnizian and the Kantian phenomenalisms is discussed in
the main body of the paper.
"Dialectic", said Plotinus,

is the science which can speak about everything in a reasoned and orderly way, and say
what it is and how it differs from other things and what it has in common with them; in
what class each thing is and where it stands in that class, and if it really is what it is, and
how many really existing things there are, and again how many non-existing things ...
with certain knowledge about everything and not mere opinion .... It stops wandering
around in the world of sense and settles down in the world of intellect, and there it
occupies itself, casting off falsehood ... to determine the essential nature of each thing,
and to find the primary kinds, till it has traversed the whole intelligible world; then it
resolves again the structure of that world into its parts, and comes back to its starting
point. .. (Enneads, 1.3.4: [11[1, 157-9).
Here, at last "it busies itself no more, but contemplates, having arrived
at unity". We observe something of an inconsistency in this description
of "dialectic"; for although Plotinus contrasts "settling down" in the
world of the intellect with "wandering around" in the world of sense, it
appears that dialectic (or "metaphysics") is fairly busy with "wandering
around" the intelligible world until it achieves closure, unity, and rest.
This conception of an intelligible world which can be explored by
non-sensory means was one which Kant programatically rejected, while
retaining the spatial metaphor. Metaphysics in its first stage is "a
boundless sea in which progress leaves no trace and on whose horizon
227
228
CATHERINE WILSON
there is no destination ..." ([6] 51). When brought to perfection it is "a

sphere whose boundary turns back on itself" ([6]135). For Kant too,
metaphysics "comes back to its starting point", but by this turn of
phrase he means to stress the narrowness of this realm and not its
superb vastness. It is, above all, made and not discovered, and, like
pure logic, "neither calls for nor permits additions" but requires only
to be maintained against "spiders and wood sprites" ([6] 149f). Meanwhile, Kant thought, it is misleading and in poor taste to borrow, as
some of his contemporaries did, the term mundus intelligibilis with its
connotations of beauty, permanence, and divinity to describe, for
example, theoretical astronomy by contrast with the purely observational type. Of course the intellect is required to deal with what
appears. The question is whether it can deal with what does not and
cannot appear, with purely intelligible objects ([4] A256f!B312f).
Kant's attack on certain uses of the notions of intelligible objects and
intelligible worlds is, I hope to show in this essay, particularly interesting in light of his criticism of Leibniz. For Leibniz's Monadology was
the best example Kant knew of a philosophical system of intelligible
objects: in this case, a community of non-sensible beings, monads,
"living mirrors", existing in a relationship of pre-established harmony,
ordered by their distance from God. But it would be utterly false to
suggest that Kant simply pushed Leibniz into the negligible category of
Plotinian dialectical wanderers. And there were two reasons he could
not do so. The first was the well-known problem of the ideal or nonideal nature of space. The second was that Kant was transfixed by an
idea that is rooted in the Symposium and that has always driven the
idea of an intelligible world forward - the idea that the intelligible
world is the world of compensation. And although it is true in a general
sense that Kant showed the Leibnizian constructive metaphysics to be
impossible, this model case of philosophical progress is full of perplexities. 2 Kant stated once that the Critique of Pure Reason constituted
the "true apology for Leibniz". The interesting question is not whether
he truly meant this, but what the statement itself could have meant.
1. SENSIBLE AND INTELLIGIBLE IN LEIBNIZ
Leibniz saw a clear relation - though it is not always easy to say what
this relation was - between his own discoveries about the conservation
SENSIBLE AND INTELLIGIBLE WORLDS
229
of forces, and the discoveries of others about the new world revealed
by the microscope, and his metaphysical teachings. The Specimen
dynamicum, for example, moves back and forth between "empirical"
argument about pendulums, rotating and falling bodies, rebound, and
relative motion, and apparently non-empirical conclusions concerning
the relation between mind and body, the goodness of God, and the
distribution of vital principles throughout the universe. 3 In the New
Essays, he indicates that microscopical investigations into plants,
insects, and the comparative anatomy of animals should strengthen the
likelihood of his doctrines. "I am merely going on beyond our observations", he explains in this connection, "not restricting them to certain
portions of matter or to certain kinds of action ... the only difference is
that between large and small, between sensible and insensible"
([81 474).
Such presentations suggest that the "intelligible world" of the
Monadology is given simply by detailed attention to the consequences
of empirical observation and experiment: Leibniz is often concerned to
show that the ontology of the "Cartesians" or of the corpuscularians
who ascribe only magnitude, figure, and motion to their ultimate
particles runs into conflicts with experiment and observation. He tries
to show that the only way to bring the "sub-sensible" world into
alignment with the results of physical experiment is to make major
modifications in this ontology. But this involves in turn negating some
of the more evident revelations of sense-experience, notably its tendency
to turn up only objects possessing magnitude, figure, and rest-ormotion. When this pattern of argument is run out we can see that
Leibniz sometimes argues in a quite direct way for the extension of
observed properties into the subvisible realm (from animalcules to
unextended living beings); sometimes in an indirect way (from the
behaviour of billiard balls to the denial that, at some ultimate level,
substance is billiard ball-like). But these patterns of argument differ in
turn from the Leibnizian presentation of the sensible world as a
"picture" of a non-sensible world. Thus, in the New Essays, we read,
As for motion: it has only phenomenal reality, because it belongs to matter or mass
which is not strictly speaking a substance. Still there is a semblance of action in motion,
as there is a semblance of substance in mass .... I give to bodies only a semblance of
substance and of action, because something made up of parts cannot, strictly speaking,
be accounted a substance any more than a herd can .... Its unity comes from thought
([81 211 ).
230
CATHERINE WILSON
On this scheme matter may be said to hint at the existence of substance,

motion at the existence of activity. But we know on analytical grounds,
Leibniz says, that substance cannot actually be material or the motion
we observe and experience be real, because no piece of matter could
satisfy the requirement of being an unum per se, and because the
statement "X has moved" has a sense only when a particular (arbitrary)
reference-frame has been designated. 4 Are we here on our way to what
Nietzsche calls a disparagement of the real world? 5 Not precisely.
"Only phenomenal reality" sounds vaguely prejudicial, but the picture
here is still one of a physics supplemented by metaphysical considerations.
At times however, this picture theme takes on another aspect with
Leibniz suggesting that the sensible world is dreamlike or even that its
existence is in some way doubtful. This position is normally put forward
as a gloss on Plato, but it surfaces often enough to force us to take it
seriously:
What the ancient Platonists have said is ... quite true and worthy of consideration that the existence of intelligible things, particularly of the I who think and am called a
mind or soul, is incomparably more certain than the existence of sensible things, and
that it would not be impossible, speaking with metaphysical rigour, that there should
exist at bottom only intelligible substances, of which sensible things would be only the
appearances. 6
This doctrine is mysteriously ascribed to Locke in the New Essays

([81 442). It appears again in a fragment of questionable date:
(Bodies in themselves are no self-states but shadows which flow away.) Corporeal
things are but shadows which flow away, glimpses, shapes, truly dreams ... 7
Clearly Leibniz finds this position attractive. But he is reluctant to give

himself over to it. He dislikes mysticism when it takes on quietistic
overtones; he observes more than once that metaphysical experiences
have a value but are unrelated to the active acquisition of knowledge.
People subject to such experiences ought to be protected and preserved
"just as one preserves a curiosity or a cabinet piece" but one ought not
necessarily try to follow their example. Leibniz typically stresses the
danger of failing to combine metaphysics with "physics and mathematics" citing for example the cases of Steno and Swammerdam who
became enthusiastic converts and abandoned anatomy and natural
history as vanitas. 8
231
There is however one area in which Leibniz is clearly dissatisfied

with what the world of immediate experience has to offer and that is
ethics. Here the results of observation, experience, tests and trials, are
only discouraging, for reward and punishment seem to be meted out in
a less than perfectly systematic way. We see the good suffering while
the evil pursue their wicked plans unchecked. Brave and decent actions
go unrecognized, while purely self-serving actions get rewarded. Resistance to tyranny, Leibniz notes, may be ineffective: all one can do is
protest. For these complaints to be meaningful, "men must mean something else by justice and right than that which pleases a powerful being
who remains unpunished because there are no judges capable of
mending matters".9 We could say that the meanings of "good", "evil",
"just", etc. can be fixed only by reference to an intelligible world. The
Monadology attempts to show what they mean by describing a supersensible moral order. The Monadology states, of course, that in the long
run, the sensible world will prove to have been a system after all. But in
the meantime, we humans, qua monads are already part of a moral
system in which justice and right are not defined in relation to temporal
power, or excellence in relation to temporal rewards.
From this brief sketch it should be apparent that the intelligible
world of immaterial substances endowed with life, perception and
activity, occupying well-defined positions in the hierarchy of the
kingdom of grace is defined in and through its relation to the world of
experience, replicating certain of its features, contradicting others. That
the procedure is already illegitimate might be suspected by anyone who
observes that the "choice" of where to replicate, where to contradict, is
ultimately determined by dogmatic purposes. But this judgment is one
of hindsight. No mind is equipped from birth with the distinction
between critical and dogmatic metaphysical systems, which it was
Kant's glory to have discovered. The question is - where did Kant find
the foothold in Leibniz's theories which enabled him to develop it?
It is not easy to reconstruct the details of Kant's development of a
full-blooded critique of Leibniz, but we can identify roughly the
following critical stages: (1) Kant's scepticism about the possibility of
knowledge of such "spiritual" unobservables as the souls of the living
and of the departed; (2) His insistence on a separation between the
faculties of sensibility and thought, between observable phenomena and
unobservable noumena (following naturally on the concerns of (1 )); (3)
His rejection of the claim that sensation is a form of "confused
232
CATHERINE WILSON
cognition" (a corollary of (2; (4) His attack on Leibniz's theory of

space and time (which might be seen as requiring this "confusion"
theory); and, (5) His general criticism of the Monadology. Thus, Kant's
attack on what (he thought) the Monadology says - that experience is
a confused apprehension of an intelligible order - turns into a
magnificent attack on what the Monadology is - an intellectual fantasy,
a dream-kingdom. And what has emerged along the way is what Kant
described in his famous letter to Herz as "the key to the whole secret of
hitherto still obscure metaphysics"; notably that "conformity" between
mind and object might be ensured in either of two ways: either by the
objects being the cause or source of our idea as in the case of sensory
experience; or by the mind's being the source of the object ([3] 71).
This discovery furnished him both with a critical principle and a
constructive tool.
Taking now for granted steps (2) and (3) which are set out in the
pre-critical treatise "On the form and principles of the sensible and
intelligible world" ([2]45-92), let us look more closely at step (4),
Kant's true point of entry into the Monadology.
2. KANT'S CRITICISM OF
LEIBNIZ'S APPEARANCE-AND-REALITY SCHEMES
Kant's most sustained criticism of Leibniz's doctrines is found in the

"Amphiboly" section of the Critique of Pure Reason. It is initially
difficult, however, to match up what he says there about Leibniz with
what Leibniz says. For example, he charges that the latter
took the appearances for things-in-themselves, and so for intelligibilia, i.e., objects of
the pure understanding (although, on account of the confused character of our
representations of them he gave them the name of phenomena), and on that assumption
his principle of the identity of indiscernibles ... certainly could not be disputed. But
since they are objects of sensibility, in relation to which the employment of the
understanding is not pure but only empirical, plurality and numerical difference are
already given us by space itself, the condition of outer appearances ([41 A2641B320).
Because, "in the concept of pure understanding matter is prior to

form", Kant explains,
Leibniz first assumed things (monads), and within them a power of representation, in
order afterwards to found on this their outer relation and the community of their states.
. . . The intellectualist philosopher could not endure to think of the form as preceding
233
the things themselves and determining their possibility - a perfectly just criticism on
the assumption that we intuit things as they really are, although in confused representation ([41 A267/B323).
Here Kant has combined his two fundamental criticisms of Leibniz,

that he did not make the correct distinction between appearances and
things-in-themselves, and that he regarded perception not as a separate
faculty from that of cognition, but as a confused or "darkened" form of
approximate cognition. How far do these criticisms touch at the heart
of Leibniz's procedures? To answer this question we must first look at
the variety of stances Leibniz adopts on the question of "representation" or "the truth about sensible things" and its relation to intelligible
concepts. We have already noted the presence of one scheme in the
New Essays: the sensible world is a picture or a semblance in some
respects of an intelligible world. Call this "Platonic" scheme Scheme A.
It is quickly apparent that the New Essays themselves contain various
other schemes relating reality to appearance. Two are particularly
prominent: neither one mentions or requires intelligible objects such as
monads.
Scheme B: Sensible Reality as Regulated (i.e., Intelligible) Coherence:
When not in a Platonizing mood, Leibniz is impatient with the sceptical
hypothesis that the world is a dream. I cannot prove, he observes
against the academic sceptic, that the world is not an hallucination
induced in me by an omnipotent demon. But it is fully possible and
appropriate nevertheless to distinguish between real and imaginary
phenomena. "Imaginary" sequences of phenomena are characterized by
their "incoherence", by their "irregularity", by their irreproducibility.
The most powerful criterion of the reality of phenomena, he says in
tum, "sufficient even by itself, is success in predicting future phenomena
from past and present ones, whether that prediction is based upon
reason, upon a hypothesis that was previously successful, or upon the
customary consistency of things as observed previously". The coherence of phenomena consists in the possibility of explaining them all by
reference to "some sufficiently simple hypothesis".l0 Dream material is
incoherent - I could not expect to find reproducible sequences of
events, or occurrences explainable by simple hypotheses, or information useful to others. If someone now maintains the metaphysical
hypothesis that, for all I know, I am being deceived by an omnipotent
demon who produces regular, coherent, durable sequences of phe-
234
CATHERINE WILSON
nomena, this hypothesis, says Leibniz, changes nothing in scientific

practice.
Although this position implies that no "proof of an external world" is
needed to justify behaviour in ordinary life, or the practice of science,
the higher-order principles "coherence", "consistency", "cause-andeffect", and "sameness" are needed to distinguish real from imaginary
phenomena. Note that here there is no talk of an intelligible world;
these are simply "intellectual principles", which, as long as they distinguish between well-ordered and badly-ordered strings of phenomena,
cannot themselves be termed "sensible". But although Leibniz sometimes appears to consider this scheme fully adequate for scientific
practice, this is only in the context of academic or sceptical threats.
When a threat emerges from another direction, in the form, for
example, of Newton's theory of gravitation, he abandons it entirely.
Newton's theory of gravitation as either a mathematical fiction or a
spiritual force is coherent, consistent, and links the phenomena together
admirably. But Leibniz finds it wholly unacceptable. This leads us to an
entirely different presentation of the relation between experience and
intelligible concepts, viz., Scheme C: Sensible Reality as explained by
Mechanical (i.e., "Intelligible") Causes.
Leibniz's third theory looks for the truth about sensible things in
their underlying structure. All properties and powers - taste, colour,
odour, gravity, magnetic attraction - are the products of the sub-visible
arrangement of moving particles. This is the usual seventeenth-century
critical scheme, embraced by Descartes, Boyle, and Locke in addition
to Leibniz. We even find him arguing that a methodological atomism is
appropriate. The properties which do the explaining in this case Leibniz
describes variously as belonging to the common sense, as being
concepts of the imagination, or mathematical concepts. "... sense
qualities are capable of explanation and rationalization only insofar as
they have a content common to the objects of several, external senses
and belong to the internal sense. For whenever one tries to explain
sensible qualities distinctly, one always turns back to mathematical
ideas".ll
Sensible qualities and the observable modifications of bodies may be
related to their underlying causes in two ways. "Magnetology", to use
Leibniz's example, "is a science, for from a few assumptions grounded
in experience we can demonstrate by rigorous inference a large number
of phenomena". The relation between phenomena and assumptions is
235
like the relation between theorems and axioms in geometry: we cannot

expect all experiments to be explainable by our theoretical assumptions,
but neither are all theorems demonstrable by the geometer. What we
seek is, in both cases, an extension of the fit between "rational
principles" and experience ([8]453). In other cases, we have no hope of
explaining the phenomena from any set of rational principles. This is
not because there is no fit, but because of the "confused" nature of
perception itself:
... when the swift rotations of a cog-wheel makes us perceive an artificial transparency
... we are not able to discern the idea of the cause of this, i.e., the idea of the teeth of
the wheel. The wheel's rotation makes the teeth disappear and an imaginary continuous
transparent ring appear in their place; it is made up of successive appearances of teeth
and of gaps between them, but in such rapid succession that our imaginations cannot
distinguish them ([8]403).
If the confusion ceases, he now observes, if the wheel slows down

long enough for us to observe the teeth and gaps, we lose the
perception we are trying to explain. But this does not mean that the
relation between the phenomenon and its cause is "arbitrary"; we grasp
that the image is an expression of its underlying cause although we
cannot grasp the process of expression, and so see the relation as
inexplicable.
3. KANT AND THE LEIBNIZIAN PHENOMENALISMS
When Kant claims that Leibniz took appearances for things in themselves, what could this conceivably mean? Does he not distinguish
consistently between sensible appearances and their causes, between
phenomena and the reasons for them? Or even, when in his Platonizing
mood, between intelligible substances and material objects?
Kant detects, nevertheless, a problem: Leibniz treats the objects of
experience as phenomena, acknowledging thereby that they depend on
our powers of representation. Yet he fails to see the implications of his
own position. For, he goes on to suppose that that whose existence is
not dependent on our powers of representation is an object of a special
type, and that the properties of these objects must call up spatial and
temporal relations. Had he realized that this power of representation in
us itself calls up spatial and temporal relations, he would not have
needed to seek the basis of these relations in the representations of the
236
CATHERINE WILSON
monads. What he was looking for in these remote objects lay all the
time close at hand.
Kant can, moreover, explain this delusion. Leibniz failed to see that
the principle of the identity of indiscemibles was entirely otiose in its
application to objects of sense. We do not require concepts to differentiate between, say, two matching armchairs. Such objects are already
perceived as having distinct spatial locations. What blinded Leibniz to
the consequences of this fact was his determination to find qualitative
differences even between two drops of water or two atoms. He
supposed, in other words, that water-drops (appearances) are differentiated by concepts (like things-in-themselves). Had he admitted that
water-drops and atoms can be identical in all respects except spatial
and temporal location, Kant thinks, Leibniz would have realized that
space and time are "prior to" substance and thus that monads were
neither objects nor the foundations of space and time. But why did
Leibniz suppose that the identity of indiscemibles should apply to the
objects of experience?
Here I think Kant has detected, in a way which is less than fully
explicit, the dual nature of the Leibnizian intelligible world as extension
and replacement. The principle in question shows this tension. Leibniz
sometimes presents it as a generalization from experience. In the
gardens of Herrenhausen, he argues, no two leaves are alike; and "two
drops of water, viewed with a microscope, will appear distinguishable
from each other". \2 So, one would be breaking abruptly with the
evidence of the senses if one were to assert that a qualitative uniformity
underlies all this variety. In this sense the Monadology just extends
absolute differentiation to the sub-sensible level. But the principle is
also asserted dogmatically in order to exclude the possibility of
identical atoms even supposing the microscope were to "reveal" them.
Given two identical-looking dust particles, the Leibnizian will argue
that this is only appearance, with apparent homogeneity disguising a
true heterogeneity. But given two dissimilar oak-leaves, the Leibnizian
will seize on this appearance as a true one, with evident heterogeneity
proving an unlimited one. This is not entirely fair.
Kant sees, moreover, that the extensive machinery of absolute
differentiation at the monadic level has been introduced in the hope of
giving a sense to the assertion that although A and B are not in fact
distinguishable by means of sense except by reference to spatial
position or temporal location, they are distinct in fact. The pathos is
237
that for all of this extensive machinery, Leibniz cannot even explain
what distinguishes the left-handed glove from the right-handed glove.
Taking space and time as "forms" and as prior to objects is a
hypothesis, which unlike Leibniz's, solves a problem.
Kant thus intends to show in this discussion that (a) the identity of
indiscernibles considered as a theory of logic could not be used as an
argument against atomism considered as a physical theory; (b) the
representations of the monads can never explain but only multiply the
instances of the problem of the nature of space and time; and, (c)
monads can not be objects, not even "intelligible objects". But the
attack is more comprehensive: Kant is concerned to show that Leibniz
has no tenable Appearance-and-Reality Scheme whatsoever.
The problem is that in doing so Kant appears to blend together
Platonizing scheme A and Mechanical scheme C. At one point he
ascribes to the Leibnizians the doctrine that "if one were completely
conscious of all the representations contained in the intuition of a body
that would ... provide a concept of it as an aggregate of monads"
([61 89). No doubt he was thinking of passages like the following in the
New Essays:
sensory ideas depend on the shapes and motions, which they precisely express, though
the mechanical motions which act on our senses are too small and too great in number
for us to sort out this detail within the confusions. But if we had arrived at the inner
constitution of bodies, these [sensible qualities] would be traced back to their intelligible
causes ([8] 403).
It looks as though Kant might simply have misread this paragraph as

expressing Scheme A, here where it is in fact simply a statement of the
more matter-of-fact Scheme C. If so, then Leibniz never maintained
that the sensible world is a confused representation of an intelligible
Monadology, this being the excited extension of his followers. And this
is generally in accord with Leibniz's carefully articulated distinction
between "composition" and "foundation". As he explains to de VoIder:
Accurately speaking ... matter is not composed of [these] constitutive unities but
results from them, since matter or extended mass is nothing but a phenomenon
grounded in things, like a rainbow or the mock-sun, and all reality belongs only to
unities. Phenomena can therefore always be divided into lesser phenomena which could
be observed by other more subtle animals and we can never arrive at smallest
phenomena. Substantial unities are not parts, but foundations of phenomena. I 3
238
CA THERINE WILSON
Thus when Leibniz confesses to Bernoulli, "[b]ow far a piece of flint

must be divided in order to arrive at organic bodies and hence at
monads, I do not know",14 he does not mean that mechanical division
elicits monads; rather mechanical division elicits animalcules which,
qua living beings, possess souls or dominant monads. Further mechanical division of animalcules yields them to infinity. "Intelligible causes"
are then, in the New Essays passage, mechanical causes. This distinction
between foundation and composition is one Kant himself later acknowledges. "Is it really believable", Kant asks Eberhard, a "Leibnizian",
that Leibniz, the great mathematician, held that bodies are composed of monads ... ?
He did not mean the physical world, but its substrate, the intelligible world, which is
unknown to us. This lies merely in the Idea of reason, and in it we must certainly
represent to ourselves everything which we think as a composite substance as composed of simple substances. He also seems, with Plato, to attribute to the human mind
an original, though now only obscure, intellectual intuition of these super-sensible
beings. He wishes these latter to be considered as relative to a special mode of intuition
... thus, as mere appearances in the strictest sense, dependent upon (specific, particular) forms of intuition. We cannot, therefore, be disturbed by his explanation of
sensibility as a confused mode of representation ... ([51158).
Note that what Kant had identified as Leibniz's central error in the
Critique is here described as something which should not disturb us
once we have understood its point. But Kant of course does not
endorse the theory; he cannot accept that in addition to the mechanical
process of aggregation, and the psychological process of "composition"
belonging "not to sensibility's receptivity, but to the understanding as a
a priori concept" ([6] 85-7), there exists some wholly different type of
founding-procedure which takes simple substances as its material and
makes out of them divisible physical objects. The Leibnizians may have
got the doctrine all wrong in supposing the monads to be components
of physical objects, but there is no way of getting the doctrine right.
Kant is now in a position to introduce his own Appearance-andReality Scheme which differs significantly from A and C, while showing
certain noteworthy affinities with Leibniz's scheme B. Kant's intelligible
objects are limiting cases. They do not cause experiences, they do not
compose the objects we handle, they do not affect us. For the real,
intelligible object is simply the object of experience thOUght of in
abstraction from all possible empirical determinations, including causal
powers. Thus Kant can say, effectively shattering even the powerful
SchemeC:
239
Neither bodies nor motions are anything outside us; both alike are mere representations
in us; and it is not, therefore, the motion of matter that produces representations in us;
the motion itself is representation only, as also is the matter which makes itself known
in this way ([41 A387).
All this critical activity has its price. For both Leibniz and Kant - as
for the majority of seventeenth-century philosophers and perhaps for
everyone since - sensory experience presents a mystery. Why yellow,
why sweet, why violet-scented? Kant took Leibniz (who had no
tolerance for metaphysical mysteries) to be arguing that sensory
experience is possible and assumes the different forms it does because
it is a variety of something less inherently puzzling, namely thinking.
(Nobody asks: Why is it the nature of the mind to think?) In insisting
that cognition is not merely "logically" (i.e., as a matter of degree)
distinct from sensation, but typologically, Kant deprives himself of this
means of solving the mystery. The "secret of the source of our
sensibility" lies hidden "in the mind . . . so deeply concealed that we,
who after all know even ourselves only through inner sense and
therefore as appearance, can never be justified in treating sensibility as
being a suitable instrument of investigation ..." ([4] A278/B334).
4. KANT'S RECOVERY OF THE INTELLIGIBLE WORLD
When he is concerned with perception, sensation, and the subject

matter of science, Kant's aim is to break off the Leibnizian bonds to the
intelligible world. But this aim is subject to a remarkable reversal in the
later sections of the Critique where, on the backs of his unknowable
noumena, Kant develops a theory of intelligible agency which provides
a bridge from the sensible to the super-sensible. This bridging operation, he argued, was indeed the aim of metaphysics; only it was to be
undertaken not as a "dangerous leap", "not continuously", but with
"checks at the boundaries of both realms" ([6]79).
Like Leibniz, Kant thought that experience did not show ethics to be
a system. At the same time, he could not allow it a dependence on
another trans-empirical system as Leibniz had. "Does not the heart of
man contain its own moral laws?" he asks at one point, "... and is it
really necessary to start up the big engines from the other world to
induce man to move in the direction of his destiny?" ([1]97). Kant
wanted to answer "yes" to the first question and "no" to the second. But
240
CATHERINE WILSON
this seemed hopeless. This conflict induces a long, exhausting, and

unresolved struggle, during which he attempted to show that the
intelligible world had no reality, but a use.
Kant's struggle with the Leibnizian problem began well before his
critical period. The case of Swedenborg, whose fantasies and hallucinations had brought him an enviable celebrity in the 1750s and 1760s,
provided him with a focus. Swedenborg had allegedly learned through
his instruction by angels, as well as in conversation with "communities"
of beings inhabiting different members of the human body, that there
were two worlds: a "spiritual world where spirits and angels are and a
natural world where people are". The latter, he was informed, is
"intrinsically dead", except insofar as it is infused with life from the
spiritual realm. "This takes place with human beings and animals, and
constitutes the life force of trees and plants". The two realms are
thus actually interpenetrating, but this can be experienced only unconsciously. Spiritual beings are in direct non-physical communication with
one another through the technique of reading each other's memories. ls
Swedenborg's two kingdoms, his "communities", and his memoryreading spirits recall the basic features of the Monadology. This is, no
doubt, the basis of Kant's early equation in the Dreams of a Spirit Seer
of 1663, between the sensory dreams of the religious visionary and the
metaphysician's "dream of reason". But the most interesting feature of
this early work is that it affords Kant a direct opportunity to give the
issue which is on his mind an airing, knowing in advance where the
limits of believability lie. In an opening chapter entitled "A Fragment of
Occult Philosophy", he offers a semi-satirical sketch towards the
establishment of a "community of spirits", which anticipates in its main
lines his later moral theory. Assume, he says, that matter possesses
inertia, permanence, solidity, shape, and extension, and lends itself to
mechanical explanations. Now consider that living beings do not make
any addition to the total quantity of matter or extension, but neither do
they obey laws of contact and impact. They are capable of animating
themselves and of animating inert matter. Thus they must constitute an
immaterial world (mundus intelligibilis) and be related to each other
not only through their bodies - this ordinary mode of communication
being secondary and contingent. A soul thus inhabits two worlds: it can
have clear perceptions of the sensible one but also "receive and
transmit the influence of pure immaterial beings". Finally, suggests
241
Kant, becoming more serious, the existence of this intelligible world

explains the moral impulse:
Among the forces which move the human heart we may distinguish some which are
very powerful but originate outside human nature ... these ... refer to those
tendencies which cause us to shift away from our personal motives to those of other
rational beings .... In this manner originate our moral impulses which tear us away
from the pressure of self-interest and become the stronger law of duty and the weaker
law of benevolence .... On account of this we realize that we depend in the manner of
our innermost motivations upon the rule of a Universal Will and this is what gives rise
to the moral unity of all thinking beings and to a systematic constitution based purely
on spiritual laws ([ 1] 51).
Participation, moreover, in this moral world has a compensatory

function. "The true intentions, the secret motives of so many defeated
plans and frustrated ambitions ... the slyness of seemingly beneficial
actions" are inappropriately met in this material world, hence we must
posit "pneumatic laws" under which bodies occupy a given moral
position just as "material bodies of this universe occupy a definite
position in relation to one another according to the laws of motion and
their materialforce ..." ([1) 52).
Later in the work, the existence of spiritual influences and the
spiritual community of moral beings is said to be indemonstrable: I
shall never know, Kant insists, whether I am in contact with other
minds by pneumatological laws, or whether my only connection is
through physical laws, all contact being mediated by the body, in which
case the rest is only a "poetic fiction". But, fiction or not, the intelligibile world survives, and is reconstituted as a practical-dogmatic
necessity in the Critique where it ensures that morality is a system.
Happiness, therefore, in exact proportion with morality of the rational beings who are
thereby rendered worthy of it, alone constitutes the supreme good of that world
wherein, in accordance with the commands of a pure but practical reason, we are under
obligation to place ourselves. This world is indeed an intelligible world only, since the
sensible world holds out no promise that any such systematic unity of ends can arise
from the nature of things.... [It constitutes] an order which in the world of sense is in
large part concealed from us ([4) A8141B842).
The "intelligible world" is the moral world "in which we leave out of
account all the hindrances to morality" (the desires) creating thereby a
system "in which happiness is bound up with and proportioned to
morality". This "can be conceived as necessary" ([4) AS09/BS37). God
242
CATHERINE WILSON
and a future life are thus "inseparable from the obligation which that
same reason imposes on us". Accustomed to Kant's minimalism in the
derivation of metaphysical necessities, we may be surprised to find that
the incommensurability between the ceremonies associated with human
generation and the spiritual dignity of man permits indeed the following
overwhelming transcendental hypothesis, a Plotinus-Leibniz-Swedenborg
gala production, to be asserted "defensively":
namely, that all life is, strictly speaking, intelligible only, is not subject to changes of
time, and neither begins in birth nor ends in death; that this life is an appearance only,
that is, a sensible representation of the purely spiritual life, and that the whole sensible
world is a mere picture which in our present mode of knowledge hovers before us, and
like a dream has in itself no objective reality ([4] A780/B808).
It is hard not to suspect Kant of over-dramatizing - but nothing in the
text supports this suspicion. On the contrary, the conviction expressed

is that problem and solution are fully commensurable.
5. CONCLUSION
From our external perspective, Kant's treatment of intelligible worlds

shows the same tendency to reverse operations at critical points,
revealing them as now useless, now as indispensable, breaking down
there, building up here the links between sensible and intelligible which
led us to suspect, in Leibniz's push towards the Monadology both as
extension and as replacement, that the terms of the inquiry were less
than free. Kant asserts - to be sure - that we know nothing about
matters in the intelligible world, that it is inaccessible. Yet the notion
remains at the centre of the general scheme of the Critique; the bit of
irritating grit around which its beautiful layers are spun. This fact - and
the details surrounding it - serve as a reminder that the "refutation"
and the detection of "error" on which philosophical progress is supposed to depend are awkward terms for the shifts, inversions, and
appropriations which characterize the actual historical example. As
Kant replied to Eberhard:
The Critique of Pure Reason can thus be seen as a genuine apology for Leibniz ... just
as it can be for many different past philosophers, to whom many historians of
philosophy with all their intended praise only attribute mere nonsense. Such historians
cannot comprehend the purposes of these philosophers because they neglect the key to
the interpretation of all products of pure reason from mere concepts, the critique of
243
reason itself. They are thus incapable of recognizing beyond what the philosophers
actually said, what they really meant to say ([ 5J 160).
NOTES
I The author is grateful to the A. von Humboldt-Stiftung for support during the writing
of this paper in 1986.
2 Thus H. J. Paton presents Kant's analysis as follows: "because of his [Leibniz'sJ
uncritical assumption that pure concepts unaided by sense can give us knowledge of an
intelligible world, he is able to construct a theory - a very queer theory - of reality as
it is in itself" ([1 OJ 86). Kant does indeed describe pre-established harmony as "the
queerest invention" (das wunderlichste Figment). But Paton ignores the question
whether Leibniz did in fact uncritically assume that pure concepts unaided by sense
give us knowledge of an intelligible world, and as it turns out, Leibniz held no such
position .
.1 Specimen dynamicum (1695) ([7J 435-52).
4 Leibniz repeats these arguments against matter-and-motion frequently. See especially
the Correspondence with Arnauld ([7J 338,343); Specimen dynamicum ([7J 445).
5 [9J 39. Nietzsche described all such intelligible alternatives as "moral-optical illusion".
We discuss this problem below.
6 'On What is Independent of Sense and Matter' (1702) (Letter to Queen Sophie
Charlotte, 1702) ([71 549).
7 'On the True Theologia mystica' (c. I 690?)([71 368).
x 'Reply to the Thoughts on the System of Preestablished Harmony Contained in the
Second Edition of Mr. Bayle's Critical Dictionary, Article Rorarius' (1702) ([7J 584).
Y 'Reflections on the Common Concept of Justice' (1702?) (17J 563).
10 'On the Method of Distinguishing Real from Imaginary Phenomena' ([7J 364).
II 'What is Independent of Sense and Matter' ([71 548).
12 Fourth Letter to Clarke (1715-16) ([71 687).
IJ
14
Letter to de Voider (1704)([71536).

Letter to Bernoulli (1698) ([71512).
15 [121 228f. This work was published after Kant had written his piece, which relied on
the 12-volume Arcana coelestia (1747-1753).
REFERENCES
I. Kant, Immanuel, Dreams of a Spirit Seer (1765), tr. J. Manolesco (New York:
Vantage, 1969).
2. Kant, Immanuel, 'Inaugural dissertation', in Kant: Selected Pre-Critical Writings, tr.
and ed. G. B. Kerferd and D. E. Walford (Manchester: Manchester Univ. Press,
1968).
3. Kant, Immanuel, 'Letter to M. Herz', in Philosophical Correspondence, 17591799, tr. and ed. A. Zweig (Chicago: Chicago Univ. Press, 1967).
4. Kant, Immanuel, Critique of Pure Reason, tr. N. Kemp Smith (London: Macmillan,
1929).
244
CATHERINE WILSON
5. Kant, Immanuel, 'On a discovery according to which any new critique of pure
reason has been made superfluous by an earlier one' (1790), in The KantEberhard Controversy, ed. H. E. Allison (Baltimore: Johns Hopkins Press, 1973).
6. Kant, Immanuel, What Real Progress has Metaphysics made in Germany since the
Time of Leibniz and Wolff? (1793-1802?), tr. and ed. T. Humphrey (New York:
Abaris, 1983).
7. Leibniz, G. W., Leibniz: Philosophical Papers and Letters, ed. L. E. Loemker (2nd
ed., Dordrecht: Reidel, 1969).
8. Leibniz, G. W., New Essays (1765), tr. P. Remnant and J. Bennett (Cambridge:
9. Nietzsche, F., Twilight of the Idols, tr. R. J. Hollingdale (Middlesex: Penguin,
1968).
10. Paton, H. J., 'Kant on the errors of Leibniz', in Kant Studies Today, ed. L. W. Beck
(La Salle IL: Open Court, 1969).
11. Plotinus, Enneads, 6 vols, tr. A. H. Armstrong (London and Cambridge MA:
Heinemann, 1965).
12. Swedenborg, Emmanuel, Soul-Body Interaction (1769), in Emmanuel Swedenborg:
The Universal Human and Soul-Body Interaction, tr. and ed. G. F. Dole (New
York: Paulist, 1984).
OLIVER N. H. LEAMAN
TRANSCENDENTAL REASONING AND THE

INDETERMINACY OF THE HUMAN POINT OF VIEW
Abstract. The human point of view is clearly an important topic for Kant, but is rarely
connected with his important discussion of indeterminacy. He makes a sharp distinction
between the precise notions of mathematics and the indeterminate notions of philosophy. Leibniz, by contrast, believed that a philosophy could usefully be constructed
consisting of determinate notions, with its paradigm God's clear and perfect view of
reality. What we perceive as confused is just a result of our limited view of what is
really clear and exact. Kant and Wittgenstein argue that we are obliged to base our
most important practices upon facts and not on a metaphysically sure foundation. Our
presuppositions will doubtless vary over time as our nature changes and develops, and
the validity of transcendental arguments will be accordingly limited. Justification must
take place within the same limited context as every other human activity, and cannot
transcend the bounds of possible experience.
In the many discussions which have taken place about what Kant means
by "object" or "transcendental idealism" or "the human standpoint" a
vital feature of Kant's thought is almost entirely ignored. This is his
thesis of indeterminacy. It is sometimes mentioned that Kant has something to say about indeterminate judgments, but this is never regarded
as being an important aspect of the critical philosophy. To a certain
extent this is because Kant's most developed comments on indeterminacy are placed towards the end of the Dialectic of the first Critique,
in the Discipline of Pure Reason, a region of the work which is
generally believed not to contain much of interest. Another reason for
treating this section with suspicion is that it is virtually identical to part
of a pre-critical work, and so on the "Patchwork Thesis" favoured by
Kemp Smith and many other commentators it may be regarded as not
integral to the critical thesis. But it will be argued here that it is
impossible to come to a complete appreciation of what Kant has to say
about idealism and objectivity without also taking into account what he
has to say about indeterminacy. In particular, his account of indeterminacy is closely related to his notion of the human standpoint and the
limitations which are involved in transcendental arguments. Yet this
doctrine of indeterminacy is generally ignored or even bluntly denied,
as in Korner's claim that "Kant also failed to consider the difference
245
246
OLIVER N. H. LEAMAN
between exact mathematical and inexact empirical judgments" ([4]104).

This is an incredible comment to make, considering Kant's continual
emphasis upon precisely that difference.
Kant establishes his account of indeterminacy by discussing the
difference between mathematical and philosophical terms. Mathematics
is said to start generally from definitions, which are patently adequate
because they are self-evident axioms, and reaches intuitively certain
conclusions by following a demonstration. Such definitions represent
the complete concept of a thing. For example, the algebraist achieves
his results by using symbols in accordance with certain rules, and he
could not achieve those results without an analogous intuitive representation of his concepts ([2] A 717 IB7 45). It is this particular feature
of mathematical reasoning which provides that procedure with its
clarity and connects it to sensibility - the human form of intuition.
Kant claims that "This method, in addition to its heuristic advantages,
secures all inferences against error by setting each one before our eyes"
([2] A7341B762). For instance, the way one finds out that 7 + 5 = 12
is by a process like counting, going from 7 to 12 by successive
additions of 1, thereby operating with a particular instance of a group
of five objects which can only be given in intuition ([2] B15-16,
A142/B182). Since intuition is an immediate relation to objects,
mathematics based upon intuition seems also to be immediate knowledge which, even though synthetic a priori, does not require the same
sort of complicated justificatory argument that the categories do
([2] A87/B120).
According to Kant, the nature of definition in mathematics and
philosophy is different. The latter defines analytically, the former
synthetically. Analytic definitions analyse a concept already given and
so may go wrong by failing to include essential characteristics or by
including inessential characteristics ([2] A 732/B760). A synthetic definition, on the contrary, synthesizes a concept, i.e., it constructs a
concept by a combination of elements. Mathematical concepts make
the concept defined and so start with definitions. But in philosophy we
can never start with definitions since philosophical definitions have the
purpose of analysing given concepts. So "The definition, in all its
precision and clarity ought, in philosophy, to come rather at the end
than at the beginning of our inquiries" ([2] A 7311B759).
Kant first argued that there is nothing like synthetic definition in
philosophy in the important Prize Essay of 1764, 'Investigation of the
TRANSCENDENTAL REASONING
247
clarity of the principles of natural theology and morality', which is very

little altered in his Discipline of Pure Reason in its Dogmatic Employment, which appears in the Methodology of the first Critique. In the
Prize Essay ([1]2) he concentrates upon the role of signs in mathematics. He says that "Mathematics considers in its solutions proofs, and
infers the universal through the signs in abstracto" ([1]2, 278; cf.
[2] A714/B742). But the key passage is this:
Since the signs of mathematics are sensible means of knowledge, one can know with the
same confidence with which one makes sure of what one sees with one's own eyes that
one has not left any concept out of consideration, that every equation has been derived
by easy rules, etc.; thereby attention is made much easier in that it has to consider only
the signs as they are known individually, not the things as they are represented
generally ([1] 2, 291).
Mathematical signs are used in accordance with rules which we have

laid down, and operating with signs in this way, even without attending
to what they signify, is itself a sufficient guarantee of correctness (he
does not think that this is true of geometry ([1]2, 291) this being one
aspect of the Prize Essay which he changed his mind about later in the
first Critique).
But he did not alter his opinion about the uselessness of mathematical-type definitions in philosophy, and considers several different
kinds of non-mathematical concepts in an attempt at showing that such
definitions are either unobtainable or besides the point. For example, in
the first Critique he argues that no empirical concept of a natural kind
can be strictly defined. We can explain what the substances gold and
water are, but cannot define them since such concepts are always liable
to change where new observations and discoveries may add or remove
predicates from the concepts of such natural kinds ([2] A 728/B756).
Even pure concepts of the understanding, like cause and substance,
cannot have definitions which we are justified in thinking are complete.
These concepts are "given" to us in a confused form in perceptual
experience, and the parts of the definitions of such concepts can only
be imperfectly appreciated by the limited nature of our experience.
There is only one way in which we can be certain that our definitions
are right and adequate, and this is where we construct such concepts
ourselves. Since there are no self-evident synthetic propositions in
philosophy, there is no point in looking for axioms in the mathematical
sense in philosophy (as Wolff and his school did). The Leibnizian
248
OLIVER N. H. LEAMAN
approach to philosophy which tries to copy the methods of the

mathematician is invalid ([2] A713/B741, A725/B753). Only in trivial
and uninteresting cases in philosophy can a necessary connection
between two concepts be perceived by merely reflecting on them and
comparing them. With mathematical concepts, on the other hand, we
may construct in imagination an instance of the two concepts and
discern a necessary connection between them. Both mathematical
concepts and the a priori forms are known, but Kant claims that the
latter are only known problematically. This aspect of his distinction
between produced and given concepts places him on one side of the
present-day dispute about how to go about philosophy. According to
Kant, philosophers are obliged to define ordinary concepts in such a
way that they maintain a certain faithfulness to ordinary usage. He
criticises the attempt at constructing arbitrary and merely "grammatical"
definitions of philosophical concepts, at over-refining the essentially
confused.
Fortunately for philosophy, many of these confused concepts can be
partially rescued from confusion. He claims that,
With philosophical definitions ... the concept of a thing is already given, though
confusedly or insufficiently determined .... Everyone has, for example, a concept of
time; this is to be defined .... The mathematician deals with concepts ... but he takes
such a concept [space in general] as given in accordance with the clear and ordinary idea
of it ([1] 2,277).
Perhaps this suggestion might be taken up by Kant to defend his

transcendental arguments which seek to show that one philosophical
concept (e.g., an external object) is a necessary condition of some
activity (e.g., the determinability of the order of a series of impressions).
If all aspects of such concepts and activities (which are undeniably
given according to Kant) were confused and vague, then how could we
be sure that we had established the correct sort of conceptual connection between them? After all, future considerations might lead us to
revise our confidence in the perception of the connection as being of
one rather than another kind. This is what we might well expect to
happen in the case of unclear concepts. In reply, Kant might take up the
above comment from the Prize Essay, that just because a term is largely
unclear in the sense that no quite definite conceptual relations can be
established between all its aspects and those of other concepts, it does
not follow that it has no features which cannot be clearly related to
other clear features of similar terms.
TRANSCENDENT AL REASONING
249
Such a defence of conceptual connections between unclear concepts

seems well-founded. Locke notes that one may have a clear idea of a
relation without having a clear idea of the things related:
Thus, having the notion that one laid the egg out of which the other was hatched, I have
a clear idea of the relation of dam and chick between the two cassiowaries in St.
James's Park; though perhaps I have a very obscure and imperfect idea of those birds
themselves ([71 II. xxv 8).
Kant seems to adopt a similar view when he refers to transcendental

arguments involving the presupposition of a Grand Designer as arguing
from analogy in a particular sense, where "knowledge by analogy, which
means not, as the word is commonly taken, an imperfect similarity of
two things, but a perfect similarity of two relations between quite
dissimilar things" ([1]4, 357). But Kant does not show that such terms
have clear aspects, he rather assumes it. He is impressed, as was St.
Augustine whom he quotes, with the apparent paradox that "I know
what time is, but when someone asks me, I do not know it" ([1] 2, 283).
He is clearly of the opinion that there is no point in developing nominal
definitions of time and related concepts since possessing such definitions would not help us at all to understand what "time" means, and we
possess such an understanding without any evidence that we adhere to
such definitions. But this fact, upon which Wittgenstein also remarks
([10] para. 89), does not at all show that we all share the same basic
clear meaning of "time". The penumbra of properties which surround
this meaning may be different in each case for different people (as
Russell suggests ([8])). There may even be inconsistent features present
in the basic sense of words like "time", i.e., a feature which contradicts
another part of the meaning. In the latter case we should be using an
incoherent concept as though there were nothing wrong with it. We
might regard Berkeley as emphasising the looseness and consequent
inconsistency of our use of the concept of a material object. We apply it
both to what we think causes our impressions and those impressions
themselves. Pointing to the use of a concept as loose or vague is one
traditional way of challenging its credentials. Kant calls the "first and
chief rule" of metaphysical certainty not to start off with definitions of
concepts and then derive the consequences of those definitions. Yet
one advantage of such a method is that it will attempt a demonstration
of the consistency of a concept. After all, non-mathematical concepts
may have no exact definition, and yet nonetheless be shown to be
consistent.
250
OLIVER N. H. LEAMAN
Kant claims that empirical concepts are indefinable, and so indeterminate with respect to mathematical concepts. His definition of an
object of experience, though, is given in terms of the determination or
synthesis of a manifold in intuition. Once the elements that make up
experience are combined according to rule they make up an object, and
the rules only determine an object if they combine the experiences in
one way rather than another, so that they constitute one object or
sequence rather than another. The synthesis of a manifold in intuition
shows that an object can only be that kind of thing if it is determined in
just that way, and when undetermined the content of the intuition gives
us no idea of what characteristics the phenomenon possesses. But the
determinacy of the object of experience is only a relative determinacy;
it is relative to our point of view. Kant contrasts this limitation to a
point of view with Leibniz's notion of completeness. As we have seen,
Kant argued that a complete concept can only be given of invented
concepts. The seeds of this idea are to be found quite clearly in Leibniz.
Leibniz held that contingent propositions are only fully understood by
God, since any a priori proof of them would comprise an infinite
number of steps. We are essentially limited to a particular point of view,
or we have available to us a limited number of points of view, whereas
God knows the world from every possible point of view. It could be
said that God has no point of view, since any change in the distribution
of confusion involves a change in point of view and he is said to
perceive everything with complete clearness. It might be said that for
Leibniz's God all representations are perspicuous: nothing is hidden.
There is no confusion in the manner in which he maintains every
substance in existence by emanation, in the way that we produce our
thoughts. God is aware of the world in two ways not given to those
whose awareness is limited to a part of the world. He knows it as it
would appear to every possible point of view, as every possible monad
would perceive it, and he knows it in a way peculiar to himself ([5] sect.
14).
It is important to grasp what the nature of his perspicuity is meant to
be, since it will be used by Kant as the chief contrast notion to our own
way of knowing. According to Leibniz, when a subject-predicate expression (whether necessary or contingent) is true, the predicate is said to
be included in the subject, so that the concept of the predicate always
follows from the concept of the subject. Yet in a proposition like "All A
is P" the predicate "is a P" is not necessarily part of our understanding
251
of the concept of A and need have no part in determining its extension.

A complete understanding of the proposition as a fact involves an
infinite analysis of the concept which only God can carry out. In order
for us thus to understand any contingent fact, we should require
knowledge of all the parts of the infinite universe and the infinitely
complex relation of this fact to everything else in the universe. God is in
a position to understand the general principles which generate the
infinite series of events in the universe in the same way that we can
grasp a rule which generates a convergent series of infinitely many
terms. But grasping such a rule does not, for Leibniz, give us complete
knowledge of the contents of the series. In any case, he thought that the
idea of reaching the end of an infinite series is impossible, since he
erroneously identified such a position as only representable by an
infinitely large number - a possibility which he rejected.
There are many points of resemblance between the concept of the
thing in itself and the complete concept of an object. We are not able
to know this complete concept, but we can nevertheless think it.
Frequently Kant says that we cannot know things in themselves, but we
are not debarred from thinking of them (see his section on Phenomena
and Noumena, passim). It would not be entirely correct to say we can
think about the complete concept of a thing, since even our thoughts
are limited to a particular point of view. Finite rational beings can only
conceive of the principle of the complete concept of a thing: no
intuitions are receivable by us to determine such a concept. That is why
Kant says that from our point of view the thing in itself is indeterminate,
whereas from the point of view of a purely intelligible intelligence it
would be a perfectly determinate concept ([2) B307). We cannot render
such a notion theoretically determinate, yet there are concepts which we
cannot theoretically establish but which, as members of the intelligible
world, we can establish practically. Within the context of moral action,
for example, such principles are no longer thought indeterminately nor
problematically. Once we recognize that the possibility of acting
morally involves these principles, we definitely know that the intelligible
world and its contents are real "... from the practical point of view,
and this determinateness, which would be transcendent (extravagant)
for theoretical purposes, is for practical purposes immanent" ([1] 5,
105). Yet how can a term in this way both be indeterminate when used
in one context and become perfectly determinate when the context is
altered? After all, the cases which Kant has in mind here have nothing
252
OLIVER N. H. LEAMAN
to do with disambiguation, or shifts of meaning taking place relative to

context and emphasis. What is even more confusing is that we are often
told that we can form no determinate concept of the thing in itself, and
now we are informed that such a concept is not only possible but also
necessary with respect to certain of our most important practices.
To take another example, the principle of searching for unity makes
objectively real the unity of the "objects" (i.e., the theories of phenomena) of nature ([2] B693). This is only true from a particular point
of view, that of the "special interest of reason" ([2] B676), and so is
useless in establishing such propositions as that the world actually
possesses such unity, or that it displays economy, diversity, etc. Even
though it is permissible, and even necessary for the notion of the unity
of scientific theories, to regard the laws of nature as though they
displayed teleological characteristics, this cannot help us appreciate
God's purpose in forming nature in the way in which he has. We cannot
move from our conception of final ends in nature to "a determinate
conception of an intelligent world-cause" ([3] 470). Kant suggests that
we can only have an indeterminate conception of an intelligent cause
but we are allowed to take a quite definite view of the laws created by
this cause. Perhaps he thinks that while the concept of the cause itself is
unclear, this does not prevent the concepts of the effects of the cause
from being quite clear. The unity of the principles of mechanism and
teleology cannot be established determinately, but only accounted for in
the sense of establishing the possibility of both principles. The conclusion must be that if we are to seek determinate connections between
concepts with the status of noumenal and phenomenal concepts, as
Kant surely did, we can only connect clear aspects of the concepts with
each other rather than the entire concept with equally unclear entire
concepts.
Kant does not distinguish accurately between two aspects of his
theory of indeterminacy. In the first place there is the relatively
uninteresting (for Kant) type of indeterminacy in which we might in
future wish to revise our notion of the meaning of a term, either
because we have made new discoveries about the referent of the term,
or because we had made the term more clear. (These are the sorts of
cases which interested Waismann ([9]).) These are cases where we
receive an incomplete set of intuitions and so cannot determine an object
completely. From the point of view of transcendental idealism, this is
the case for all our non-mathematical concepts. There are in addition
253
other sorts of cases in which we cannot receive intuitions at all, where a

term is indeterminate because "it can never be definitely cognized or
clearly specified for employment in particular cases as they arise"
([3] 412). Concepts are problematical if they can have no intuitions
corresponding to them. For instance, we have no experience of a
supersensible ground, nor could we, in spite of the fact that it has the
function of making possible the co-existence of apparently incompatible
notions of the unity of nature. We cannot say that such a ground exists
nor even what its properties are without any intuition of such an object.
When Kant calls the concept of such a ground indeterminate he means
that it cannot be a possible object of knowledge for us, and even if we
regard it as such we still cannot treat it like an empirical object. We
cannot ascribe perceived properties of the object to it in the way we do
with possible objects of experience.
Experience will not help us come to any decision concerning the
properties of such an object, and without experience we shall have no
more reason to apply one predicate than another - which is why such
objects are called indeterminate. The law of bivalence is empty in the
realm of pure reason, we cannot say that a judgment is true or false
since we cannot empirically establish whether a judgment applies or not.
It is then rather misleading of Kant not to differentiate more exactly the
varying uses he has for the notion of indeterminacy. Calling the
supersensible ground indeterminate suggests that it is an empirical
object which shares its characteristic of indeterminacy with other
concepts of empirical objects, in contrast to the determinate nature of
mathematical concepts. But more is involved in the indeterminacy of
concepts of noumena than just dealing with given as opposed to
invented concepts. The thing in itself is indeterminate from our
(theoretical) point of view. From the point of view of the intuitive
intellect, however, such an object is perfectly determinate. This is one
of the contrasts of which Kant was so fond, and which he thought most
useful in explaining the nature of our point of view and its limitations.
Intellectual intuition is described by Kant, in keeping with the Leibnizian
notion, as a form of knowledge distinct both from human understanding
or discursiveness and sense intuition. As a form of intuition it is a
representation which stands in direct relation to its object. It creates its
own object - "it can provide itself with the existence of its object"
([21 B72) - so that "if all that is manifold in the subject were given by
the activity of the self, the inner intuition would be intellectual"
254
OLIVER N. H. LEAMAN
([2] B68). It has concepts which Kant calls "synthetic universal" as

opposed to our own "analytic universals", since the former themselves
determine which objects fall under them ([3]405-10). We are dependent upon objects presenting themselves to us in appearance; we rely
upon sensibility for any knowledge of objects to be possible. The
highest level of knowledge accords, on the Leibnizian account, with this
description of intellectual intuition, where knowledge is clear and
distinct, adequate and intuitive. Knowledge is perfect and so intuitive if
at every stage of a proof we are explicitly aware of the analysis of the
simple ideas appearing in the proof. It will be recalled that it is this
aspect of mathematical concepts, their perspicuity, which Kant claims
secures all intuitive inference against error and sets them off from
philosophical proof ([2] A 734/B762).
Kant argues that the categories have no meaning for the intuitive
understanding, since they are merely rules for an understanding which
brings the synthesis of the manifold given to it in intuition to the unity
of apperception ([2] B135, 138-9, 145-6). Now, this suggests a
question. How can even an intuitive intellect be conscious of itself
without some form of consciousness? After all, Kant is quite consistent
in his claims that "Relation to the identity of the subject does not exist
because I accompany every representation with consciousness, but
because I join one to another and am conscious of the synthesis of
them" ([2] B 133). It may be that an intuitive understanding spontaneously produces the contents of its own consciousness, but it cannot
evade the necessity that its experience should be unified. Kant draws
the contrast incorrectly, since it is not the contrast between an understanding which can do without principles of unity and one which
cannot, but between an understanding for which such unity is immediate and an understanding which is obliged to elaborate gradually and
imperfectly the unity of experience which consists of items which come
to it diversely, and with features missing which are necessary for the
understanding of their unity. We are to think of a consciousness which
would not be limited to awareness of a part only of the totality of what
exists, but would be conscious of the whole in a single uninterrupted
activity, in one grasp of apprehension. This apprehension of the unity of
the whole would not involve synthesis, the way of knowing appropriate
to the understanding whose consciousness is piecemeal and must
construct the unity of its experience. This latter consciousness is subject
to temporal conditions at every stage of which it apprehends only part
255
of the whole. Our understanding, as opposed to the divine understanding, is finite and does not possess an immediate recognition of its
own unity. Kant's distinction should be between two different kinds of
unity, not one kind of unity and something entirely other than unity.
This contrast between tbe intuitive intellect and the finite consciousness
tben can be used to bring out more clearly the nature of tbe limitations
of our point of view.
One might wonder, as Kant often does, how we could talk about the
intuitive intellect or the divine understanding, since is is clear that it is
not an object of possible experience. We might then insist that not even
the modal categories apply to it, since the categories apply only to what
is given in experience. The synthetic trutbs like the Axioms of Intuition,
the Analogies of Experience and the categories are necessary "relative
to tbe possibility of experience" ([2) A217/B264). To say that such
truths hold only of possible experience is to say that we can know
propositions like
(x) (x is an extended magnitude), and
(x) (if a state of x changes, that change in x has a cause)
to hold only if the values of tbeir variables are explicitly restricted to
empirical objects. But Kant argues in the Dialectic of the first Critique
and throughout the other two Critiques that the notion of God is
logically possible, and the use that can be made of it as a presupposition of the search for unity in nature and the possibility of moral action
being fully rational entitles us to regard it as actual and necessary as
well. The Fourtb Antinomy is supposed to prove that the notion of God
as creator of tbe world is free from contradiction, and so possible. Kant
often repeats this conclusion, that the concept of God is not selfcontradictory and that we may make it clearer by analogous reasoning,
yet it is evident tbat he does not in tbis way seek to derive the God of
religion. The last few chapters of the third Critique are particularly
explicit and almost violent on this point (see, for instance, the description of praying to God as "idolatry" at [3) 459). When Kant claims that
a particular concept of a thing in itself, such as God, the causa
noumenon, immortality, etc., acts as a necessary condition of the
possibility of adopting a certain point of view, he does not mean that
the entire concepts of these things in themselves must be presupposed.
That is why his discussions of concepts like that of God seem often so
256
OLIVER N. H. LEAMAN
hollow. For instance, at one point he suggests that the meaning of the
regulative Idea of God is to be understood in terms of "divine wisdom"
or "a principle of its [nature's] systematic and purposive unity"
([2] B727). We are supposed to use just those aspects of the concept of
the thing in itself which are necessary conditions of what we seek to
justify, and ignore the rest of the concept which has no such use. We
are thus forbidden to "allow an exacted presupposition to masquerade
as a free insight" ([ 1] 8, 13 7).
There is, then, good reason to think that Kant felt himself only
entitled to postulate the possibility of the existence of such things in
themselves, in the same way as in the Refutation of Idealism only the
possibility of an external object's existence is presupposed. His claim
that we are entitled to think of things in themselves implicitly follows an
aspect of Leibniz's approach to the ontological argument, in which the
latter would accept the Anselmian formulation of the argument only
with the addition of a proof of the freedom from logical contradiction
of the concept of a most perfect being ([1]1, 386). Kant thought that
such a proof could be provided by the unschematized categories, which
can be applied to anything at all ([2] A254/B309). It is only when these
principles are mediated by the form of space and time and thereby
related to possible experience that they are limited to possible objects
of experience. As we have seen, Kant has two notions of possibility, one
of which is the logical notion which calls possible any proposition or
concept which is in accordance with the principle of contradiction. He
argued that the Wolffians did not realize that this principle is merely
logical and abstracts from all considerations of the relation between
thought and its object. It is a logical principle which determines our
concepts of objects and so does not extend our knowledge of the
character of such objects. In the Amphiboly Kant challenged the
Leibnizian philosophy for identifying our concepts of things with the
things themselves:
Leibniz erected an intellectual system of the world, or rather, believed that he could
obtain knowledge of the inner nature of things by comparing all objects merely with the
understanding and with the separated, formal concepts of its thought [leaving] sensibility ... only a confused mode of representation ([2] A2701B326).
Kant devised a system in which

the condition of the objective employment of all our concepts of understanding is
merely the mode of our sensible intuition, by which objects are given to us ([21 A286/
8342).
257
He claims with respect to Leibniz and his school that previous

metaphysics was involved in "a transcendental amphiboly, that is, a
confounding of an object of pure understanding with appearance"
([21 A270/B326).
It is important to realize that one way of characterizing the difference between Kant and Leibniz may be expressed in terms of their
different views on determinacy. Leibniz thought that metaphysics
should consist of an investigation of concepts which are quite determinate, and since the most determinate view of the world that can be
adopted is God's, then the system of objects should be analysed from
God's point of view. God has organized the world in such a way that
the existence and nature of everything in it can be deduced from a few
supremely rational principles which represent the ground or explanation for the existence and properties of objects. Any more restricted
view which we adopt is a result of our being limited to a particular,
human, point of view, but even though we may not be capable of fully
grasping God's system and the place of everything in it, we can be sure
that behind what we regard as indeterminate there lies something quite
determinate, and what we see as inexact is really just our confused
apprehension of what is exact. Now, Kant's account is presented within
this context, and yet is radically different. For Kant, the non-mathematical concepts which we use are not indeterminate because we are
restricted in our point of view and so miss what is determinate about
such concepts. There is nothing precise "behind" the use of such
concepts, although there may be precise features of them. It is not as
though such concepts referred to objects, the hard and fast contours of
which the concepts imperfectly picture. Kant obviously thinks that
philosophical explanation in terms of such concepts is perfectly adequate, and that the implied Leibnizian move from the looseness of a
concept to its unsatisfactoriness is invalid. As he points out, this
Leibnizian move involves regarding objects of possible experience as
though they are logical objects, or things in themselves. Kant argues
that there is a vast difference between an object of experience and a
thing in itself. But Kant was enough of a Leibnizian to think that we
could only talk about being limited to a point of view if we can
conceive of not being limited to a point of view, and that talking about
objects as appearances presupposes being able to contrast them with
what a complete or perfect view of them would reveal.
Kant actually develops this point when he presents an account of the
indeterminate concept which characterizes aesthetic judgement. He uses
258
OLIVER N. H. LEAMAN
the indeterminate concept to suggest that starting from different empirical premisses would lead to different conclusions about what the
conditions of possible experience would be. Aesthetic judgement is said
to be based on indeterminate concepts because in a type of aesthetic
contemplation no specific concepts are involved. We react to the
aesthetic object freely, where the imagination and understanding operate
in a free way in producing aesthetic pleasure. Our pleasure in beauty is
precisely the result of viewing an object as coming under some concept
or other. That is, we are free to see a work of art from different points
of view; some concept must be employed, but it is indeterminate which.
It is possible to have this free reaction because it is possible to see the
object from anyone of an infinite possible points of view. From such
points of view we cannot derive information nor produce objectively
valid judgements (although we do make claims for the assent of others).
This is why Kant thinks that art is the expression of the basic but
unavailable concept of man. The best means we have of grasping this
concept is through whatever it is that produces images of changing
patterns of purposive wholes without an underlying concept of purpose.
It is important to emphasise the freedom with which we construct such
patterns; there is nothing necessary about the sort of pattern we apply
to the object of aesthetic contemplation. We can produce no generally
valid theoretical propositions about the object as seen from such points
of view. The contingency of each act of aesthetic contemplation, in the
sense that it necessarily contains no particular concept, represents the
contingency of the human substratum. Art reveals what man is, but
never finally, because the fact that we have to use indeterminate
concepts means that nothing universally valid is produced, nor could it
ever be produced. The use of indeterminate concepts in aesthetic
appreciation reveals the lack of determinate knowledge we can have of
human potentialities, and so of the sorts of experience persons could
have. A note of caution should be introduced at this point. Kant is not
arguing that there are no limits on the type of experiences which we
could have of objects - he would be very much opposed to such a
thesis. He is arguing that our reactions to certain kinds of aesthetic
objects is such that no determinate concept is to be used as a description of such objects.
If our concept of man were determinate and perspicuously related to
other concepts, then we should not be limited in our point of view. The
determinate concept of man would be the same for anyone seeing it
259
from any point of view, and aesthetic objects would completely be

determined to have one set of characteristics rather than another.
Leibniz also influenced Kant considerably in his theory of aesthetic
experience. According to Leibniz, the source of our pleasure in
aesthetic experience is our clear and confused perception of a unity in
variety. (Knowledge is clear if it enables me to recognize just what is
represented, and confused where I am unable to enumerate all the
various marks which distinguish the object known from other objects,
even though the object may possess such marks.) As Leibniz claims:
Pleasure is the feeling of a perfection or an excellence, whether in ourselves or in
something else.... For the image of such perfection in others, impressed upon us,
causes some of the perfection to be implanted and aroused within ourselves ([6J 425).
The source of our pleasure in aesthetic experience is our clear but

confused perception of the perfection of a unity in the sorts of varieties
which confront us in the world. Every perception is a greater or lesser
realization of the universal harmony which we represent on the model
of the world as a whole combining the simplest hypotheses with the
richest compos sible variety of phenomena. Our knowledge at each
stage is characterized by the harmonious resolution of apparent incompatibilities. The successful work of art emulates this best of all possible
worlds in achieving the greatest expression of a felt unity in variety
compossible within the receptivity of the world which the artist tries to
fashion.
Now, Kant famously holds that an important aspect of the judgement
of taste is its complete independence of the idea of perfection. The
harmonious interplay of our cognitive powers that characterizes such
experience is based upon a purposiveness without purpose. Objective
purposiveness or the conceptual reference of the object to a criterion of
what it ought to be, has no place in the judgement of free beauty. Kant
argued that we are not capable of discovering which metaphysical
principles characterize the world. We can, however, think about such a
possible characterization and use it as a regulative principle in the
construction of science, but we cannot claim to know that it describes
anything. His account of aesthetic concepts brings this out: the variety
of points of view means that there is a variety of possible responses to
what is seen from such positions, and the Leibnizian theory that
appreciation involves a feeling or perception of a perfection (a perfection which may, moreover, be objectively defined) ignores this aspect of
260
OLIVER N. H. LEAMAN
variety. If we had a concept of perfection and insisted that our aesthetic

feelings must be directed towards it, then we could only account for the
variety of possible aesthetic responses which people actually do make
by calling them mistakes. We need not accept that they are mistakes,
according to Kant, because the idea of objective purposiveness itself is
not a comprehensible object of aesthetic contemplation. We can accept
different responses to the same object, because our selection of concepts to bring to the object is subjective and indeterminate; ultimately,
this procedure reveals the contingency and indeterminacy of our
concept of man.
In a similar vein, Wittgenstein remarks:
Concepts with fixed limits would demand a uniformity of behaviour. But where I am
certain, someone else is uncertain. And that is a fact of nature ([10J para. 374).
Given the actual nature of our behaviour, we should not represent it as

though it were, or should be, more precise that it really is. What we
might say in particular contexts, about beauty or even knowledge for
instance, need not agree with what others would say. The approach of
Kant and Wittgenstein here is to discuss the nature of knowledge and
language from our point of view, not from the point of view of a perfect
observer. Kant argues that one way in which we can grasp the diversity
of possible points of view which are different manifestations of human
essence is through examining aesthetic contemplation and the indeterminate concepts which that involves. Wittgenstein also considers that
examination of the facts which underlie our system reveals that our use
of indeterminate concepts is appropriate to our behaviour and purpose.
For neither philosopher, though, do the facts justify our concepts.
Wittgenstein refers to certain very general facts which are such that
were they to change, then we would probably be led to accept different
concepts and different systems. This is because the ways in which we
behave are related to our perception of such important facts, and the
purposes behind our actions (such as measuring, for instance) would
not be realized were such facts not to hold. It does not follow, though,
that such facts enter into our language games, or that we believe in their
truth, for it to be correct to refer to them as grounding our systems. We
may well be unaware of them even though our systems would be
useless were they to be different. It is the whole point of asking what
relation holds between such facts and our systems that we are led to
unravel some of the ways in which our practices are not unique, but
depend on the world being a certain way:
261
If the formation of concepts can be explained by facts of nature, should we not be
interested, not in grammar, but rather in that in nature which is the basis of grammar?
- Our interest certainly includes the correspondence between concepts and very
general facts of nature.... But our interest does not fall back upon these possible
causes of the formation of concepts; we are not doing natural science. . . . I am not
saying: if such-and-such facts were different people would have different concepts (in
the sense of a hypothesis). But: if anyone believes that certain concepts are absolutely
the correct ones, and that having different ones would mean not realizing something
that we realize - then let him imagine certain very general facts of nature to be
different from what we are used to, and the formation of concepts different from the
usual ones will become intelligible to him ([ II] II. xii).
We should not take it that just because a change in such very general
facts has been noted that our practices must change - after all, we may
not consider such a change as being important. The point of discussing
such facts is to see what changes in our practices become intelligible. It
is not as though our practices are justified by their correspondence to
certain facts, until perhaps it is discovered that such facts no longer (or
never did) hold. When such facts break down we often, although
certainly not always, have a choice as to how to go on. There is no
general purpose with which our practices must accord, and so these
facts are not in general to be regarded as justifying such practices.
Such a state of anarchy would not be permitted to exist from a
perspicuous point of view, since the nature of both the world and our
concepts would be perfectly well known. There would be no room
for the uncertainty and various possibilities which exist in our limited
point of view with respect both to the world and to our concepts.
Wittgenstein's many discussions of the loose and yet somewhat obscure
relationship between the facts and our concepts brings out this point
well. If the notion of our point of view is to be taken seriously, and not
merely as something which requires analysing away into something
more perspicuous, then the indeterminate view we have both of the
world and of our concepts must be taken seriously too. On the
perspicuous point of view it is clear that the beliefs which are held are
certain and based upon something rational and necessary, such as the
metaphysical character of reality, and it would be illogical to challenge
such beliefs. The physical world, in Leibniz's view, has no reality in
itself but is merely a confused representation of something whose
essence could be apprehended by pure intellect alone. The monad is
not a sensible but an intelligible entity, and so could be viewed as the
ultimate subject of predicates. Reason can inform us of the principles
which characterize the world, and so we may class the confused view of
262
OLIVER N. H. LEAMAN
the world which we often have as an inaccurate representation of the

truth. To return to the aesthetic case, we may then disparage those who
are at the stage of being unable to appreciate the perfection which
characterizes excellent art:
Likewise we sometimes see painters and other artists correctly judge what has been
done well or badly; yet they are often unable to give a reason for their judgement but
tell the inquirer that the work which displeases them lacks "something, I know not
what" ([61291).
The supremely rational basis which Leibniz provides for our criteria
of aesthetic appreciation, and for our investigation of the nature of
reality itself, is not available to us in Kant's view. From our limited
point of view only a very restricted class of propositions are logically
undeniable, and these do not significantly extend our knowledge in the
ways that Leibniz thinks possible. In addition, most of the propositions
which we call certain are based ultimately on an empirical fact, namely,
that we have certain characteristics. These characteristics and our
training lead us to accept that some propositions stand fast, while
others are dubitable. Yet as Wittgenstein argues in On Certainty, the
contingency of a proposition is no bar to its certainty and basicity in a
system. Such propositions may be doubted, but all that doubting does is
produce
concepts the possibility of which is altogether groundless, as they cannot be based on
experience and its known laws; and without such confirmation they are arbitrary
combinations of thought, which although indeed free from contradiction, can make no
claim to objective reality ([21 A223/B270).
Kant distinguishes between real and logical possibility, where a concept

of an object is really possible if it can "agree with the formal conditions
of experience", thought and intuition ([21 A218/B265). A concept is
logically possible if it does not contain contradictory predicates, but the
concept may be without an object and so only an "ens rationis"
([21 A292/B348). Kant could then claim that the sceptical attack upon
the certainty of what we take to be knowledge which has the form of
pointing out that no contradiction arises from the denial of such
knowledge claims is correct in so far as it is based upon concepts which
are only logically possible. But logical possibility has a purely general
reference, and what we require if we are to deal with concepts which
refer to knowledge of objects are concepts which are really possible,
263
and which therefore possess the characteristics outlined by him in the

Analytic of the first Critique. Similarly, Wittgenstein distinguishes
between false beliefs and mistakes, between challenges to our epistemic
claims which are logically possible and those which are in addition
provided with a comprehensible context. Of course, it it always logically
possible to question any such epistemic claims, but Kant would not
count all such questioning as relevant to such claims, since not all such
questioning is really possible. Concepts can be shown to be really
possible and objectively real even though it is logically possible to
challenge them. As Wittgenstein pointed out, in a particular case
mistakes may be ruled out, yet it is still possible that we have a false
belief.
So basing our practices on facts need not be unsatisfactory.
Wittgenstein goes too far in his refusal to accept that such basic facts
can then be justified or known. He appears to think that if they are not
independently justifiable, then they are not justifiable at all. Kant
realizes that they can be justified, but not a priori. They have no
independent justification, it is true, but they have an empirical justification which, given our limited point of view, is all that we can provide.
Unlike the rationalists, Kant thought that we could not base our
practices on any metaphysically sure foundation:
The concepts of reality, substance, causality, even that of necessity in existence, apart
from their use in making possible the empirical knowledge of an object, have no
meaning whatsoever, such as might serve to determine an object. They can be employed,
therefore, to explain the possibility of things in the world of sense, but not to explain
the possibility of the universe itself. Such a ground of explanation would have to be
outside the world, and could not therefore be an object of a possible experience
([2J A677 18705).
But it does not matter that the whole system of transcendental

argument depends upon the world being a certain way, i.e., upon the
specific character of human beings. Kant is quite ready to admit that the
starting point of his arguments is of this contingent nature, and he
argues that the starting point cannot be of any other type, given the
limited nature of any possibly human point of view. What makes the
conclusions of transcendental arguments necessary, if such arguments
work, is the logical relation between the presuppositions and what they
make possible. Such arguments will be "uniquely" valid of any practice
which requires a presupposition to be acceptable as a possible and
rational practice. If the practice changes, as well it might if the
264
OLIVER N. H. LEAMAN
characteristics of the agents change, then different presuppositions will

be involved, and these too will be "uniquely" valid with respect to the
new practices. Kant is unambiguous in his view that since we cannot
specify an eternal and fixed human essence, we cannot tell how human
beings might change and develop, nor what categories they might
require in the future. The conclusions of transcendental arguments are
not supposed by Kant to hold for ever whatever concept of human
being is used; such conclusions are derived unashamably from a
particular notion of a person with a particular way of achieving
knowledge, acting morally and aesthetically. Thinking that Kant was
interested in any stronger sorts of argument could result only from a
failure to appreciate the strong emphasis in his thought on the notion of
indeterminacy and its importance in articulating what it is to be limited
to a point of view. Justification must take place within the same limited
context as every other human activity, and cannot transcend the bounds
of possible experience.
REFERENCES
Preussische Akademie der Wissenschaften (Berlin: Reimer and de Gruyter, 190283).
2. Kant, Immanuel, Immanuel Kant's Critique of Pure Reason, tr. N. Kemp Smith
(London: Macmillan, 1929).
3. Kant, Immanuel, Critique of Teleological Judgement, in Critique of Judgement, tr.
J. C. Meredith (Oxford: Clarendon Press, 1952).
4. Korner, Stephan, 'On the Kantian foundation of science and mathematics', KantStudien 56 (1966), 463-73; reprinted in The First Critique, eds. T. Penelhum and
J. Macintosh (Belmont CA: Wadsworth, 1969),97-108.
5. Leibniz, G. W., Discourse on Metaphysics, tr. P. G. Lucas and L. Grint (Manchester: Manchester Univ. Press, 1953).
6. Leibniz, G. W., Philosophical Papers and Letters, ed. L. E. Loemker (2nd ed.,
Dordrecht: Reidel, 1969).
7. Locke, John, An Essay Concerning Human Understanding, 2 vols, ed. A. C. Fraser
(Oxford: Clarendon Press, 1894).
8. Russell, Bertrand, 'Vagueness', Australian Journal of Psychology I (1923),84-92.
9. Waismann, F., 'Verifiability', Proceedings of the Aristotelian Society Supplementary
Volume 19 (1945), 119-50.
10. Wittgenstein, Ludwig, Zettel, tf. G. E. M. Anscombe (Oxford: Blackwell, 1967).
11. Wittgenstein, Ludwig, Philosophical Investigations, tr. G. E. M. Anscombe
(Oxford: Blackwell, 1953).
MICHAEL K. POWER
BUCHDAHL AND RORTY ON KANT AND THE

HISTORY OF PHILOSOPHY
Abstract. Richard Rorty's recent challenges to the professional philosopher flow from
his distinctive interpretation of the modern tradition. He argues that its deepest and
most questionable presupposition is the notion of the mind as a mirror of nature - a
notion that requires "deconstruction" if philosophy is to return to the good faith of its
ancient role. Against Rorty I argue that his programme cannot be sustained in the
context of his account of Kant. I draw upon Buchdahl's reading of the Kantian text
which seems to offer an alternative "constructivist" path between the extremes of
foundationalism and relativism. Buchdahl's idea of philosophy as a multi-levelled and
hermeneutic enterprise underpins his model of the structure of Kant's thought. In turn
this model facilitates an interpretation that cuts though a number of the Gordian knots
that have obsessed Kant scholars - particularly those of the "thing-in-itself" and of the
concept of matter. On closer inspection there may be little here that Rorty's pragmatism
can dispute.
1. INTRODUCTION
Generalizations about the history of philosophy inevitably contain

implications for its present and future development. Indeed the more
compressed and aphoristic the view the more critical may be the
challenge to the current self-understanding of philosophical research.
For example, we are traditionally taught that modem philosophy begins
with Descartes who was the first to write with a clarity and rigour that
is the basis of its development today. In stark contrast there is the idea
(attributed to A. N. Whitehead) that philosophy is but footnotes to
Plato. Or perhaps the history of philosophy is a history of attempts to
end philosophy given that numerous treatises purport to offer the final
word? It could be said that Wittgenstein had two quite different
attempts at this. Taken together these and other snapshots of the
history of philosophy offer an ironic reminder to each other of the
open-ended character of historical interpretation. In this essay I wish to
dwell on two such images of the modem tradition offered by Richard
Rorty ([12]) and Gerd Buchdahl ([3]).
To talk of images in this sense is important if one follows the
hermeneutic insight that access to an intellectual field requires the prior
265
266
MICHAEL K. POWER
understanding that a succinct metaphor may be able to provide.

However the implication is not that understanding is merely imagerelative, thereby undercutting the possibility of criticism. Indeed, in the
context of the interpretations of Kant offered by Rorty and Buchdahl, I
shall argue that Rorty's metaphor of the mirror of nature, while
powerful in many respects, is unsatisfactory.
2. THE MIRROR OF NATURE
It is tempting to look back upon the history of philosophy as a quest for
solutions to "perennial" issues. Confidence in such a view has recently

been undermined by Richard Rorty (see also [13]). This confidence, he
argues, crystallized after Kant and gave shape to "professionalized"
philosophical research thereafter. While thinkers such as Descartes and
Hobbes are regarded paradigmatically as philosophers, they could not
have seen themselves in this light. In their own terms they were
contributors to research in mathematics and science. Rorty claims that
it is only after Kant that the distinction between philosophy and science
emerges and that the concept of a "theory of knowledge" as a discrete
subject gathers momentum:
once Kant had written, historians of philosophy were able to make the thinkers of the
seventeenth and eighteenth centuries fall into place as attempting to answer the
question "Howis our knowledge possible?" ([12]132).
Specifically, Rorty regards the revival of Kantianism in Germany

following the widespread rejection of the Hegelian framework as crucial
in the development of the professional self-image of the philosopher. In
his view, the turn from Hegel's legitimate and refreshing historicism was
a vital episode allowing the problem of "representation" to take hold as
a non-empirical project independent of physiological discovery.
Rorty's history is not, as MacIntyre ([8]) has argued, primarily a
sociological thesis concerning the way in which philosophy departments
have fallen prey to general trends toward bureaucratization and professionalism: rather, it is based upon a view internal to the history of
philosophy ([9]). At the heart of Rorty's analysis are his claims concerning the metaphor of mind as a "mirror" of nature.
The model of the mirror of nature implies that the mind or self has a
status that is ontologically distinct from its environment. According to
Rorty it is Descartes who is largely responsible for inventing this
BUCHDAHL AND RORTY ON KANT
267
notion. By conceptualizing a domain of "inner space" the way is open

for the Kantians to develop epistemology:
The Cartesian change from mind-as-reason to mind-as-inner-arena was not the triumph
of the prideful individual subject freed from scholastic shackles so much as the triumph
of the quest for certainty over the quest for wisdom ([12[ 61).
Rorty claims that it is a short step from the "invention" of the mind as
an inner arena to the notion that such a mind can know itself and its
inner contents with (quasi-mathematical) certainty. By contrast, knowledge of an external world, a world that the mind's ideas purport to
represent or "mirror" faithfully, cannot match this degree of certainty. It
is the difference between these two modes of knowing that generates
the problems of scepticism. If all that the mind can truly know is its
own immediate contents, then how can it claim to go beyond? How can
we be sure that its ideas accurately represent the world as it is?
According to Rorty most modern philosophy addresses this problem
of "accurate representation" in one form or another. In contrast, those
thinkers who reject or deconstruct this central concern come to
recognize that philosophy has no special set of problems which give the
discipline any systematic unity. Rorty utilizes this schematic division
to categorize the leading historical figures. For example, the later
Wittgenstein, Heidegger, Nietzsche and Dewey were not centrally
concerned with the problem of accurate representation. Of their work it
has often been asked "But is it really philosophy?". The work of Hume,
Kant and Carnap does not evoke this question because it deals with a
recognizable set of philosophical issues at the heart of which lies that of
representation.
It is interesting to note that Rorty's categorization cuts across some
traditional distinctions between "Continental" and "Anglo-American"
philosophy. For example he concentrates upon the similarities between
positivists (e.g., Schlick) and transcendentalists (e.g., Husserl) which
serve to distinguish them from those features shared by pragmatists
(e.g., Dewey) and structuralists (e.g., Derrida). This provides a stimulating challenge to prevailing intuitions about the differences between
the Continental and Anglo-American traditions. However, the cutting
edge provided by the image of the mind as a mirror of nature is not
self-justifying and requires support from more localized analysis. I
propose therefore to consider Rorty's interpretation of Kant. While this
is essentially a separable discussion and not intended to prejudge his
268
MICHAEL K. POWER
detailed claims elsewhere, it remains of general importance for at least

two reasons. Firstly, Rorty's history of philosophy relies heavily upon
an interpretation of Kant, whose work he regards as the most influential
statement of the "mirror of nature" orientation. Secondly, Rorty argues
that it is Kantianism which is first to invest professional respectability in
philosophy thereby demarcating the theory of knowledge as a specialist
Fach in its own right.
Rorty argues that some version of a distinction between "content"
and "scheme" is central to the Kantian programme which he understands to be realist in intention ([11 D. "Content" is another term for the
"given" in contrast to the idea of "scheme" as that which organizes the
raw data to yield knowledge/experience proper. An initial difficulty
with Rorty's strategy at this point is that he rather loosely assimilates
the scheme!content distinction (which he borrows from Davidsonian
semantics) to various Kantian distinctions such as those between
concepts and intuitions, understanding and sensibility and the a priori
and the a posteriori. The scheme!content distinction cannot possibly
capture the nuances of these oppositions and while such a loss of local
detail might be permissible for the sake of the grander argument, the
importance of Kant in Rorty's reconstruction of the history of philosophy militates against this. For example, he sometimes talks as if the
scheme aspect of cognition actually creates the content. However, Kant
was at great pains to distance his position from this kind of generative
or dogmatic idealism ([6] B 274-5).
Notwithstanding such interpretational difficulties Rorty offers a blunt
and powerful challenge to the Kantian philosophy. He launches a
Deweyian form of attack on Kant's notion of "experience in general"
which, he argues, disguises its highly specialist nature. Such a notion has
its place in what Rorty calls the "epistemological language game". The
terms of the game and the rules for their employment are internal to the
whole Kantian enterprise. By implication epistemology is a game that
we can simply accept or reject - there are no arguments to compel us
to play it.
Deliberately echoing Wittgenstein, Rorty proceeds to question the
authority of key elements in Kant's architecture:
The notions of synthesis and the concept-intuition distinction are thus tailor-made for
one another, both being invented to make sense of the paradoxical assumption that
runs through the first Critique - the assumption that manifoldness is given and that
unity is made (112] 152-3).
269
It is true that Kant seems to offer no arguments for his claim that
"combination" does not come to us through the senses. It is more a
central assumption in his system around which other claims are
constructed. Rorty pursues the point by arguing that the whole idea of
synthesis postulates an inner space in which the mechanics of combining take place. Thus Descartes had invented an inner space which
Kant further develops by ascribing to it an active role in the constitution of objectivity. However,
Why should we think sensibility "in its original receptivity" presents us with a manifold,
a manifold which, however, cannot be represented as a manifold until the understanding has used concepts to synthesise it? ([ 12]153).
We cannot appeal to introspective evidence to justify such an

assumption and everyday experience seems to show that objects come
to consciousness fully combined as unities. What therefore is the
justification for Kant's assumption that atomicity is given? As Rorty
puts it "we want to know whether concepts are synthesizers" ([121154).
Rorty's own assumptions come to the surface at this point since for
him any talk of "synthesis" would only be meaningful within some
testable causal story about the human organism and its interaction with
the natural world. He argues that this is of no special philosophical
standing but is merely to follow where our current scientific theories
take us.
3. BUCHDAHL: INTERPRETING OBJECTIVITY
Buchdahl's interpretation of seventeenth- and eighteenth-century philosophy differs markedly from Rorty's. The problem of accurate representation implicit in the mirror of nature metaphor is supplanted by the
problem of intelligibility ([21 passim). Questions about the meaning of
the central concepts of the new sciences of mechanics are conducted at
the level of a meta-discourse. Rather than the quest for indubitable
foundations, Buchdahl offers the image of an emergent two-way
hermeneutic between science and philosophy in which each of the
central thinkers proceeds by way of model and analogy to sketch a
basic grammar of objectivity consistent with scientific discovery. Rorty
alludes to the self-defining nature of the epistemological language game
but this implies that it is a closed unity without connection to other
areas of intellectual endeavour. Such a view thereby understates the fact
270
MICHAEL K. POWER
that modern distinctions between mathematical, logical and physical

propositions were not yet clear. According to Buchdahl, the philosophy
of this period is characterized by the need to create a discourse within
which to negotiate these distinctions. Thus a notion such as "synthesis",
which from Rorty's standpoint is contradicted by psychological fact, has
a legitimate place within these attempts to explicate the language of the
new sciences. In contrast to Rorty's self-styled deconstructionism,
Buchdahl offers what might be called a constructivist interpretation of
the history of philosophy:
Progress in the development of philosophical history - in so far as this idea applies will go on record as an increased degree of self-consciousness accompanying the
construction of different systems ([2) 5).
Nowhere is this counter-image to Rorty so forcefully expressed than in

the interpretation of Kant.
Buchdahl requires an heuristic distinction between what he calls
"phenomenology" and "ontology" in order to generate his interpretation
of Kant. While both terms have an established history, they are given a
distinctive twist when read into the Kantian text. To talk of phenomenology in Buchdahl's sense is to view objectivity naturalistically (for
Rorty this is the only intelligible status it can have). Borrowing the
language of Husserl, Buchdahl locates phenomenology at the level of
the lebenswelt. The latter is not merely the locally shared common
sense world of medium-sized objects, it is also the world of theoretical
physics. The breadth of such a concept might threaten its sense were it
not for the notion of ontology to which it is counterpoised. At the level
of ontology, inquiry is not concerned with the actual existence of
objects but with the question of their possibility. According to Buchdahl
([3] 41) Leibniz's question as to why there is "something rather than
nothing" is a paradigmatic instance of this form of inquiry. The general
point is that it is a question that cannot be construed as an appeal to the
natural science of the day. Its more profound task is to interpret the
status of the objects with which natural science deals.
The distinction between phenomenology and ontology so described
is quite fundamental to Buchdahl's interpretation of Kant. In marked
contrast to Rorty, Buchdahl states that
the danger here is that one may simply be telling how it really is. Instead of ferreting
out a descriptive metaphysics one ends up with a descriptive physics ([3) 41).
271
The heart of Buchdahl's account of Kant's transcendental method is

to be found in his notion of a Reduction-Realization Procedure (RRP).
In Wittgensteinian terms the RRP may be regarded as an attempt to
characterize the "grammar" of objectivity. Buchdahl claims that the
RRP is merely a more explicit refinement of a structure whose contours
are already clearly visible in the Kantian text. "Reduction" is the
intellectual process by which the "possibility" of the world is placed in
doubt in abstraction from its phenomenological actuality. This process
of "bracketing" or epoche bears a strong resemblance to Cartesian
methodological scepticism.
Reduction in this specialist sense is simply a method of interpreting
objectivity. Indeed to talk of an object in a "reduced" sense is simply to
offer one interpretation of the world. It is not to posit another realm of
objects behind the scenes. Kant's "transcendental object" is not therefore a hazy entity which bears a mysterious causal relation to the
common world of appearance. It is an important nodal point in the
Kantian scheme. This is the meaning of Kant's condensed claim that
the concept of an object in general taken problematically is the supreme concept with
which to begin a transcendental philosophy ([6] A290/B346).
The opposite of "reduction" is "realization" under which the bloodless

interpretation of the object in a reduced sense is built up again. The
details of this process are to be found in Kant's arguments for space
and time as conditions of the possibility of sensible experience and for
the categories of the understanding as conditions of experience as such.
The fundamental "critical" move is to reinterpret the object as an entity
whose nature is "to appear". Hence the significance of the faculty of
sensibility in the Kantian structure as receptivity to an object which
appears.
It is difficult to describe Kant's position in a natural language since,
as has been stated above, it is a strategy whose purpose in part is to
illuminate the deep structure of that language. Buchdahl's RRP model
provides a quasi-dynamic representation of the argument structure.
One of the most fruitful applications of the RRP model is to the
problem of the thing-in-itself. It is important to accept that all this talk
of reduction and realization is indeed something of a game. There is
only one world subject to a process of interpretation and in this sense
the thing-in-itself is not a mysterious kind of object to which we are
incapable of having cognitive access. Buchdahl draws attention to
272
MICHAEL K. POWER
Kant's idea that human agents could only be capable of such access
counterfactually, i.e., if they were capable of intellectual intuition. This
counterfactual notion retains the logical form of rationalism without its
metaphysical substance. It only seems to posit a non-sensible realm of
"bjects until it is understood as Kant's explication of a necessary
component in the language of objectivity.
While objectivity is interpreted as that which appears, the "esse" of
objects is not solely "percipere". As is well known, Berkeley was forced
to modify his radical idealism and to utilize divine perception to
characterize reality. In other words he recognized that the ontological
status of objects could not be exhausted by their appearing relative
solely to human perceivers. Kant's notion of the thing-in-itself must be
understood as playing a similar role in serving to characterize that
aspect of experience for which we are not responsible, i.e., its givenness.
Appearing is always the appearing of something.
Buchdahl's distinction between phenomenology and ontology is
particularly important in making sense of this notion if it is not to
collapse under the almost Heideggerian pressures placed on the discourse. It can be said that in experience we are affected by objects.
Thus subjects do not (normally) create the experiences they undergo
but are affected by them. At the level of phenomenology such "affection" can only be understood as an oblique reference to the possibility
of a causal account being given of the physiological transactions
between perceiving subjects and a natural environment. While Locke
tries to invest this level of inquiry with metaphysical significance
Buchdahl wishes to distinguish it from Kant's position at the level of
ontology. Affection in this ontological sense is an aspect of objective
cognition as such. Part of the deep structure of objectivity is its capacity
to "affect" or to appear to SUbjects. Grammar appears to commit us to a
mysterious something that does the affecting, but Kant's intention is to
disregard any metaphysically realist interpretation of such a something,
retaining only its essential (transcendental) logic.
On the view expounded, the issue of the timelessness of the foundations of knowledge is secondary to the programme of providing a deep
interpretation of objectivity. In fact this is none other than a commitment to explicate the pragmatist's notion that "[m)ost of the world is as
it is whatever we think about it" ([13) xxvi). From Rorty's perspective,
attempts to supply interesting philosophical accounts of this claim are
riddled with confusion. From Buchdahl's perspective this is too super-
273
ficial: it is the "facticity" of the world as celebrated by pragmatists that

demands an interpretation.
4. CONSTRUCTION OR DECONSTRUCTION
In contriving a confrontation between Buchdahl and Rorty on the basis

of their interpretations of Kant it is evident that each would regard the
burden of proof as lying with the other. This is to some extent a
product of differences in style. For Buchdahl, there is no faithful point
of access to the Kantian text without a model of whole such as the
RRP. Despite Rorty's own allegiance to elements of the hermeneutic
tradition, his stylistic commitment is, in contrast, analytic.
Rorty's critique of the "epistemological language game" might equally
well be directed against Buchdahl's notion of ontology. To put it
crudely, what is the point of a reduction as a prelude to realization if we
are back where we started? Wittgenstein has said that "philosophy
leaves everything as it is" and it does seem true that one cannot define
Buchdahl's notion of ontology independently of questions of the form
"how is knowledge/experience really possible?". That is to say we do
not accept the contours of this problem unless much else is assumed,
e.g., the "givenness" of atomocity/manifoldness and the related distinction between concepts which have an intellectual origin and intuitions
which are sensible. From Rorty's point of view the hermeneutic significance of traditional philosophy is that of a closed circle, and his
pragmatism is simply a determination to stand outside it. Only a
particular model of the mind and its relationship to nature ever
persuaded us to enter the circle.
Does the interdependency of key terms commit us to an appraisal of
Rorty's kind? Buchdahl's interpretation of Kant suggests that the circle
is very much open. One way of demonstrating this is to take up Rorty's
claim that either the notion of the "world" is a Kantian thing-in-itself
and therefore empty, or it is "well within some particular theory about
how the world is" ([14]14). Stated thus, the idea is that a "descriptive
physics" is our final word about a world which is:
either the purely vacuous notion of the ineffable cause of sense and goal of the
intellectual or else a name for the objects that inquiry at the moment is leaving alone
([141 15).
The challenge is stark and powerful yet, necessarily perhaps, overstates
274
MICHAEL K. POWER
the choice of interpretations. Buchdahrs distinction between questions

of ontology and of phenomenology could be said to correspond to
those asked about experience and those asked within experience. The
former (metaphysics) are typically posed by philosophers whereas the
latter (physics) are not. If my reading of Buchdahl's approach is correct
then questions about experience are a special mode of question internal
to experience, and this seems entirely consistent with the pragmatist
viewpoint. In modern terms one could argue that Kant's notion of the
thing-in-itself is compatible with Putnam's "internal realism" ([10]). The
key issue, it would appear, is whether such a notion is "vacuous" or a
label for the objects that inquiry is currently leaving alone. However,
this dichotomy has its source in Rorty's image of normal science. The
idea of being "well within" current theory discloses both an implicit
naturalistic bias and an understatement of the nature of conflict in the
natural sciences. For example, Rorty claims that if we abandon the
dualism between scheme and content upon which Kant's arguments
depend then we are left with unmediated contact with the familiar
objects of the lifeworld. As Hesse has stated, it is pertinent to ask
whose objects these are, whose familiarity is to count and how is
unfamiliarity to be negotiated ([5]). If Rorty's pragmatism seeks to
deconstruct a specialist employment of the concept of experience, then
in turn it disguises the manner in which conflict may generate the need
for meta-discourse about the objects of experience. That epistemological
inquiry emerges in this way from specific issues in the interpretation of
scientific concepts seems to be a claim that is consistent with Rorty's
historicism. However, Buchdahl's interpretation of Kant attempts to
show that this is not merely to relativise epistemology. To understand
this idea imagery is quite essential. A quotation from Cassirer contains
a suggestion:
the question was no longer the discovery and confirmation of individual, definitive facts
but rather a fundamental conflict in the interpretation of recognized phenomena ... ,
here not just isolated observations and data but the principles on which the examination
of nature is founded and their diverse areas of jurisdiction had to be weighed against
one another. Kant always formulated his particular question in the light of this general
task ([4] 27).
This movement from specific to general questions could be represented

by the image of a spiral without beginning or end. Just as there is no
ultimate point of foundation neither is there an ultimate point of
275
relativisation (effectively the mirror image of foundationalism). Of

course such a metaphor must prove itself and Kant's discussion of the
concept of matter is an appropriate context.
In the preceding discussion, attention has effectively been focused
upon Kant's "general metaphysics" of nature, i.e., his theory of experience/knowledge in general. In the Metaphysical Foundations of Natural
Science ([7]) Kant develops what he describes as a "special" metaphysics of nature. The difference between these two levels of the
Kantian enterprise is a critical issue in Kant scholarship ([2) 672). In the
special metaphysics Kant addresses the problem of a realist interpretation of "matter" as conceived by Newtonian physics. Berkeley and
Hume had purged matter of secret powers but Kant could not accept
the implied reduction of force-discourse to motion-discourse. According
to Brittan there are two strands to Kant's strategy ([1) 133-4). Firstly,
as an extension to the arguments in the refutation of idealism Kant
claims that spatial extension alone is insufficient as a criterion of the
existence of external objects:
we are acquainted with substances in space only through forces which are active in this
and that space, either bringing other objects to it (attraction) or preventing them
penetrating into it (repulsion and impenetrability) ([6] A265/B321).
The second strand of Kant's position is concerned directly with the

problem of action-at-a-distance. It is well-known that Locke and Clarke
regarded action-at-a-distance as unintelligible despite its inductive
support. For Kant this problem of intelligibility is a problem of real
possibility. That is, while action-at-a-distance is not a logical impossibility or contradiction as some had maintained, predictive confirmation
was not of itself adequate.
For Kant the problem of intelligibility is resolved by "constructing"
the concept of matter in order to generate a distinction between it and
empty space. This is tum involves the construction of two fundamental
forces of attraction and repulsion. Construction means something like
"ostensive construction" ([6) A7171B745) or "display in intuition". It is
essentially mathematical and illustrated paradigmatically by the spatial
addition of velocities ([7) 492). Kant's problem is that he does not
believe forces can be constructed in this way (although he concedes
that one day they might be!) An approach to understanding his difficulty is to distinguish metaphysical from mathematical construction.
The former is the conceptual unpacking process such that the possibility
276
MICHAEL K. POWER
of displaying a concept as a spatial magnitude is demonstrated. A

mathematical construction is the spatial representation itself. Thus
while a mathematical account of forces in the required sense can be
given, e.g., a parallelogram of forces, Kant cannot find a conceptual link
between space and forces, i.e., the link between forces and acceleration
cannot be cognized a priori.
The account given above may be somewhat crude. However, the
intention is not to contribute to existing studies of Kant's position in the
Metaphysical Foundations but to draw out its general features as a way
of continuing the interplay between Rorty and Buchdahl. The following
points can be made. Firstly, whatever else can be said of Kant's position
it can hardly be assimilated to the representationalism that Rorty
attacks. Secondly the need for an account of the concept of matter such
as to make action-at-a-distance intelligible arises from within normal
science.
A pragmatist is unlikely to be impressed by these points since he
may not accept the problem of "real possibility" that is being addressed.
The idea of real possibility is a modality which is intended to be distinct
from logical possibility. This is particularly crucial for Kant for whom
the critical reorientation ties real possibility to the limits or boundary of
sensibility. In contrast his pre-critical rationalism effectively conflates
real and logical possibility. Brittan ([1]13) and others connect logical
possibility with the notion of truth in all possible worlds. Accordingly
the analytic/synthetic distinction is then a distinction between truth
in all possible worlds and actual empirical truth. Against this, real
possibility concerns the limited set of worlds of which we could have
experience. The Quinean challenge can then be read as stating that we
cannot say what a really possible world must be like and at different
times and for difference purposes the limits of what is really possible
may ebb and flow. This is surely something that Rorty would accept.
We identified above two sources of Kant's account of the concept of
matter. The difference between the two could be said to correspond to
the difference between Kant's general and his special metaphysics. The
former is concerned to explicate a concept of matter in general that
squares with the experience of impenetrability. The latter addresses the
real possibility of matter as conceived by classical physics. What is the
relationship between these two levels? As Brittan ([1]137) puts it, one
of the difficulties with Kant's concept of matter is that it is intended to
playa number of different roles. As an a priori concept it corresponds
to the otherness of experience. As an a posteriori concept it neces-
277
sitates a construction to show how it is really possible. What Brittan

calls weak and strong concepts of matter effectively correspond to the
distinction between Kant's general and special metaphysics. Reichenbach
([1]137) makes a similar distinction between matter as that which is
there over and above the forms we impose on it and matter as
construed by classical physics. However, the analytic/synthetic and a
priorila posteriori dualisms cannot adequately generate all the distinctions that Kant requires. For example, mathematical construction and
synthesis are both a priori, yet they operate at different levels of Kant's
arguments. With charitable modem hindsight we can read this not as a
confusion but in the spirit of Quine's holism. What is a priori is
therefore relative to its place in the network. Specifically, construction
is a priori at the level of the Metaphysical Foundations of Natural
Science, yet it presupposes the general account in which synthesis is a
priori. The determinate concept of matter as conceived by classical
physics and reconstructed by Kant presupposes the indeterminate
concept of matter that is a necessary element of experience. If this
suggestion makes sense it has important consequences. Firstly, given
that there are enough passages where Kant is explicitly empiricist about
science to outweigh the notorious rationalist statements; he cannot be
read as attempting to deduce Newtonian physics a priori since the
relationship between the various levels of his philosophy is presuppositional and not deductive. Buchdahl calls this a "looseness of fit"
between such levels. Therefore the explication of the concept of matter
at the level of a general metaphysics makes possible, and is a precondition of, the special account but does not necessitate it. This is true
despite the obvious fact that the general account is modelled or
structured by the concerns of Newtonian physics. Secondly, the metaphor of the spiral structure in transcendental philosophy gains plausibility on this holistic reading of Kant's methodology. It is the relationship of presupposition which links the elements of the spiral.
It could now be argued that this account is only as strong as the
notion of presupposition on which it relies. Brittan ([1] 36) has drawn
attention to the importance of distinguishing presupposition from
deduction and of not attempting to model the former upon the latter.
Consider the following modus tollens:
A presupposes B
not B
Therefore not A
278
MICHAEL K. POWER
If presupposition is construed as implication this does not hold.

Similarly:
A presupposes B
notA
ThereforeB
If one takes this as a schematization of "not-A is true" entails "B is
true" then this is valid as a presuppositional but not as a deductive
argument. This at least establishes some preliminary plausibility for a
distinctive notion of presupposition which is not reducible to implication. A more detailed discussion is beyond the scope of this essay but it
is worth stressing that even if a logic of presupposition holds together, it
is the content given to its symbols which conveys the substance of any
argument. For Kant's critical philosophy it is the idea of "capable of
being experienced by us" which guides that content. This has the effect
of moulding Kant into a verificationist until one recalls the specific
manner in which he requires a construction of the concept of matter in
sense experience. The concern for intelligibility or real possibility would
be unlikely to impress a verificationist for whom the phenomenological
fact of action-at-a-distance would probably be sufficient.
5. CONCLUSIONS
I began by reflecting upon the role of images in providing an overview

of the history of philosophy and pursued this in the context of Rorty's
and Buchdahl's interpretations of Kant. Rorty's deconstructionist challenge to the Kantian enterprise flows from his image of the mind as a
mirror of nature. Yet it is a challenge that is unconvincing for the
following reasons. Firstly, it is not ultimately plausible to mould Kant's
position in the Critique of Pure Reason into a form of representationalist philosophy. Buchdahl places the quest for intelligibility at the
centre of Kant's concerns which allows a reading of the continuities
between the pre-critical and critical work in a way that Rorty does not.
One need only read Kant's Third Analogy ([6) A2111B256) in parallel
with the Nova Dilucidatio to recognize this. Secondly, it is not possible
to claim that Kant's epistemology, for all its flaws, is a vacuous selfdefining game since its concerns can be seen to emerge from the
science of the day. This emergence does not merely relativise the
Kantian project because it does not seek to provide a deductive proof
279
of the laws of classical physics. Kant's philosophy is a system of

presuppositional logic which may be construed holistically and thereby
as an interpretative rather than as a foundationalist project.
Rorty's remaining ploy is to claim that he is simply not interested in
the Kantian approach. He says as much by implying that transcendental
philosophy is simply an attempt to give an interesting account of those
objects that science is not currently questioning. This is a weaker
argument against both Kant and the self-image of philosophy. An
interesting footnote to this is provided by Rorty's didactic claims for the
future of philosophy as the quest for new vocabularies. Following the
spirit of Buchdahl's account, Kant's transcendentalism represents an
attempt to forge a vocabulary about objectivity. That Rorty finds Kant
vacuous says much about his own assumptions. One is tempted to
conclude that the only telling difference between the transcendentalist
and the pragmatist is the former's lack of reverence for common sense.
REFERENCES
1. Brittan, G., Kant's Theory of Science (Princeton: Princeton Univ. Press, 1978).
2. Buchdahl, Gerd, Metaphysics and the Philosophy of Science. The Classical Origins,
3. Buchdahl, Gerd, 'Reduction-realization: A key to the structure of Kant's thought',
Philosophical Topics 12 (1982),39-98.
4. Cassirer, E., Kant's Life and Thought, tr. J. Haden (New Haven: Yale Univ. Press,
1981 ).
5. Hesse, M., 'Epistemology without foundations', in Philosophy, Its History and
Historiography, ed. A. J. Holland (Dordrecht: Reidel, 1985),49-68.
6. Kant, Immanuel, Critique of Pure Reason (1st ed. 1781; 2nd ed. 1787), tr. N.
Kemp Smith (London: Macmillan, 1933).
7. Kant, Immanuel, Metaphysical Foundations of Natural Science (1786), tr. J.
Ellington (New York: Bobbs Merrill, 1970).
8. MacIntyre, A., 'Philosophy and its history', Analyse und Kritik 1 (1982), 101-15.
9. Nielson, K., 'How to be sceptical about philosophy', Philosophy 61 (1986), 8393.
10. Putnam, H., Meaning and the Moral Sciences (London: Routledge and Kegan Paul,
1978).
11. Rorty, Richard, 'Transcendental arguments, self-reference and pragmatism', in
Transcendental Arguments and Science, eds. P. Bieri and R.-P. Horstmann
(Dordrecht: Reidel, 1979), 77-103.
12. Rorty, Richard, Philosophy and the Mirror of Nature (Oxford: Blackwell, 1980).
13. Rorty, Richard, Consequences of Pragmatism (Brighton: Harvester, 1982).
14. Rorty, Richard, 'The world well lost', in 1131 3-18.
DWIGHT C. BARNABY
THE EARLY RECEPTION OF KANT'S

METAPHYSICAL FOUNDATIONS OF
NATURAL SCIENCE
Abstract. As the traditional criteria for the evaluation of hypotheses in terms of the
conceivability of their "inner nature" came into question towards the end of the
eighteenth century, Kant proposed in his Metaphysical Foundations of Natural Science
a new set of rational standards for the design of scientific explanation. Although this
work attracted a wider audience than has previously been suspected, it was poorly
understood, typically being caricatured as an a priori assertion of what must exist in
nature. The misconstruction of Kant's philosophy of science stemmed from radical
mistakes about the transcendental setting of the fundamental structures of the Critique
of Pure Reason, which prepared the way for Kant's views later to be merged with those
of the succeeding philosophical movement, Naturphilosophie.
The interest of the present study is to trace the historical influence of

the Metaphysical Foundations of Natural Science (MFNS) in Germany
in the thirty years after its publication, concentrating on the many
misunderstandings of its transcendental position in the whole of Kant's
system, rather than on its influence in scientific practice. Since these
misinterpretations pulled Kant's ideas into the current Naturphilosophie,
that philosophical movement inspired by Schelling and immediately
following the Critical System, the hidden sympathy between Kant's
position and Schelling's, especially as it was historically revealed, will
constitute our central problem. Because the earliest confusions about
Kant's theory were very similar to the popular misunderstanding of his
views even today ([40]205f), it will be necessary first to consider the
actual meaning of the MFNS before investigating its reception. Because
a false picture of the Critique of Pure Reason (CPR) caused much of
the difficulty in understanding the MFNS, we must first examine the
foundation of the Critical Philosophy.
Kant's contemporaries were obsessed with an epistemological dualism
whose question was how separate and self-contained self and world
fruitfully come into conceptual relation, even though their respective
"natures" are thoroughly opposed. Self and world, informally determined by the traditions of philosophy and of the ordinary belief
community, were taken as givens, the necessary starting points of all
philosophical analysis. Epistemological questions thus had to be
281
282
DWIGHT C. BARNABY
answered by an "ontological geography", "locating" stages of the

explanation of knowledge in self or world, according to their imagined
ontological character, however poorly this was defined. The result was
more the picture of a solution than an actual answer, since with the self
and world opaque or at best informally sketched, the key actors in the
argument were themselves inaccessible to investigation. It is then not
surprising that the typical result of such considerations was that
between a self and a world initially defined as opposed, no bridge could
be fashioned, and scepticism was the predetermined consequence. The
analogy between the eighteenth-century philosophy and science is thus
clear: informal ontological traditions, irrelevant to either discipline and
vaguely formulated, constrained the growth of epistemological and
scientific argument, burdening both with entities or ontological criteria
resistant to rational development. Although after Wittgenstein, a reconstruction of self and world so as to permit their obvious relation to be
explained seems an easy reply to this dilemma for epistemology, the
rejection of the pseudo-problem of fundamental dualism by the CPR
was in its day so radical an invention that Kant himself could only
formulate it in a mythic tale of the thing-in-itself, appearance, and a
variety of aspects of the knowing subject. The Critical Philosophy
begins not with irreconcilable endpoints but with their relation, conceived as experience-as-unity, as objectivity.l This experience is then
subjected to an analytical reduction, in which the conditions of its
possibility as the coherent world of a consciousness receptively affected
are removed in sequence, being referred either to subject or object,
accordingly as they are spontaneous or given, and to concept or
intuition, as they are discursive or immediate ([9]). Distinct self and
world are thus only first made by the analysis, not given at the outset,
so there can be no problem of their relation, at least not at this most
basic level, especially since they are both built by us out of a common
root, experience-as-unity. Since the unpacking of the elements constitutive of objective experience is a deliberate, theoretical undertaking, the
stages of the reduction should not be pictured as generating real things,
whose own interrelation would then in turn come into question, unto
vicious regress, as though the CPR were an empirical psychology,
describing merely the physically independent steps by which objects
come to consciousness ([29] A85ff/Bl17ff, A346fIB404f). The project
of the CPR can fail by abstracting the structures constitutive of
objective experience too loosely, so that, when reconstructed, they no
KANT'S METAPHYSICAL FOUNDATIONS
283
longer searnlessly add up to objectivity; but it cannot fail because these

abstracted structures defy it, so that their connection cannot be found
because the real nature of the elements remains occult. The world is
characterized as "appearance" to give it an initial ontological deficiency,
bringing it into dependence on the self and so assuring that its
analytical reduction of it will yield elements of both mind and world in
relation ([29) A128f, A165f1B204f). The main achievement of this
analysis is that self-awareness and world-awareness both first emerge
to "visibility" through the same conceptual structure, standing between
both, as it were, (and not "warehoused" in the mind of the individual
subject!) factoring each to such determination and unity that it makes
sense to distinguish a stage of the world and a world, a subject and an
object, and so to have that experience we know we have: experience as
coherence, our starting point. Because this argument is so similar to
those of Wittgenstein and Strawson, it is tempting to assert that this
middle conceptual structure, the categories, is public language or the
determining influence of a belief community, but Kant never associates
the categories with anything real, although followers of his such as
Dilthey ([12) 19, 44) and Simmel ([72) 20) were much later to attempt
this.
Instead for Kant the categories are defined only in reference to his
reduction procedure ([9]). When experience is reduced to the point
where it is nothing more than- a fugue of sensations, that ordering which
elevates them only to the level of determinate succession is called the
work of the categories. This provides us with nothing but a fixed array
of sensation which then, at a much higher stage of the gradual
rebuilding of experience to that form in which it is ordinarily encountered, must be further fashioned into the world described by language
and science. The concepts of the categories have only functioned at the
aboriginal stage as models of determinate order; in generating this
order they have not further built themselves into the ordinary world as
concepts, but are only braces for stabilizing the given. Science and
theory of any sort will require a second application of the categories,
only this time as ordinary empirical concepts operating with purely
hypothetical force ([8) 133; [29) A126ff, A159/B198, A269f1B325f,
A356, B164f; [32) 495), like any empirical description. This is in
contrast to their initial application, which was conceived as an irresistable imposition on a given at that stage so dimensionless that no
ordering of it could be said either to be supported or refuted by the
284
DWIGHT C. BARNABY
data. Because the categories proved their worth as data bridges, as

synthetic a priori devices, at this primordial level, they have now been
validated in this role, and so their use as concepts in and for the
ordinary world - where they cannot visibly be supported by the given
material at every point, but must project beyond it - is similarly made
plausible ([53) 586-9). But they are not for that reason necessary, and
indeed each use of them must face ordinary empirical tests. This now
answers the question posed at the beginning of the CPR, how synthetic
a priori statements, or metaphysics generally, may be possible: such
structures projecting beyond the direct endorsement of the data have a
relative warrant in their ordinary empirical use, since they were necessary for that initial ordering implicit in our having any coherent
experience at all.
But now even though Kant's system began as a monism of mind and
world, principally in order to dismiss from the outset the unsolvable
epistemological dualism of the day, and even if the first-stage operation
of the categories was conceived as constitutive, controlling the given
material without resistance, now, at the second-level employment of the
categories in ordinary experience, the gulf between subject and object
fully reopens ([29) A19/B33, A166f/B208f, A176/B218). Radical
scepticism about the possibility of any conceptual projection can
perhaps be combatted by reasoning that such projection is implicit in
our having any experience we can share, and a given set of concepts
have been shown to have such use in the foundation of experience; but
whether these will have any new, higher-level employment remains an
open question, depending on how concept-resistent experience proves
to be. While the mind/world dissonance has been tamed at the basic
level at which traditional epistemological dualism insisted on it (the
transcendental level), ordinary experience (the empirical level) remains
untamed, and the individual, empirical subject could still find himself
lost in a world proof against all sophisticated description. The grounding
of theory in this range of experience would thus have to remain merely
suasive, not necessitating, not directly derived from the results of our
analysis about the possibility of any experience at all, which concerns
only the possibility of determinate succession, and not of science.
It is here that the MFNS enters the picture. Recognizing that a
purely formal, instrumentalist science would be difficult to achieve
([29) A847/B875 note a), Kant wishes to argue for the real possibility
of certain types of hypotheses by showing them to re-use that set of
285
concepts already validated by their foundational employment, the

categories, and by exhibiting them as comprehensible in terms of those
spatial structures which are also used to construct nature from its
reduced to its realized state in the work of the categories. Since
empirical data are open to any number of interpretations, whose
universality and necessity can never be established solely by the data
acting to delimit this range, scientific explanation needs anchoring not
only below in the given but also above in a canon-guiding hypothesis
evaluation ([29) A85fIB117, A9lfIB124). This can be achieved by
supplementing the criterion for the acceptance of hypotheses already
available, logical consistency, by tests of the real possibility of hypotheses, defined as their experienceability in space and describability in
terms of the validated categories ([29) A159/B198, A222f/B269f,
A290ff/B347f, A370-3, A769ff/B797ff, A850fIB878f; [32) 474n,
563; [34) 325f). One model of such hypothesis-formation, surveying
the whole basic structure of the science, should be developed, yielding
in the process one set of "really" possible hypotheses consistently
arranged. Although the science under consideration is empirical, a
concept of it can be chosen and constructed a priori: defined, not
given, as the foundational structure for the science ([8) 141;
[29) A341ff/B399ff, A617/B645-A620/B646; [32) 470, 472f, 482).
That necessary status associated with scientific laws can now be
grounded, if not in fact in the data, then at least within the core
structure codified by this analytical process. For physics, this foundational device is the concept of matter as arbitrarily designed in the
MFNS.
But what is the status of such a system with respect to the actual
practice of scientific theorising? Kant is not, as he was sometimes
caricatured, a scientific deductivist, and for him nature remains the
arbiter of what is real ([29) A276/B332, A770IB798; [30) 168;
[32) 477). The MFNS's design of the concept of matter is a priori only
in the sense of being a deliberate, theoretical outline. Although its status
is more secure that that of an ordinary hypothesis, because the real
possibility of its features has been carefully referred to those structures
fundamental to all knowledge detailed in the CPR, it remains only
that: possible ([29) A17lf/B213; [32) 525, 532). Experience will have
to decide how valid for describing nature and how convenient for
organizing physics it may be. But even in his own scientific ventures,
Kant does not always explain strictly in terms of the designs set out in
286
DWIGHT C. BARNABY
the MFNS ([31) 221 nl). The actual is much richer in detail than the
possible, and nature may always surprise us with subsets within,
refinements of, or developments from the merely possible, in which
instances our framing of hypotheses would have to advance beyond its
basic foundation in the MFNS, following nature into its own manifestations of the possible ([29) A87/B 120, A92/B 124, A 127, A165f/B206f).
Indeed, the significance of a metaphysics of nature is essentially negative ([29) A847f/B875f), showing what explanatory strategies would
not be able to argue for their acceptability by pointing to their relation
to the foundational factors of experience; though "non-possible" hypotheses of this sort are not for that reason impossible, just as "possible" hypotheses are not for that reason necessary ([32) 505f, 524).
Having thus detached the MFNS from the higher, necessary structures of the CPR above, it is now equally important to separate it from
the specific explanations of empirical physics below; for just as the
MFNS is not an integral part of the transcendental argument, so too it
does not belong to ordinary physics. Even the design of matter
proposed as a model in the MFNS only serves as a canon for the
development of lower-level, detailed hypotheses for physics - not
logically deduced from the argument of the MFNS, but merely confident that the imitation of its forms will at least yield "really possible"
hypotheses ([29) A847/B875 note a; (32) 525, 532). Specifically, the
picture of material extension offered, as consisting of opposed attracting and repelling forces in equilibrium, is not to be taken as positing
these forces as things really to be found in matter, or anywhere in
nature; rather, the MFNS only describes how the filling of space by
matter can be conceived: simultaneous movement in opposing directions, frozen in equilibrium, yielding extension. Forces cannot in any
case be posited a priori, but can only be given in experience
([29) A20f/B35, A206f/B252, A648ff/B676ff). When Kant urges that
the goal of simplification of empirically-given forces be their reduction
to such fundamental forces as are employed in the conceptual characterization of material extension, this does not mean that those fundamental forces are the general, physical basis of all others. Rather it is
just that the fundamental forces, having already shown their role in our
understanding of the most basic concept of physics - matter - now
emerge as well-founded constructs for the rest of physical theory, and
so are clearly to be favoured as concepts guiding the simplification
of nature. Kant himself seems to have taken his notion of fundamental
287
force from a specialized use by Aepinus, in which that term designated

not a real force at all, but only a formal classification, offered as a first
step in organizing phenomena pending discovery of their real causal
mechanisms ([33] 254n). Thus such "fundamental" forces were strictly
conceptual and superficial, not physical and underlying. Almost all
contemporary critics of the MFNS complained that its two fundamental
forces were too thin a fabric to support all physical description; Kant
would have agreed. The foundational role of the MFNS with respect to
actual physics is thus like that of the CPR with respect to the MFNS.
The CPR establishes certain concepts as having possible synthetic
projection over ordinary empirical data by having shown that their
similar projection over the initial fugue of impressions with constitutive
force is implicit in the objective experience we have. The MFNS
establishes certain hypotheses as having possible use for explanation in
ordinary physics, by having shown their use in a design of material
extension (in terms of structures validated by the CPR) which yields a
concept central to all physical theory ([32] 473-6).
The particular object of the MFNS is to show that the concept of
attraction, so long rejected as inconceivable in itself, is rooted in the
very basis of our picture of nature, the concept of matter, and so its
conceivability, at least relative to the foundational commitments of
physics, is established. No longer can Newtonian attraction be effectively circumvented by a theory of material impact, since matter itself
entails, if not the physical force then at least the idea, of attraction
([32] 486f, 499, 51H, 518, 562). Thus with the MFNS Kant sets out
criteria for the evaluation of hypotheses relative to their setting in the
vast structure of commitments both essential and incidental to our
experience of objectivity and our understanding of nature, and so
provides an explicit theory where once only the crude opposition of
conceivable and inconceivable natures decided on the acceptability of
explanations (d. [36]168).
Before turning to the misunderstandings of Kant's position which
eventually led to its being conflated with a very distinct Naturphilosophie, it will be useful to assess its historical impact quantitatively. The
assessment of the MFNS as too difficult was nearly universal, and a
reluctance to discuss critically and creatively so imposing a work was
widespread ([3] 37f; [86]31). Even the usually opinionated Lichtenberg declined an offer to review it, excusing himself as not wanting to
enter its associated controversies, though he was quite possibly simply
288
DWIGHT C. BARNABY
daunted by its difficulties. 2 Various theories were advanced as to why it

had called forth so little response: Schwab, an opponent, thought
physicists reluctant to consider such a work, suspecting an a priori,
speculative system in it; because it fell between philosophy and physics,
students of neither discipline felt prepared for its challenges ([70]
i-iv).3 Jakob, one of Kant's supporters, wrote to him that "to hear
Klugel [Professor of Mathematics at Halle] speak about ... the Metaphysical Foundations of Natural Science is almost unbearable, and to
chemists and physicists it seems to be altogether incomprehensible how
one cannot be satisfied with proofs from induction and analogy".4 The
MFNS found not only perplexed sympathy but determined opposition
when defenders of atomism like Busse took to the field against it, "to
prevent the loss of time which young, talented men among my auditors
are inclined to devote to the Kantian system" ([10] xi). Yet although
each of the favourable commentators individually imagined himself to
be nearly alone in his interest in the MFNS, there were in fact more
than forty authors discussing it in Germany up to 1810. Johann
Schmitt, Professor of Mathematics at Heidelberg, was the first to
lecture on the MFNS in 1792-3, and by 1795 the influence of Kant's
philosophy had begun to spread more rapidly ([75] 184). By 1806 the
chemist and historian of science Trommsdorf could even assert that
"none of the rich epochs of philosophy has worked so profoundly on
chemistry as the Kantian, which as a purely critical endeavour promises
great results for the field of chemical analysis and its development as a
science" ([79] 3, 135). As was the case with the Critical System
generally, however, some of the supporters of the MFNS were its own
worst enemies. The institutional dissociation at this time of the natural
sciences from their foundation in the philosophical faculty led to a
search for new foundations, such grounds enhancing the status of
newly-independent disciplines, so the MFNS was often seized upon
with neither interest nor understanding as a useful prop ([75] passim).
Kant's technical terminology was sometimes employed merely to lend
the appearance of philosophical dignity to ordinary natural histories
([22]37-48) or undisciplined speculations ([14]) and sometimes to
conduct ordinary physics in the language of transcendental inquiry
([84] 369). In some quarters however the MFNS enjoyed the authority
of a dogma to be respected and even feared, so that Fischer revised his
dictionary of physics to incorporate Kantian notions ([15]179), while
Metzger apologized for his failure always to "harmonize with Kantian
KANT'S METAPHYSICAL FOUNDA TIONS
289
principles", since departing from them was something he "would not

dare to do" in his work on anthropology ([46] 6f).
Naturphilosophie, following on the MFNS by eleven years, first
opposing it, then co-opting it, confused its impact. Some Naturphilosoph en dismissed Kant as too "empirical" ([64]412), while others
mistook the MFNS as deriving from a Kantianism caricatured as
empirical psychology ([60] 149f). The contest between Kantians and
Naturphilosophen in German academia was vigorous and even "vehement" ([77]1, 116), and both movements, competing for the same small
community of philosophically-interested scientists and scientificallyliterate philosophers, saw in each other a more serious opponent than
in the surrounding empiricism ([47]53; [77]1, 117, 147f; 2, 33, 38,
60f). Steffens, a follower of Schelling, wrote to his mentor that the
University of Halle would be "fatal" to him, because of its "old, half-decayed Kantians who hate Naturphilosophie like the plague".5 But the
Kantians themselves, already unclear about many of Kant's own ideas,
proved unequal to the struggle with a new generation quite sure of its
commitments. Schelling saw himself as beginning his system where Kant
had left Off,6 being the natural heir to Kant's followers. Trommsdorf so
conflated these movements he was even convinced that Kant was the
father of Naturphilosophie ([79] 3, 134ff), while Graffe viewed Schelling
as the principle expositor of the MFNS ([23] iii-iv). One student wrote
in 1803 that he would only first attempt Kant after having read
Schelling, an inversion certain to render Kant's finer distinctions
invisible ([77] 2, 145). The drama and flamboyance of Naturphilosophie
against the discipline of the MFNS clearly counted to the advantage of
the former, making Kant's precision seem to represent a poverty of
imagination, for "Kant proceeds from postulates, only ... with the
difference that he has not, like Schelling, drawn them together into a
single colossus" ([77]1, 131). Schelling could even triumph in 1801
that Kantianism was already extinct ([60]147). Unfortunately for the
fate of the MFNS, this eclipse and absorption of Kant's philosophy of
science by Naturphilosophie came just as Schelling's movement was
itself about to fall to an empiricist reaction ([7] xxvii), and the MFNS
was discredited together with its former opponents ([10] xi-xii;
[39]562).
Discouraging as the story of the influence of the MFNS in philosophy may be, its influence within a scientific community mistaking it for
a work of physics rather than philosophy was considerable. Kant
290
DWIGHT C. BARNABY
achieved his most immediate object, encouraging the acceptance of

Newtonian action-at-a-distance ([68J 35; [87J 116ft). His sanction of
dynamical explanations gave physics new latitude in hypothesis-design,
and the MFNS contributed to a general formalizing of imponderable
fluids, no longer seen as automatically preferable to dynamical accounts,
and so helped establish the dynamical theory of heat ([21 J 546f;
[63J 168, 182f, 284f). The dissolving of absolute, self-subsistent material bodies as the ultimate explanatory units of physics into continuous
fields of action also first emerged under Kant's influence ([83J 60f). The
obsession of Kantian scientists with the model of forces in dynamic
equilibrium focused the attention of physics on those reflections
essential to the evolution of the concept of conservation of force, and
thus to that of energy ([78J 28f; [84J 372ft, 376f).
Because matter was now taken as itself the product of forces, the
great historical block to balancing the ledger of force transformations in
impact - i.e., the supposed absolute loss of force within inert matter was now removed ([63J 164). The most dramatic contribution to
science by the MFNS was its influence on Oersted's discovery of
electro-magnetism in 1820. While this is often attributed merely to
Oersted's confidence in the interconvertability of forces, a truism of the
time he could have drawn from any number of sources, in fact the
influence of the MFNS on him was more precise. The image of forces
acting directly on other forces without the mediation of matter,
balancing each other in empty space, sustaining in that opposition a
persisting occupation of space, was what prepared Oersted to interpret
his data, long obscure to the Newtonian mind, as he did ([50J 138ft). At
a time when forces were supposed simply to radiate from supporting
material particles, the independence of forces from matter, supporting a
complex and lasting configuration in space, was a picture available only
in the MFNS. 7
To investigate now how the MFNS was incorrectly located by its
interpreters within Kant's transcendental framework and so made ripe
for conquest by Naturphilosophie, it is first necessary to consider how
the Critical Philosophy was generally misunderstood. While Kant had
sought to give new transcendental location to established concepts such
as "subjective" and "objective", "ideal" and "real", his interpreters held
fast to traditional usages, constraining Kant's own argument to fit within
the old analytical perspectives. 8 That epistemological dualism which
291
Kant's initial perspective was intended to avoid was so entrenched in

the mind of his readers that they imposed its forms on the CPR. There
was a strong inclination to see the Critique as a work of empirical
psychology, taking an inventory of the powers of the mind of the
individual subject as they imposed themselves upon a variously resistant
nature ([5] 165f). The logical a priori of Kant's analysis was made into
the temporal a priori of a real process of conceptualization, operating
through time, so that the categories could only be innate ideas
([24] 34-7). The most serious error was to mistake the given on which
the categories act for ordinary perceptual experience, when in fact
determinate perception for Kant was only the result of their operation,
the initial affection of the "mind" having as its source only the most
indeterminate "promptings" and "occasionings," certainly not empirical
objects! ([35]71-2). As a result the interface between the thing-in-itself
and appearance, intended by Kant to mark the very edge of meaningful
experience, was imagined to emerge within ordinary empirical experience. The temptation to this mistake was almost irresistable, since
philosophy and science then operated under the assumption that all
conceptual limitation derived from the confinement of our knowledge,
of our perception, to the surface radiations of things, or to their
intellectual and sensible relations to us, leaving the underlying or inner
"objects themselves" - a lower realm occupied by real causes, inner
natures and substance platforms - forever obscure ([71]581, 596f).
The problem of knowledge thus became one of failed correspondence
to the ultimately real object,9 while the challenge to conceptual characterization of nature became the existence of real causes outside of a
mind which had in its grasp only their effects. Even Humean scepticism,
it was thought, could be overcome if only knowledge were not limited
to the result of the interaction of mind and world, and we could instead
penetrate to the realperse ([80]30-3; [81]42).
Since the work of the categories was typically promoted up one
level, to operate not in forming determinate succession out of a
dimensionless fugue, but rather to inject conceptual material into
ordinary perceptions, the whole array of Kantian structures was now, in
a domino effect, similarly moved one step higher. The thing-in-itself,
instead of remaining merely an empty token marking where knowledge
may not reach, moved up to become a significant opacity within
experience. Appearance was now the character of experience because
292
DWIGHT C. BARNABY
this was known only in its relation to the individual, perceiving subject.
The categories were confined to appearances because these were taken
as all that the mind can grasp. But now an essential problem for the
MFNS emerges from this transcendental dislocation. Because reason,
understanding, and the transcendental determination of fundamental
order, together with ordinary scientific theorizing, all now find themselves displaced into the same location - the world of appearances "in
the mind" - there is no way to introduce that re-opening of the gulf
between subject and object in experience, which was so important to
Kant's argument. Because the control or lack of control of concept over
object was typically explained not logically but ontologically, once the
first and second applications of the categories were both crowded into
the same ontological setting, i.e., appearance for the individual subject,
there could be no difference between them ([20] 234f; [54] 97f). With
some few exceptions ([2]537; [55]2, 234f), Kant was thus seen as an
idealist ([80] 43), \0 and the problem of the CPR became one of bridging
from the functions of the mind and the appearances it "possessed" on
the one side to the world of "real objects" on the other ([69] 374ff). The
irresistible habit of conducting all philosophical analysis in terms of a
presupposed opposition of self and world, automatically taken as the
relevant framework for every topic, was the ultimate source of this
misinterpretation ([3] 36). Even the refutation of the idealism offered in
the second edition of the CPR was habitually analysed in terms of
the opposing self-in-itself and world-in-itself that very argument was
designed to overcome ([76] 225f; [82]178-183). This dualism reappeared as the ultimate context defining every argument, even if within
the argument the dichotomy had been eliminated. Armed with nothing
more than the crude opposition of "the mindlike" and "the thinglike",
Kant's contemporaries missed his fine distinction of varying levels and
modes of the encounter between the conceptual and the empirical,
essential to an understanding of the relation of the CPR and MFNS.
The ultimate result of the promotion of the work of the understanding, with its necessitating force, into the realm of ordinary
theorizing about nature - thus overlapping with or even displacing the
work of the MFNS which properly occupies this territory - is our
central concern. The first avenue to the promotion of the understanding
was the difficulty in visualizing how the first promptings to action on
the categories could be as dimensionless as Kant needed them to be.
The simple solution was to imagine them to be ordinary objects
293
([5]164). Kant's use of examples from ordinary life to illustrate, not to

define, the operation of the understanding was taken to transpose the
categorical ordering into ordinary experience ([42]328; [43]188).
Everyday perceptions were forced into the role of Kant's "occasionings" as the material upon which the understanding operated ([4412.1,
349f, 359f; 4, 268f; [55]2, 47f). While a multiplicity of objects,
corresponding to the various degrees of reduction or realization of
objective experience in general, are essential to Kant's approach, his
followers found only one object, that of ordinary experience. While in
fact the extent to which the object was reduced determined the nature
of the matching conceptual structure needed to restore it, with only a
single object the post-Kantians could have only a single type of
conceptualization with one modality of application, where Kant had
had many. The pressing desire to solve Hume's problem, to which
German philosophy had been alerted in the last years of the eighteenth
century, encouraged the illusion that the necessitating, constitutive force
of the initial application of the categories must extend all though the
encounter of mind and experience, so that Kant could be used to defeat
scepticism: "This principle of the Critical Philosophy would thus, in
Hume's language, say that we can know a priori and without any
experience whether and according to what rules impressions are
combined .... There are thus propositions in human knowledge which
apply to objects and have necessity." ([11 224-7) The categories were
thus advanced into the area proper to the hypothetical operation of
reason. The universality and necessity of inductive projections could be
effected by their description in terms of the categories, as though saying
two phenomena are causally related rather than just associated could
validate the assertion. I I From this step, the inflation of the MFNS into a
prescription for nature of a priori certainty was inevitable ([49] 12lf).
Mellin, one of Kant's closest followers ([13] 2, 450), had an extreme
misconception on this point: Because causality was one of the categories
of the understanding, first "making experience real", Mellin imagined
that any positing of a cause or causal status, even if only ventured from
speculation or hypothesis, necessarily established the real existence of
whatever conceptual structures were associated with this causality.
Thus, since the dynamics of the MFNS were causally characterized, all
its associated posits must be real ([44]1.2, 622ff, 642)!
The picture of the general misconception of Kant now having been
sketched, we can look more closely at three of those who struggled with
294
DWIGHT C. BARNABY
the relation of the CPR to the MFNS - a close disciple of Kant, an

independent scientist, and a Naturphilosoph - to explain how the link
to Naturphilosophie was forged. The disciple, Johann Schultz, holding
with Beck the status of Kant's official expositor, made his first misstep
in taking the logical a priority of the MFNS for a temporal one, leaving
himself no choice but to regard its doctrines as a prediction of what
must exist in nature, rather than the merely conceptual development of
what nature had already given:
In the strict sense however one understands by the theory of nature . . . simply the
theory of bodies. If this is not to be a mere system of empirical laws of nature, thus not
merely an empirical theory of nature, but instead a true natural science, it must contain
laws of nature not simply taken from experience, which never carry strict necessity and
universality, but also a priori principles, i.e., such as precede these, independently of all
experience, and first make experience possible. A priori objective principles from mere
concepts are called metaphysical ... ([68] vii).
He is here about to blunder into a conflation of the CPR and the

MFNS on the basis of a mere pun, for while the work of the categories
in constituting objective experience in general is indeed called "general
metaphysics of nature", this is in fact entirely separate from the
"metaphysical foundations of natural science". While Kant designs the
concept of matter so as to exhibit its conceivability in terms of opposing
forces, the concept itself comes from empirical sources, whether experience of nature or of the contingent history of physical theory, and so is
"a priori" only in the sense that it is theoretical, not because it predicts
what must be. But Schultz reasons:
Thus phoronomy belongs to pure mechanics, and must go first in every system of
mechanics as its foundation. But with dynamics it depends on whether the forces of
matter are knowable solely in experience, or if it admits of proof a priori from mere
concepts that certain forces must originally reside in matter. ... But after Kant ... had
proved a priori that matter as matter must necessarily possess two original forces ...
dynamics belongs, insofar as it is confined merely to these two original forces, similarly
to pure mechanics ... ([68] iv-v).
Schultz is then already well on the way to making the MFNS an

imposition of the mind on the world, even before its conflation with the
categories. But he also positions the categories at the level of ordinary
experience for good measure, so that all those factors which fix the
object in general into determinate order are now seen to transform
295
empirical matter into the mere embodiment of the categorical concepts,

making it the physical copy of the purely transcendental notion of
substance ([68] 37f, 40),12 Thus made a mere creature of the mind, any
amount of theory can be packed into matter, and <;0 Newtonian
commitments, later built into it, are made "apodictically certain"
([68] ix-x).
Schmid, a physician with an interest in psychology and philosophy,
infamous for his debate with Fichte, at first promises to do better. He
understands Kant's notion of the role of the MFNS in establishing only
the "real possibility" of the hypotheses, and the separation between the
fundamental empirical application of the categories seems at times
dimly visible to him ([65]2, 192f, 203f; [67]1, 209, 214). Schmid sees
that metaphysical principles cannot put anything into experience, since
"natural science does not have compulsory force in itself to explain
fundamental forces, but only to show their necessary presupposition for
the conceptualization of what is given in experience" ([65]2, 42n),
though even here, he errs in imagining the possibility of force-explanation established in the MFNS to be a necessity.
Such insight as there was was short-lived, however. For Schmid the
decisive mistake permitting the MFNS to be seen as continuing the
work of the understanding is that Kant's distinction of empirical and
transcendental contexts, of the conceptualization with hypothetical and
with necessary force, is mislocated so as to coincide with the lines
of opposition of traditional, dualistic epistemology, so that Kant's
"empirical" context is that of the object-in-itself, while Kant's "a priori"
range becomes a mental structure superimposed on that object and its
visual phenomena. As a result there can be no effective distinction
between the principles of the MFNS and the CPR, now both equated
by their identical ontological base in appearance ([65]2, 10ff). Schmid's
reasoning displays the classic chain of displacements of Kantian structure: First the constitutive subjugation of the manifold to lawlikeness by
the understanding is seen as the fashioning of individual, empirical
objects to lawlikeness. This then forces ordinary objects to constitute a
scientific structure, now invested with all the properties characteristic of
the original work of the understanding. With this, science is made
necessary for the possibility of experience, the MFNS being made part
of the transcendental conditions of the CPR. Behind all of this
inference, as usual, is the assumption that Kant is talking about
296
DWIGHT C. BARNABY
processing functions in the mind of the individual, imposed on things.

When all knowing is thus reduced to the action of a single type of real
structure, it can then operate only with a single modality, necessity:
Rules are general representations of the uniform connection (unity) of a manifold. They
are rules of nature when this manifold is given as appearance in our sensible perception; they are laws of nature when they at the same time carry the character of
exceptionless generality and necessity in themselves .... The knowledge of the rules of
an object of sense is knowledge of nature; the knowledge of its laws is natural science.
. . . That animal phenomena [the particular science under consideration is zoology] like
all phenomena in general, constitute a nature, that is, that they are represented as
connected according to rules, is as necessary as their existence. For they are objects of
our mind (and it is just in this that their existence as appearances consists); this mind as
the principle of all-encompassing unity subjects everything given to it ... to its unifying
activity - through it everything is connected, subjected to rules and laws.... This
insistence on the necessity of a nature in general, insofar as there are to be objects of
intuition and activity of the understanding on our mind, is shown in the strictest sense
throughout the Critical Philosophy ([6511,25-8).
Kant's separate empirical and transcendental contexts, specific nature

and nature in general, scientific law and constitutive, fundamental
order, here all become effectively one. Schmid describes each separately,
but because he then cannot reproduce Kant's argument defining that
separation philosophically, he instead distinguishes them only trivially,
and winds up with a monism. This happens because the old tradition of
explanation via ontological geography is superimposed on Kant's
entirely new philosophy, so that everything which can be ontologized as
"of the mind", whether appearance, categories of the CPR or foundations of the MFNS, for that reason alone becomes related without
resistance. For Kant, in contrast, the significance of describing experience as "appearance" is exhausted as soon as epistemological dualism
has been overcome, and all further tension between concept and its
objects occurs within appearance - even if the CPR does understand it
as something for the subject - not over the gulf between appearance
and the thing-in-itself.
Although Schmid understands the difference in Kant's system
between constitutive and non-constitutive imposition of concepts, he
cannot now find a place in his theory for it. The only solution he can
contrive is to distinguish the categories from the scientific laws of
nature by making the former general, the latter specific, as though this
were the entire difference between the CPR and MFNS:
297
Pure natural science contains merely general laws a priori, without which no nature,
that is, no whole of the objects of experience can exist. It arises through the development of the necessary conditions under which an object can be sensibly intuited (given)
and conceived, that is, under which its manifold can be connected in a determinate
fashion. The subjectively necessary conditions of knowledge (experience) of an object
which lie in the human sensibility and in the human mind, are referred to things as
objects of sensibly understood knowledge, represented as conditions of their existence,
that is, a.s appearances. A priori, however, we cannot determine anything of the
particular kinds of natural objects, of what necessarily applies to them ... as natural
objects ([65[ 1,49f)Y
And in his dictionary guide to Kant, he can only formulate their difference as follows:
Theory of Nature: (a) General, pure, transcendental metaphysics of nature, ontology the investigation of the general and pure laws which make a nature in general possible,
without reference to particular, given natural things.... (b) Special, applied, metaphysical natural science; ... the application of those general laws of nature a priori to
particular, empirically-given objects of sense ([67] 277f, 392).
But, because as mere specifications of the work of the understanding

the laws of science cannot dilute its necessary force, the MFNS is again
fundamentally misconceived. Schmid fares little better in explaining the
difference between necessary and contingent physical laws, viewing the
former as distinguished by a mere act of will, our resolve to "think" of
them "as general and necessary" ([66]1,16).
Eschenmayer, a scientist initially Kantian but later quite closely
associated with Schelling, commits all the errors already catalogued. (In
fact, only Beck ([4) 2, 39lf, 396) and an obscure physician, Kollner
([37] 317ff, 349f, 356f; [38) 312f), even approached an understanding
of Kant on this vital matter, yet neither was able to instruct his
contemporaries effectively.) But for Eschenmayer the dualism of
experience and the thing-in-itself is within, not at the limit of, experience ([16) 56ff); the logical a priority in Kant's system is taken for a
temporal one ([17) 3ff); and the MFNS becomes a dogma from which
the specific detail of nature may be described a priori ([17] 7lf). The
categories are seen as logically necessary factors governing the ontology
of the real and thereby controlling the "being" of matter as described in
the MFNS ([17]13-16). The actual existence of the fundamental
forces constitutive of matter in that matter is then seen to follow, and
the purely cognitive stages by which Kant designs matter are reified as
298
DWIGHT C. BARNABY
the real forms through which empirical matter must come into being.
Then, curiously, all the Critical conceptual moments having been
hypostatized as the real, "empirical object of matter," this is in turn
taken as constraining us to interpret it as we do in the MFNS, because
it is "really given" to us as such ([171 SOf)!
Yet Eschenmayer is remarkable for having embraced the long
history of incorrect interpretation of Kant's philosophy of science and
transformed it into the beginning of a new perspective, a deliberate
confounding of the CPR and MFNS, opening a path by which Kantian
scientists could cross to Naturphilosophie. Where other interpreters of
the MFNS had wandered by error, Eschenmayer, like Schelling, would
go by design, and the following passage marks the historical transition
from the MFNS to Naturphilosophie. Only because he understood the
resistance of the MFNS to its accepted interpretation could he clarify
the nature of the step away from it:
In these various passages [in which Kant seemed to support both deductivist and
non-deductivist understandings of the MFNS[ a certain swaying from one side to the
other is noticeable. On the one hand it emerges that the possibility of all explanations in
nature should follow from the principles presented .... But on the other hand it is said
"that no law either of attracting or repelling forces may be ventured on a priori
assumptions, but that everything may be inferred only from the data of experience" ....
It suffices to understand the necessary assumption of those fundamental forces ... if we
put ourselves in the position of transcendental philosophy, giving to each of these
forces the exact qualities which could apply to it alone, in order to have sufficient
material for constructions in natural science. If we could still not comprehend ... the
possibility of the [application off the categories ... there must still be somewhere an
absolute proposition in which problem and solution, statement and proof coincide, in
which the conceivable comes to its conclusion ([ 17] 65-70).
This confidence that the first moment of the real is the identity of
concept and object is the credo of Naturphilosophie; but in contrast to
Kant, who sees this integration only at the foundation of transcendental
analysis and not in ordinary experience, Naturphilosophie, operating
with only one stage of analysis, sees the possible realization of concepti
object identity throughout experience. Eschenmayer continues:
There is no more in the concept of matter than we ourselves can put into it, that is, the
whole manifold of appearances belonging to material nature. .., [T]he sphere of
possible laws in general must be correctly delimited in a metaphysics of nature, and it
must contain the moments for every possible construction in natural science. Were
there to be other moments not in it and yet necessary for explaining natural phenomena, the question arises, in what area are they to be found? ... [T]he possibility of
299
concluding all explanations of nature in the metaphysical principles and of carrying

over to the former the necessity of the latter, cannot be disputed; although the ...
difficulty of actually achieving this is not for that reason diminished. Kant observes that
one should take care not to pass beyond that which makes the general concept of
matter possible and to wish to explain a priori the particular, or even the special
determinations of it. It seems to me as if exactly the special determinations of matter
must be contained in the general concept of it, that one should not need to advance
beyond the former to arrive at the latter, but only has to go to the concept, and that the
analysis of the general concept should at the same time make available the principles of
construction for their specific determinations.
Eschenmayer is asking, why set out the conceptual interpretation of

nature without carrying this conceptualization to completion, so that
every detail of the given expresses itself as an idea; why restrict
metaphysics to some small kernel, explicitly surrendering the ambition
to comprehend all of nature? To answer his question we must first
return to the most fundamental aspects of Kant's thinking, comparing
these with the new perspectives of Naturphilosophie. Kant begins with
one "Copernican turn", in which the given is made absolutely pliant to
its conceptual treatment; this occurs at the level of the operation of the
categories in generating determinate sequence, and is intended to
eliminate the traditional epistemological dualism from its foundational
position. But of course, above this, in ordinary experience, the gulf
between subject and object, concept and world is permitted to re-open.
In this range truth is defined as correspondence between idea and thing,
and "thinghood" occupies its traditional position as the ground of
ontology. Yet the possibility of a second "Copernican turn" must come
to mind. If at the transcendental stage, Kant confines the world to that
which would have significance for the subject, characterizing experience
as appearance for the subject and the thing-in-itself as meaningless, why
not repeat the procedure within the ordinary empirical context and
affirm, with an exaggerated Kuhnianism, that "appearance for a subject"
is not only what affects us according to conditions of space and time,
but what is "meaningful" for us at a higher level, what is consistent
not just with the bare modes of receptivity of the subject but also
with his scientific theory? In this view the ground of ontology would
no longer be thinghood but meaningfulness, the basis of truth no longer
correspondence to objects but the realization of theoretical interests in
the visible. Generalizing Kant's transcendental procedure, Fichte, the
ultimate father of Naturph ilosoph ie, holds that "intelligence, as such,
300
DWIGHT C. BARNABY
looks at itself", that objects, "are in me only insofar as 1 look at them:

Looking at and being are inseparably united" ([18]18). Coleridge,
writing under the influence of Schelling, remarks, "I wonder why facts
were ever called stubborn things? ... Facts, you know, are not truths;
they are not conclusions; they are not even premisses, but in the nature
of parts of premisses" ([11]165). Schelling, who never seems to have
understood Kant's basic position ([62] 91nf), adopts just the approach
suggested here ([56]534; [58] 142f, 342f; [59]245; [61]199). The unity
of self and world, concept and object, for Kant only found at the
transcendental level, is for Schelling the presupposition for all contexts,
and there are for him not two stages of application of the categories,
but one. Thus what began as the historical misconstruction of the
MFNS becomes the deliberate assertion of the new philosophy. This is
not to say that Schelling does not recognize that the mind of the
individual does not control or anticipate the world in all its detail
([62] 92f), a point admitted even by Fichte ([19] 43f, 43n). His choice is
simply to frame philosophical discussion so as not to comprehend
within it the conflict between data and theory, but rather to leave
behind epistemological questions, locating philosophy within the
conceptual elaboration of data already within the net of theory. The
issues of interest are now not so much those of the possibility of
knowledge as of order, the goal being the fullest realization of theory in
its objects ([61] 18lf). So, returning to Eschenmayer's question, there is
no reason not to build the MFNS so as to comprehend all of nature,
because the data/theory interface is no longer to be codified in our
theory, which therefore need not be limited to some small, wellfounded core to support hypotheses, but can instead be a limitlessly
free and flexible arranging of the data already in our possession. The
difference between Kant and Schelling is thus essentially one of
philosophic taste: Kant can formulate the concept/object opposition in
his system, while Schelling can focus his attention purely on questions
of theory-design rather than theory-confirmation, entering areas of study
very close to twentieth-century interests ([62]105f). That Eschenmayer
and Schelling both formulated the initial statements of Naturphilosophie at the same time, each in his own field, although not entirely
independently, shows that this was not a solution imposed by philosophy on a reluctant science, but represented instead the confluence of
ideas developing in both disciplines: in philosophy as the logical
extension to science of Fichte's autonomy of reason in morality, and in
KANT'S METAPHYSICAL FOUNDA TlONS
301
science as the result of the misconstruction of the MFNS. The essential

element in this misconstruction can be detected in the more recent
example of Simmel's philosophy which, in setting out a priori structures for a wide range of disciplines, not just for the initial constitution
of objectivity, follows a course similar to Naturphilosophie (e.g., [74] v).
If, as both Simmel and Schelling saw it, the a priori concepts of the
CPR are just asserted as the basic principles of objectivity, rather than
precisely deduced as the minimum necessary implications of the
experience of self and world ([73] 32), then why not extend this
assertion, providing general principles for the specific sciences as well,
of the same status and with the same right as those of the CPR? Kant
would reply that there must be a hierarchy of a priori principles, with
those of the CPR clearly superior, since only they constitute preconditions of discourse, only they are rigorously defined by the projection of
self and world, and are thus more secure that the historically-conditioned first principles of any discipline, abstracted from their science
and touted as "necessary".
While the interpreters of the MFNS simply blundered into their
invention of a "second Copernican turn" for Kant, the same cannot be
said of Schelling, who understood the internal strains of the Critical
Philosophy which themselves suggest a Naturphilosophic resolution.
Kant himself takes part of this step in his moral philosophy when he
affirms the reality of freedom over the fact of causality solely because
reason's interest in freedom is of higher value, of greater meaningfulness, than the affirmation of universal causal conditioning demanded by
the things of the world ([28] 42f, 48f, 54f, 89-107). But why should
meaningfulness be the ground of ontology in this context and not in the
scientific realm, where reason has to accept the limiting power of an
ontology grounded exclusively in bare thinghood? In fact, even here
Kant makes a tentative move towards Schelling's position. He remarks
that an ideal of reason "must offer no more warrant for its reality than
the need by reason to complete all synthetic unity through it"
([29] A614/B642), and a number of other passages show more interest
in supporting reason on an ontology of meaningfulness than in impeaching it through an ontology of thinghood. It is the systematic unity
sought by reason which is the "criterion of truth for its rules", not its
correspondence to resistant data ([29] A646ff/B674ff, A668/B696,
A772ff/B800ff; [30]168).14 But when we read that the right of reason
to require us to accept its posit of unity is that without it there would be
302
DWIGHT C. BARNABY
no coherent use of the understanding ([29] A650f/B678), we are then

returned to reflection on the origins of the Critical Programme, which
was to examine the conditions of the possibility of coherent experience
generally. Was Kant's choice then correct, to make the operation of the
categories alone necessary for the possibility of objective experience,
leaving the higher ordering functions of the MFNS and reason merely
ancillary in the constitution of experience as coherence? While some
structures are obviously more fundamental in this constitution than
others, was it correct to have classified them into two sharply separated
types, necessary in the understanding and hypothetical in reason, or
would it have been better to present them as all of a single type, in
practice gradually fading from constitutive to hypothetical application,
though all capable in principle of perfect realization in their objects?
Certainly the early interpreters of the MFNS found it easier to see in
the Critical Philosophy only a single type with potentially necessary
force at all levels of experience, while Naturphilosophie was to build an
entire system out of what had begun as a mere misunderstanding of the
MFNS.
NOTES
I Heidegger begins from a similar perspective, holding subject and object to be termini
only first isolated by analysis from what is phenomenologically always a single, unified
experience of subject-object relation in the process of knowing. Among Kant's contemporaries, Sieyes seems to have understood this same interpretive strategy (see [26]14,
484f).
2 Letter from G. C. Lichtenberg to Heyne, 27 April 1788 ([41] 2, 335).
3 Cf. Letter from D. Jenisch to Kant, 20 April 1796 ([27]12, 76).
4 Letter from L. H. Jakob to Kant, 24 January 1792 ([27)11, 305f).
5 Letterfrom H. Steffens to Schelling, January 1808 ([57]1,401).
6 Letterfrom Schelling to Hegel, 1795 ([52)1, 73).
7 Letter from C. S. Weiss to Oersted ([51]1, 309).
8 E.g. letter from C. G. Selle to Kant, 29 December 1787 ([27]10, 516f).
9 Cf.letter from Heinrich von Kleist, March 1801 ([48]8).
10 Letter from Reimarus to Mendelssohn, 14 June 1784 ([45]197).
II Asserted, for example, in Werner's paraphrasing of Girtanner's Ausfiihrliche
Darstellung des Brownischen Systems (Gottingen, 1797) ([85]1, 215f).
12 The same position is still being defended today (see [6]144, 147-150, 157).
13 A similar notion has been advanced recently (see [25]53).
14 Letterfrom Kant to J. S. Beck, 18 August 1793 ([27]11, 441).
303
REFERENCES
1. Anonymous, '1st Hume's Skepticismus durch die Kritik der reinen Vemunft
widerlegt? (Gegen Aenesidemus)', Philosophisches loumai einer Gesellschaft
Teutscher Gelehrten 4.2 (1796), 205-56.
2. Anonymous, 'Kant's Metaphysische Anfangsgriinde der Naturwissenschaft', Allgemeine Literatur-Zeitung (Jena), 261 (24 August 1789), 262 (25 August 1789),
537-52.
3. Anonymous, 'Ueber den Begrijf der Wissenschaftslehre ... von ... Fichte', Allgemeine Literatur-Zeitung (Jena) 1,5-8 (4-8 January 1789),33-71.
4. Beck, Jacob, Erliiutemder Auszug aus den critischen Schriften des Herm Professor
Kant, 3 vols (Riga, 1793-6).
5. Brastberger, M. G. U., Untersuchungen fiber Kants Critik der reinen Vemunft
(Halle, 1790).
6. Brittan, Gordon, Kant's Theory of Science (Princeton: Princeton Univ. Press,
1978).
7. Bryk, Otto, Entwicklungsgeschichte der reinen und angewandten Naturwissenschaft
im XIX. lahrhundert (Leipzig: J. A. Barth, 1909).
8. Buchdahl, Gerd, 'Kant's "Special Metaphysics" and the Metaphysical Foundations
of Natural Science', in Kant's Philosophy of Physical Science, ed. R. E. Butts
(Dordrecht:Reidel, 1986), 127-61.
9. Buchdahl, Gerd, 'Reduction-realization: A key to the structure of Kant's thought',
Philosophical Topics 12.2 (Fall, 1981),39-98.
10. Busse, Friedrich, Immanuel Kants metaphysische Anfangsgriinde der Naturwissenschaft in ihren Griinden widerlegt (Dresden, 1828).
11. Coleridge, Samuel Taylor, Table Talk, ed. H. Milford (Oxford: Oxford Univ. Press,
1917).
12. Dilthey, Wilhelm, Gesammelte Schriften, 12 vols in 11, eds. Georg Misch et al.
(Leipzig: Teubner, 1914-58).
13. von Eberstein, Wilhelm, Versuch einer Geschichte der Logik und Metaphysik bey
den Deutschen, 2 vols (Halle, 1794-9).
14. von Eckhartshausen, Karl, Probaseologie oder praktischer Theil der Zahlenlehre
der Natur (Leipzig, 1795).
15. von Engelhardt, Dietrich, 'Einheitliche und umfassende Naturdarstellung in der
Naturwissenschaft urn 1800 und Hegels Philosophie der Natur', Rete 1.1 (19712),167-92.
16. Eschenmayer, Adam Karl, 'Spontaneitat = Weltseele, oder das hochste Princip der
Naturphilosophie', Zeitschrift for speculative Physik 2.1 (1801), 1-108.
17. Eschenmayer, Adam Karl, Versuch, die Gesetze magnetischer Erscheinungen aus
Siitzen der Natur-Metaphysik, mithin a priori, zu entwickeln (Tiibingen, 1797).
18. Fichte, Johann, Versuch einer neuen Darstellung der Wissenschaftslehre (Jena and
Leipzig, 1797-8; Hamburg: Meiner, 1975).
19. Fichte, Johann, Ober den Begrijf der Wissenschaftslehre (Weimar, 1794).
20. Fries, Jacob, 'Atomistik und Dynamik', Studien 3.1 (1807), 201-35, in Siimtliche
Schriften, eds. Gert Konig and Lutz Geldsetzer (Aalen: Scientia, 1975) 17, 22157.
304
DWIGHT
c. BARNABY
21. Gilbert, Ludwig, 'Giebt es eine Warmematerie oder nichtT, Annalen der Physik
12.5 (1802),546-73.
22. Goldbeck, Johann, Metaphysik des Menschen (Altona, 18(3).
23. Graffe, Johann, Commentar uber eine der schwersten Stellen in Kant's Metaphysischen Anfangsgrunden der Naturwissenschaft (Celie, 1798).
24. Grohmann, Johann, Dem Andenken Kants, oder die neuern philosophischen
Systeme in ihrer Nichtigkeit dargestellt (Berlin, 1804).
25. Harre, Rom, 'Knowledge', in The Ferment of Knowledge, eds. G. S. Rousseau and
Roy Porter (Cambridge: Cambridge Univ. Press, 1980), 11-54.
26. von Humboldt, Wilhelm, Tagebucher, in Gesammelte Schriften. Ed. Preussische
Akademie der Wissenschaften (Berlin: A. Behr, 1903-36) 14.
27. Kant, Immanuel, Briefwechse!, in Kants gesammelte Schriften, 29 vols in 32. Ed.
Konigliche Preussische Akademie der Wissenschaften (Berlin: Reimer and de
Gruyter, 1902-83), II-l3.
28. Kant, Immanuel, Critik der practischen Vernunft (Riga, 1788; Frankfurt and
Leipzig, 1791) in Schriften 5,1-163.
29. Kant, Immanuel, Critik der reinen Vernunft (Riga, 1781 (=A); 1787 (=B.
30. Kant, Immanuel, Critik der Urtheilskraft (Berlin and Liebau, 1790) in Schriften 5,
165-485.
31. Kant, Immanuel, Geographie, ed. Friedrich Rink (Konigsberg, 1802) in Schriften 9,
151-436.
32. Kant, Immanuel, Metaphysische Anfangsgriinde der Naturwissenschaft (Riga, 1786)
in Schriften 4, 467-565.
33. Kant, Immanuel, Naturwissenschaftliche Reflexionen, in Schriften 14.
34. Kant, Immanuel, Preisfrage: Welches sind die Fortschritte, die die Metaphysik seit
Leibnizens und Wolf's Zeiten in Deutschland gemacht hat? ed. Friedrich Rink
(Konigsberg, 1804) in Schriften 20, 253-351.
35. Kant, Immanuel, Die Streit der Facultiiten (Konigsberg, 1798) in Schriften 7,
1-116.
36. Karsten, Wenzel, 'Versuch eines evidenten Beweises der allgemeinen mechanischen
Grundsatze', Abhandlungen der Churfiirstlich-Baierischen Akademie der Wissenschaften 8 (1773), 147-75.
37. KoHner, Johann, '1st die Heilkunde, als Wissenschaft betrachtet, moglich, und wie
ist sie esT, Magazin zur Vervollkommnung der theoretischen und praktischen
Heilkunde 1.2, 303-36; 1.3, 337-87 (1799).
38. Kollner, Johann, 'Priifung der neuesten Bemiihungen und Untersuchungen in der
Bestimmung der organischen Krafte, nach Grundsatzen der kritischen Philosophie',
Archiv fiir die Physiologie 2.2,240-350; 2.3, 351-96 (1797).
39. Kriinitz, Johann, 'Philosophie', in Okonomisch-technologische Encyclopiidie,
continued by Jakob Floerke et al. (Berlin, 1773-1838), II2 (1800).
40. Lavine, T. Z., From Socrates to Sartre (Toronto: Bantam, 1984).
41. Lichtenberg, Georg Christoph, Briefe, eds. Albert Leitmann and Carl Schiiddekopf
(Leipzig: Dieterich, 1901-4).
42. Maimon, Salomon, Versuch einer neuen Logik (Berlin, 1794).
43. Maimon, Salomon, Versuch uber Transcendentalphilosophie (Berlin, 1790).
305
44. Mellin, G. S. A., Encyclopiidisches Worterbuch der kritischen Philosophie (Ziillichau and Leipzig, 1797-1804).
45. Mendelssohn, Moses, Briefwechsel, in Gesammelte Schriften, ed. Alexander
Altmann (Stuttgart: Frommann, 1977), 13.
46. Metzger, Johann, Medizinisch-philosophische Anthropologie [iir Aerzte und
Nichtiirzte (Weissenfels and Leipzig, 1790).
47. Moller, Rudolph, 'Beitriige zur Biographie A. N. Scherers', Schriftenreihe [iir
Geschichte der Naturwissenschaften, Technik und Medizin 2.6 (1965),37-55.
48. Muth, Ludwig, Kleist und Kant (Cologne: Kainer Universitiits-Verlag, 1954).
49. Nose, Karl, 'Nachschrift des Herausgebers', to Johann Schiiffer, Ober Sensibilitiit
als Lebensprincip in der organischen Natur (Frankfurt, 1793), 113-26.
50. Oersted, Hans Christian, Ansicht der chemischen Naturgesetze (Berlin, 1812).
51. Oersted, Hans Christian, Correspondance de H. C. Orsted avec divers savants, ed.
M. C. Harding (Copenhagen: Aschenhoug, 1920).
52. Plitt, G. L., Aus Schellings Leben (Leipzig, 1869).
53. Reinhold, Carl, 'Recension von Phil. Magazin von J. A. Eberhard', Allgemeine
Literatur-Zeitung (lena), 174-76 (11-13 June, 1789), 577-600. (This review
was composed by Reinhold from notes provided by Kant).
54. Reinhold, Carl, Ueber das Fundament des philosophischen Wissens (Jena, 1791).
55. Reuss, Maternus, Vorlesungen iiber die theoretische und praktische Philosophie
(Wiirzburg, 1797).
56. Schelling, F. W. J., 'Einzelne Bemerkungen', Zeitschrift [iir speculative Physik, 1.2
(1800), in Siimmtliche Werke (Stuttgart and Augsburg, 1856ff) 4, 534-48.
57. Schelling, F. W. J., Briefe und Dokumente, ed. Horst Fuhrmans (Bonn: Bouvier,
1962ff).
58. Schelling, F. W. J., Ideen zu einer Philosophie der Natur (Leipzig, 1797) in Werke
2,1-343.
59. Schelling, F. W. J., 'Kritische Fragmente', lahrbiicher der Medizin als Wissenschaft,
1 (1806), in Werke 7,245-59.
60. Schelling, F. W. J., 'Miscellen', Zeitschrift [iir speculative Physik 2 (1801).
61. Schelling, F. W. J., System der gesammten Philosophie und der Naturphilosophie
ins besondere (1804) in Werke 6, 131-494.
62. Schelling, F. W. J., Zur Geschichte der neueren Philosophie (1827) in Werke 10,
1-200.
63. Scherer, Alexander, Nachtriige zu den Grundziigen der neuern chemischen Theorie
(lena, 1796).
64. Schlegel, Friedrich, Die Entwicklung der Philosophie (1804-6), in Kritische
Friedrich-Schlegel-Ausgabe, ed. Ernst Behler et al. (Munich: Schoningh, 1963) 12,
bk.5.
65. Schmid, Carl, Physiologie philosophisch bearbeitet (Jena, 1798-1801).
66. Schmid, Carl, Psychologie (lena, 1796).
67. Schmid, Carl, Worterbuch zum leichtern Gebrauch der Kantischen Schriften (lena,
1798).
68. Schultz, Johann, Anfangsgriinde der rein en Mechanik (Konigsberg, 1804).
69. Schulze, Gottlob, Aenesidemus (Gottingen and Helmstiidt, 1792).
306
DWIGHT C. BARNABY
70. Schwab, Johann, Priifung der Kantischen Begriffe von der Undurchdringlichkeit,
der Anziehung und der Zuriickstossung der Korper (Leipzig, 1807).
71. Selle, C. G., 'De la realite et de l'idealite des objets de nos connaissances',
Memoires de l'Academie Royale des Sciences et Belles-Lettres (Berlin) 1 (for the
years 1786-7; published 1792),577-612.
72. Simmel, Georg, Hauptprobleme der Philosophie (Leipzig: G. J. Goschen, 1910).
73. Simmel, Georg, Kant Sechszehn Vorlesungen (Leipzig: Duncker und Humblot,
1913).
74. Simmel, Georg, Die Probleme der Geschichtsphilosophie (Munich and Leipzig:
Duncker und Humblot, 1923).
75. Stichweh, Rudolf, Zur Entstehung des modernen Systems wissenschaftlicher
Diziplinen: Physik in Deutschland, 1740-1890 (Frankfurt: Suhrkamp, 1984).
76. Tieftrunk, Johann, Das Weltall nach menschlicher Ansicht (Halle, 1821).
77. Tilliette, Xavier (ed.), Schelling im Spiegel seiner Zeitgenossen (Torino: Bottega
d'Erasmo,1974-81).
78. Treviranus, Gottfried, Biologie (Gottingen, 1802-3).
79. Trommsdorf, Johann, Versuch einer allgemeinen Naturgeschichte der Chemie
(Erfurt, 1806).
80. Weiller, Kajetan, Der Geist der allerneuesten Philosophie (Munich, 1804-5).
81. Weishaupt, Adam, Ueber die Griinde und Gewissheit der menschlichen Erkenntniss
zur Priifung der Kantischen Critik der reinen Vernunft (Niirnberg, 1788).
82. Weishaupt, Adam, Ueber Materialismus und Idealismus (Niirnberg, 1787).
83. Weiss, Christian, 'Abhandlung iiber die Preisfrage: 1st die Materie des Lichts und
des Feuers die namliche, oder eine verschiedeneT, Physikalische Abhandlungen
der koniglich-baierischen Akademie der Wissenschaften 1 (1803), 1-167.
84. Weiss, Christian, 'Dynamische Ansicht der Krystallisation', Anhang to Karsten's
translation of Rene Hauy's Lehrbuch der Mineralogie (Paris and Leipzig, 1804) 1,
365-89.
85. Werner, Carl, 'Beurtheilung der Schriften gegen Brown's System', in Werner,
Apologie des Brownischen Systems der Heilkunde (Vienna, 1799) J, 199-288.
86. Will, Georg, Vorlesungen iiber die Kantische Philosophie (Altdorf, 1788).
87. Windischmann, Karl, 'Grundziige zu einer Darstellung des Begriffs der Physik und
der Verhiiltnisse dieser Wissenschaft zur gegenwartigen Lage der Naturkunde',
Neue Zeitschrift for speculative Physik 1.1 (1802), 78-160.
JOHN G. McEVOY
THE ENLIGHTENMENT AND THE

CHEMICAL REVOLUTIONl
Abstract. The Chemical Revolution is becoming a focus of renewed interest among

historians and philosophers of science, some of whom wish to relate it to the wider
sociocultural context of eighteenth-century life. This context included the cultural
movement known as the Enlightenment, at the core of which was a set of metaphysical
presuppositions concerning the nature of the knowing mind and its relation to the
object of inquiry. The Enlightenment notion of the self-defining subject established a
unitary framework of regulative principles, dealing with the relation between science
and metaphysics, the method of analysis, and the relation between thought and
language, which were variously interpreted in the opposing views that Lavoisier and
Priestley developed about the ontology of chemistry, the nature of experimentation, the
reform of the chemical nomenclature, and the institutional organization of science. The
ensuing dialectic occurred within an historiographical framework in which both sides
viewed the chemical upheavals of the eighteenth-century in terms of the Enlightenment
notion of the dawning of a new age, radically different from anything that had gone
before. The interpretation of the Chemical Revolution developed here calls for a more
balanced view of the moments of continuity and discontinuity in scientific change; it
also suggests that an adequate conception of scientific change can be formulated only
within a framework provided by a robust contextual model of science, in which the
distinction between the constitutive aspects of science and the contextual factors is
rejected in favour of a relational view of each as constitutive of the other.
INTRODUCTION: THE MIND OF THE ENLIGHTENMENT
The Chemical Revolution occurred towards the end of the Enlightenment. The conflict between the antiphlogistic system of chemistry and
its phlogistic rival is usually associated with the names of two scientists,
Antoine Lavoisier and Joseph Priestley, who shared the rational and
liberal principles of the philosophes. Both men paid a high price Lavoisier with his life and Priestley with his home and country - when
they carried these ideals into the political arena. Despite the significant
philosophical and scientific differences that separated these two men,
the shared context of their similar political destinies points to deeper
levels of congruence between them. The unitary nature of the Enlightenment mind, and especially its view of the methodological unity of all
rational thought and action, signals the existence of significant links
307
308
JOHN G. McEVOY
between the scientific activities and the other views and practices of
these famous protagonists in the history of chemistry. Some of these
connections are already known to the historian, who can point to the
different rhetorical uses that Priestley and Lavoisier made of parallels
between the Chemical Revolution and the French Revolution ([9] chap.
14; [31] 57), to the influence that the philosophical views of Locke and
Condillac exerted on Lavoisier's chemical thought ([I]; [16]), and to the
relation between Priestley's theological, philosophical and political
interests and his work in electricity and pneumatic chemistry ([26][30]). With a view to a more extensive study of the Chemical Revolution
and its place in the wider scheme of eighteenth-century life, this essay
will concentrate on a comparative analysis of the thought of Priestley
and Lavoisier in order to elucidate the interconnections between some
of the core theoretical disputes of the Chemical Revolution and the
ideals and values which are generally taken as characteristic of the
philosophical mind of the Enlightenment. In this manner, the dialectical
relation between Lavoisier, the leader of the antiphlogistic party, and
Priestley, the lone defender of phlogiston, will be shown to recapitulate
the unity-in-diversity that characterized the Enlightenment as a whole.
The first point to be made about the relation between the Enlightenment and the Chemical Revolution is an historiographical one. The
concept of a Chemical Revolution and the more general notion of a
scientific revolution did not flow ineluctably from the events they
sought to describe and were not generated in a historical vacuum.
Rather they were formed by forces inherent in the sociocultural
environment of eighteenth-century Europe. The Enlightenment was a
revolutionary ideology, in spite of the political moderation of its
champions. In the name of a rationalist and progressive individualism,
the philosophes sought to abolish the prevailing social and political
order, which shackled the free development of individual human beings
([19]34-5). These aspirations played a crucial ideological role in the
French Revolution, which was the most dramatic and significant event
in a wave of democratic revolutions which shook Europe during the last
quarter of the eighteenth century ([19]74-5, 79). These political
events exerted a powerful influence on the thought of the day, encouraging the emergence and development of new concepts and theories of
soCial and intellectual revolution; and these views were applied to the
articulation of a new historiography of the sciences. Towards the end of
the eighteenth century, an "older sense of 'revolution' as a cyclical
THE ENLIGHTENMENT AND THE CHEMICAL REVOLUTION
309
phenomenon . . . or a repetition" was replaced by a new use of the

term to denote "a breach of continuity, or a secular change of real
magnitude" ([8]257-8).
The modern notion of "revolution" shaped Priestley's description of
the Chemical Revolution as a "great ... sudden ... [and] general"
change, and determined Lavoisier's denial of the possibility of a return
to the "old order" in science and society after the Chemical and French
Revolutions had been completed ([42] 35; [8] 282). Attentive to the
conjunction of the Chemical Revolution and the French Revolution,
proponents and opponents of the oxygen theory shared an exhilarating
sense of living in "an age of revolutions, philosophical as well as civil"
([42] 36). Both friends and foes of the new chemistry exploited an array
of associations with its radical political context to promote diverse
rhetorical objectives ([31] 57; [47]). Shaped by political values and
events associated with the Enlightenment, the view that the generative
phase of modern chemistry was constituted by a revolutionary rupture
with traditional chemical theory and practice has exerted a lasting
influence on subsequent generations of historians of science, providing
a thread of continuity through a maze of diverse and conflicting
historiographical sensibilities and philosophical persuasions ([12]).
In order to explore the more substantive relations between the
Enlightenment and the Chemical Revolution, it will be useful to
construct, from a variety of secondary sources and authorities, a
composite image of the mind of the Enlightenment centered on the
notion of the epistemic self, or the knowing subject ([2J 109-28; [3J
12-26; [4]; [7]3-27; [15]17-77; [17]1-23; [49]3-11). Extending
and completing the scientific and philosophical changes initiated by
seventeenth-century thinkers, the philosophers of the Enlightenment
forged the modern concept of the self-defining subject. Medieval and
early-Renaissance thought defined the self in relation to the cosmic
order and equated reason - the highest attribute of human beings with the eternal verities held in common by the human and the divine
mind. Reason, the natural light of the mind, linked the cosmic order
and the human subject in a relationship of reflection and systematic
correspondence. The real task of knowledge (philosophy and science)
on this view was the construction of metaphysical systems, based on
intrinsic, a priori links between the knowing subject and the cosmic
order, in which the method of proof and rigorous inference was used to
spread the light of certainty over derived being and knowledge. The
310
JOHN G. McEVOY
Enlightenment mind, in contrast and reaction to this perspective,

rejected the idea of an intrinsic link between the self and the cosmos:
the knowing subject encounters the world as something other, as a set
of de facto, contingent correlations. Drawing back from the world,
the self-defining subject concentrates on the nature and limits of its
own activity. The eighteenth century was characteristically an age of
epistemology, psychology, and methodology.
Within this new framework of thought, reason was no longer
identified with a sound body of knowledge; it was viewed instead as an
activity characteristic of man, as a method of inquiry which guided the
discovery of truths. Reason was not embedded in a metaphysical
system, or a vision of the cosmic order; rather it operated through the
methods of the empirical sciences. The logic of empirical facts replaced
the logic of metaphysical systems. Order, law and reason could not be
grasped prior to the phenomena, but had to be discovered in the
phenomena themselves. This, for the eighteenth-century mind, was the
essence of Locke's rejection of innate ideas and Newton's method of
analysis, which moved from phenomena to principles and not vice
versa. The method of analysis, which was regarded as characteristic of
the empirical sciences, understood an event by analysing it into its
elements, from which it was then reconstituted or synthesized. Science
could progress independently of metaphysics, which, like all other
disciplines, should subject itself to the methods and results of the
sciences. In replacing metaphysical reason, grounded in things, with
scientific reason, anchored to method, the Enlightenment mind replaced
the Renaissance doctrine of signatures, which posited an intrinsic link
between words and things, ideas and objects, with the view of language
as a representational system of signs, which are independent of the
things they represent. Within this linguistic framework, the method of
science and the language of science were necessarily linked to the
epistemological procedure of classification.
The Chemical Revolution captured something of the characteristic
unity of the philosophical mind of the Enlightenment to the extent
that, despite their considerable intellectual differences, Lavoisier and
Priestley shared an abiding commitment to the liberation of science
from metaphysics, a strong desire to deploy the method of analysis in
the understanding of nature, and a deep sense of the epistemological
significance of reforming the language of science and reconstructing the
nomenclature of chemistry. The fact that Lavoisier and Priestley
311
pursued radically different scientific objectives and reached fundamentally opposed conclusions within a shared framework of regulative
principles lends support to the interpretation of the Enlightenment as a
coherent and self-conscious movement of epistemological and methodological reform which encompassed a wide range of opinions and
doctrines about the world. As with other aspects and dimensions of the
Enlightenment, the philosophical and scientific differences between
Lavoisier and Priestley encompassed variations in cultural patterns and
procedures, as well as divergences in personal predilections and presuppositions. At the core of this dialectic was the polarity between
Lavoisier's notion of an active, social, hierarchical subject and Priestley's
concept of a passive, individualistic, egalitarian self. In this fashion, the
Enlightenment view of the self-defining subject was shaped according
to different structural and organizational features of French and British
science and society. The international unity of Enlightenment thought
encompassed considerable diversity at the levels of national and
individual consciousness ([36]).
THE ONTOLOGY OF CHEMISTRY
Lavoisier and Priestley shared the epistemological and psychological

assumptions of the philosophes ([16]112-19; [25] 349-57). They
endorsed Locke's doctrine that, at birth, the mind is a tabula rasa: the
ideas and operations of the mind are all determined by experience. Real
knowledge is obtained only when the mind liberates itself from a priori
prejudices and opinions and derives its ideas directly from experience
and observation. Speculative notions and metaphysical systems are a
hindrance to the development of science, which progresses through a
rigorous analysis of sensory experience. These epistemological considerations led Lavoisier to treat traditional attempts to determine the
nature, number, and distribution of the chemical elements in the world
as "discussions entirely of a metaphysical nature" ([22] xxiv). Lavoisier
was critical of the established mode of chemical theorizing, in which the
generic properties of bodies were explained by reference to a small
number of ultimate "principles" present in all, or almost all substances,
but beyond the reach of the senses or the analytical techniques of his
time.
Lavoisier developed a line of criticism which extended back in time
before his appearance on the historical scene ([48]). The elements of
312
JOHN G. McEVOY
early-eighteenth-century chemistry, such as Earth, Air, Fire, and Water,

were property-conferring principles which designated kinds of beings
or types of qualities. Initially thought to be unobtainable in the free
state and to be apprehendable only by their effects, as the century wore
on they came to be regarded as "obtainable in impure forms as the end
product of analysis", and were eventually transformed into classes of
specific isolable substances ([48)276). In this manner, the Earthy
principle gave way to the class of specific earthy substances, the
element Air was replaced by a multiplicity of chemically distinct airs,
and ordinary water was substituted for elementary Water ([32)).
The "century-long transition away from the metaphysical towards the
operational concept of the element" culminated in Lavoisier's pragmatic
definition of an element as the end-product of analysis and coincided
with Priestley's epistemological characterization of material substances
as "things that are the objects of our senses, being visible, tangible, and
having weight, etc." ([48)278-83; [41)4). This epistemological analysis
compelled Priestley, in 1783, to identify phlogiston with a specific
inflammable substance, namely inflammable air, which was isolable in
the laboratory and detectable by the balance ([39) 398-412).
The ontological changes associated with the Chemical Revolution
belong to a broader philosophical context. On the level of discernible
"influences", Lavoisier acted in conformity with Locke's (and Condillac's) criticism of traditional metaphysics when he replaced the ontology
of "real essences", underlying and determining the properties of things,
with the nominalist view of elements understood as "simple substances",
or "the last point which analysis is capable of reaching" ([22) xxiv;
[34)51-2). Viewed in these terms, the chemical elements do not
denote the ultimate constituents of bodies, out of which all their
properties are generated, but refer to the specific substances which are
obtained by current experimental analysis. Viewed from a broader
philosophical perspective, these ontological changes also support the
suggestion that the development of science in the seventeenth and
eighteenth centuries involved the emergence of a new logic of concept
formation. The fact that the Chemical Revolution involved the transition from an ontology which grounded phenomena in underlying
substances to one which focused on the invariant relations of analysis
and synthesis between observable things supports the claim that, "for
the growth of science, from the logician's point of view, [the) replacement of substance by law is a central issue" ([5]51; [6]; [25)370-2).
313
The structural features of the Enlightenment mind can also be

discerned in a comparative study of Lavoisier's concept of a chemical
element and Newton's theory of gravitation. In rejecting the explanatory
aspirations of Newton's gravitational theory because the concept of
action-at-a-distance was incomprehensible, the Cartesians insisted that
the aim of science is to render the phenomena intelligible by reducing
them to their constituent principles, or essences, which, because they are
based on divinely-guaranteed links between the knowing subject and
the cosmic order, are ultimate, innate, and transparent to the intellect
([7] 50-1). In response to this line of criticism Newton insisted that
although action-at-a-distance was unintelligible, or irreducible to a set
of universally-recognized, transcendentally-grounded essential principles, nevertheless gravitation, as a universal phenomenon established by
the method of empirical analysis, should be acceptt<d as an explanatory
principle ([7] 52-3). Though not an ultimate cause, gravitational attraction was to be accepted as a temporary "last" element of nature.
Lavoisier endorsed the Newtonian view of scientific explanation when,
besides eliminating from chemistry any reference to the ultimate
"constituent and elementary parts of matter", he also recognized the
provisional nature of his list of operational elements, which, "though
they act with regard to us as simple substances", may be shown at some
future date to "be composed of two, or even of a greater number of
principles" ([22] xxii-xxiv). In detaching explanation from intelligibility,
Lavoisier and Newton upheld the self-defining nature of the knowing
subject, which shares no common essence with the world, but which
encounters it, through experience, as something other. Lavoisier's
conception of a chemical element as the end product of laboratory
analysis was part of the wider movement of thought involved in the
autonomous development of the empirical sciences in the eighteenth
century.
Neither Lavoisier nor Priestley were consistent in the pursuit of their
epistemological objectives. Lavoisier's notorious failure to make a clean
break with the Stahlian tradition of "real essences" he sought to
overthrow is evident in his inclusion of caloric, the acidifying principle
and the alkaline principle in his list of simple substances ([35]).
Nevertheless, the main thrust of Lavoisier's thought was innovative, and
his followers purified his revolutionary views and aspirations ([48] 292;
[32]). Priestley, on the other hand, was wedded more firmly to the logic
of traditional chemical discourse. Thus he soon abandoned his attempt
314
JOHN G. McEVOY
to identify phlogiston with inflammable air; and by the early 1790s he

was defending "the hypothesis of water being the base of every kind of
air, the difference between them depending upon the addition of some
principles which we are not able to ascertain by weight" ([45] 21;
[40] 221). After a brief flirtation with the new concept of a chemical
element championed by Lavoisier, Priestley reverted to a chemical
ontology of "principles", or "real essences", devoid of any reference
to "simple substances". Even then, however, Priestley upheld the
epistemological values of the Enlightenment by stressing the speculative
and provisional nature of current attempts to determine "the composition and elementary parts of all substances" ([42] 8-9). He laid "no
great stress" on these "speculations", emphasising instead "the facts"
relating to the observable properties and interactions of the substances
of gross chemical experience. Priestley never tired of insisting that the
true theory of the elements would emerge inductively and automatically
from the accumulation of facts, not the exfoliation of speculative
systems ([26] 35).
THE EPISTEMOLOGY AND METHODOLOGY OF CHEMISTRY
Within a shared Lockean framework, which reduced the entire content

and activity of the mind to sensations and their transformations,
Priestley and Lavoisier upheld diverse conceptions of the nature and
function of the knowing mind. Priestley adopted Hartley's view of the
passivity of the epistemic subject: he viewed thought as the natural
product of the mechanical law of association, and he reduced discoveries in natural philosophy to the equal ability of all minds to
accumulate and inductively order "facts". These ideas led him to attack
the elitist "spirit of systems" and the predilection for "speculation"
which accompanied it ([26] 30-9). Taking his cue from Condillac,
Lavoisier endorsed Priestley's epistemic devaluation of hypotheses
which transcend the realm of observable "facts" ([22] xv-xvi); but he
did not adopt Priestley's extreme inductivism. While sensitive to the
danger that the "spirit of systems" posed for the "physical sciences",
Lavoisier was equally fearful that "by heaping up too great a number of
experiments without order one might obscure the science rather than
clarify it" ([23] 2, 225). Lavoisier accepted Condillac's view of the
active knowing mind, and he emphasised the epistemological importance of the theoretical ordering of facts in a manner that was at odds
315
with Priestley's emphasis on the patient accumulation of "new facts"

and the inductive emergence of a "general theory".
The contrasting epistemological allegiances of Lavoisier and Priestley
were intertwined with disparate interpretations of the "method of
analysis", which was taken by the Enlightenment to be the proper
scientific method, wherein truth is obtained by decomposing and
recomposing ideas from the simplest components of experience. The
"analytic and historical method" deployed by Priestley was shaped by
an empiricist methodology of enumerative induction, which contrasted
sharply with the rationalist procedure of "mathematical analysis"
underlying Lavoisier's experimental inquiries and theoretical conclusions. This dimension of Lavoisier's thought bears witness to the
influence of Condillac's Logic, which integrated Newton's method of
decomposition and Continental views of algebraic analysis into a
method of analysis predicated on the doctrine that all reasoning is
algebraic in character ([1]31-74, 110-85; [17] 17-23, 110-85;
[17]17-23). Under the influence of Condillac's methodological guidelines, Lavoisier maintained that just as "mathematicians obtain solutions
to a problem by the mere arrangement of data", so chemists should
solve the problems of chemical composition by a "true mathematical
analysis", in which the unknown composition is identified with a simple
recombination of known substances ([22] xviii; [1] 64-72, 128-34).
Claiming that "all reasoning in scientific matters implicitly contains true
equations", Lavoisier developed the first genuine chemical equations
based on the assumption, inherent in the principle of the conservation
of matter, that in all chemical reactions, "the quality and quantity of
the principles are the same, and that there are only alterations and
modifications" ([23] 3, 777-8 ; [23]1, 101; [1]139-50).
The notion that all scientific reasoning is algebraic in character
influenced Lavoisier's doctrine of the composite nature of water, which
he formulated and presented in the form of a "true equation", or a
"proof", or a "demonstration", which approached the logical rigour of
mathematical proof. Given the Newtonian principle of the conservation of matter and the oxygen theory of combustion, he found it
impossible "to doubt" the "decomposition and recomposition of water"
when "we see that, in burning together fifteen grams of inflammable air
and eighty-five of pure air, we get exactly a hundred grams of water;
and when we can, by decomposition, find again the same two principles, in the same proportion" ([21] 298). To doubt this conclusion,
316
JOHN G. McEVOY
Lavoisier claimed, is to doubt the method of analysis, the only route to

certainty in science.
Priestley was of the opposite opinion to his great rival; and he
accused Lavoisier of arriving at the goal of mathematical certainty in
chemistry only by violating the principles of the "analytic and historical"
method. In place of a rigorous observational verisimilitude, which
related "analysis" to the accumulation and inductive ordering of facts,
Lavoisier adopted, according to Priestley, the "synthetic style" of
inquiry and presentation, in which facts are rendered subservient to
theories ([26]37-8; [28]155-6). In order to present his view of
the composition of water in the form of a quasi-algebraic equation,
Lavoisier described an idealized experiment in which oxygen and
hydrogen combined together to form pure water. Besides dismissing the
acidic solution that Priestley always obtained in this experiment as
an impurity-effect due to the presence of nitrogen in the reactants,
Lavoisier developed sophisticated experimental procedures and elaborate laboratory apparatus in order to eliminate all impurity-effects
from his results and to produce the idealized data necessary for the
formulation of "true equations in chemistry" ([22]94-5, 437-41).
Priestley was sceptical of the results obtained in these idealized
experiments, which he criticised for being too "expensive" and "complex", and for using too much "computation" and "allowance" in
advancing from experimental data to theoretical conclusions ([46]103).
He accused "the Lavoisians" of introducing "opinion" and "prejudice"
into the realm of observation and experiment in such a way as to
violate the principle of the passivity of the knowing mind in the
collection and inductive ordering of "facts". In contrast to Lavoisier,
who used a preconceived theory of chemical compositions and reaction
mechanisms to distinguish genuine products from impurity-effects in
the reaction between hydrogen and oxygen, Priestley structured his
phlogistic discourse more in accord with the perceptible properties and
interactions of the laboratory substances involved. For Priestley, the
passive reception of experimental data by the mind indicated that nitric
acid was as much a product of the reaction between dephlogisticated
and inflammable air as was water ([40] 220-1). Above all, Priestley
maintained that, whatever was the correct interpretation of this experiment, the "analytic and historical" method of communication required
philosophers to provide the public with a faithful record of all their
observations, in the order and manner of their occurrence, and not, as
317
in the "synthetic" mode of discourse, to report the facts "as if everything

had been done to verify a true preconceived theory" ([37] 574).
Whereas Lavoisier interpreted the Enlightenment notion of analysis in
terms of a doctrine of experimental "proof" based on "demonstrative
experiments", Priestley (and Kirwan) related it to the examination of
nature through the use of a "multitude of experiments", performed and
recorded "impartially" ([20] 61, 307).
THE LANGUAGE OF CHEMISTRY
The contrasting strains of rationalism and empiricism in the thought of

Lavoisier and Priestley surfaced in the:r respective programmes for the
reform of the chemical nomenclature. These programmes of linguistic
reform indicate the extent to which eighteenth-century views on the
methodology of science were closely related to philosophical theories of
the language of science. A useful framework for the analysis of these
linguistic sensibilities is provided in Foucault's discussion of the emergence of the classical episteme from its Renaissance background.
According to this analysis, for philosophers of the Renaissance, language resided in the world as a system of signs enmeshed in the
resemblances and similitudes of things. Taking the external characteristics of things to signify a hidden power or a relation to other things,
the doctrine of signatures maintained that the grammar of natural signs
could be deciphered to reveal the syntax of being. The Classical
episteme, which provided the underpinnings of the Enlightenment
mind, severed this intimate link between words and things. Signatures,
which inhere in and resemble things, were replaced by signs, which are
distinct from the things they represent. The interpretation of the
language of nature gave way to the representation of nature in the
language of men. The essential epistemological procedure became the
classification of phenomena by ordering a system of signs from simple
to complex. The method of analysis wedded thought to language. The
construction of a precise language was identified with the analytical
procedure of creating a systematic science by reducing the data of
sensory experience to its representational components ([2]109-28;
[15]17-63; [18]62-72).
Condillac summed up this line of thought when he declared that
science is nothing but a well-made language ([10] 2, 469). Lavoisier
upheld Condillac's view of the analytical activity of the knowing mind,
318
JOHN G. McEVOY
in which linguistic reform and theoretical development are inextricably

linked in the progress of thought from the known to the unknown
([11) 68-92). Thus he argued that "the impossibility of separating the
nomenclature of science from the language of science" implied that any
change or improvement in the one was inextricably bound up with a
change or improvement in the other. For this reason, his work on the
formation of a new "Nomenclature" led him to write "a treatise upon the
Elements of Chemistry" ([22) xiv). Lavoisier's reform of the chemical
nomenclature was intimately related to the logic and development of
the oxygen theory; and, as Priestley complained, the former could not
be utilized without understanding the latter ([46) 104-5).
Priestley rejected Lavoisier's new nomenclature because it was based
on "principles ... not ... sufficiently ascertained" ([38) 2, 335). On the
level of specifics, he insisted that it was only Lavoisier's subjective
preference for the hypothetical principles of the oxygen theory which
justified his calling "finery cinder" the "black oxide of iron" in the
absence of any "direct evidence of its containing any oxygen at all"
([44) 251). More generally, he maintained that contemporary ignorance
of the "primary constituent parts" of bodies meant that Lavoisier's
attempt to base his chemical language "upon a knowledge of the real
constituents of natural substances" only ended in confusing "facts" with
mere "opinions" ([38) 2, 30; (46) 105). In contrast, Priestley designed
the "use of terms" to guarantee the permanence of a scientific language
anchored to the epistemic bedrock of "facts", rather than the shifting
sands of "hypotheses" and "conjectures" ([28) 154-5). In accordance
with this regulative linguistic principle, Priestley developed a chemical
language to describe the perceptible characteristics, circumstances and
transformations of perceptible substances, without any reference to
underlying and imperceptible reaction mechanisms and chemical compositions ([26)-[29)). Priestley's conception of a permanent scientific
language was shaped by the more general, Lockean view that language
is a symbolic expression, or description, of independently-existing
mental patterns and processes ([25) 354-7). Insofar as the Lockean
separation of thought and language accommodated Priestley's notion
that language passively reflects the prelinguistic, mechanical processes
of inductive reasoning, it reinforced his conception of a scientific
language rooted in the "facts" of experience, and independent of the
fluctuating products of the theoretical imagination. In so opposing
Lavoisier's reform of the chemical nomenclature, Priestley highlighted
319
characteristic differences between his own empiricist orientation towards observation and experience and Lavoisier's rationalist preference
for theorising and reasoning.
Implicit in Priestley's criticism of Lavoisier's reform of the chemical
nomenclature was a rejection of the Classical identification of thought
and language with the construction of a conventional system of signs.
This strategy did not, however, lead Priestley back to the doctrine of
signatures, but to a third, intermediary position. He did not view words
as inhering in and resembling things; nor did he view them as constituted by the system of knowledge in which they occurred. For
Priestley, signs were neither resident in nature nor constituted by
thought. Instead they were anchored in facts and derived their signification from experience, which grounded thought in nature. In this
manner, Priestley's view of the cognitive status and function of language
marks an empiricist half-way point between the Renaissance doctrine of
natural signatures and the Classical theory of conventional signs.
THE INSTITUTIONS OF CHEMISTRY
The contrast between Priestley's empiricism and Lavoisier's rationalism

can be placed in a broader sociocultural context. This context can be
characterized by the juxtaposition of two historical images, which
suggests that whereas English thought in the eighteenth century was
concrete, practical, and individualistic, French thought was abstract,
principled, and corporate ([36]1-18, 41-53). Whereas the rational
and liberal principles of the Enlightenment were articulated in England
in such a way as to link knowledge to the experience of the individual,
in France they were developed in a way that associated rationality with
the organization and administration of society. The difference between
the individualistic empiricism of the English and the corporate rationality of the French is discernible in the different views that Lavoisier
and Priestley developed about the source of knowledge, the methods
and organization of science, and the relation of thought to language.
As a leading representative of the English Enlightenment, Priestley
mingled liberal, Protestant and Baconian motifs in a vision of epistemological progress based on the cooperation of individual experimentalists, united by a network of communications encompassing diverse
views and interests, but unencumbered by the institutional accretions of
established theory and practice ([31 D. On this view, science could
320
JOHN G. McEVOY
support the forces of progress, rationality, and liberalism only if it

avoided all dogmatic modes of thought and encouraged the intellectual
and moral development of the individual. In opposing the "French
system" of chemistry, Priestley placed more epistemic value on the
observations and judgments of individual experimentalists than on a
shared body of theory established and perpetrated more by the techniques of indoctrination than by the logic of critical discourse. Criticising John MacLean for basing his opinions on what he read rather
than on what he observed, Priestley commented: "I speak from my own
experiments, and I only wish that Dr. MacLean would speak from his"
([43] 29). Priestley's rejection of the phlogiston theory was rooted in his
refusal to take on faith results he could not duplicate in his own
laboratory.
The individualistic sensibilities inherent in Priestley's notion of
speaking "from his own observations" contrast with Lavoisier's conception of the social nature of scientific thought and practice. According to
Lavoisier, he and his colleagues had used "the habit of communicating
our ideas, our observations, and our way of thinking to each other to
establish a sort of community of opinions, in which it is often difficult
for everyone to know his own" ([22] xxxiii-iv). Although Lavoisier was
the undisputed leader of this hierarchical community, the cooperative
division of theoretical and empirical labour among its members, and
their readiness to close ranks against external criticism and dissent,
betokened a collectivist enterprise which was at odds with Priestley's
individualistic preference for thinking, reading, working and publishing
on his own. A similar divergence in epistemological sensibilities underscored the difference between Priestley's Lockean view that language
expresses, or communicates, the prelinguistic experiences and meanings
of isolated individuals and the notion favoured by Lavoisier and
Condillac that the thoughts and ideas of the individual are shaped by
the communal tool of language ([14]60-1, 115). Yet again, whereas
Priestley struggled, in the name of the civil and intellectual liberties of
the individual, to minimize the power and influence of central governments and established institutions, Lavoisier was influenced, in his
scientific and nonscientific activities, by Turgot's programme for using
the methods of science to reform, rationalize and, thereby, strengthen
the administrative structure of the nation ([13]; [17]159-61). The
broader intellectual context within which these contrasting sensibilities
were located is evident in the different agendas for the development of
321
a science of man inherent in the English doctrine of associationism,

which sought to construct human nature out of the experiences of the
individual, and the programme adapted by the French philosophes for
the construction of a social science based on actuarial considerations
and the mathematical theory of probability ([25) 348-57; (17) 17990).
The difference between Lavoisier's collectivist view of knowledge
and Priestley's individualistic epistemology also highlights differences
between the institutional organization and structure of the French and
British science in the late eighteenth century ([24]). In the highly
organized and centralized scientific community of France, the pressures
of formal education, centralized learned societies, employment opportunities, and a competitive system of reward and recognition meant that
the aspiring French chemist had little choice but to follow the intellectual lead of the Academicians in Paris. In contrast to the intellectual
conformity of state-subsidized French chemists, English natural philosophers offered a more varied and entrenched opposition to the
communal dissemination of Lavoisier's theory. In this manner, they
offered the kind of resistance to the centralization of power and
authority that was characteristic of an individualistic mode of inquiry
conducted by a loose congregation of amateurs, with weaker organizational structures and fewer educational opportunities than their French
contemporaries, but with closer ties to entrepreneureal industry. From
this perspective, Priestley's opposition to the oxygen theory constituted
a cultural resistance to the emergence of chemistry as "a coherent,
professionalized and autonomous science" ([50) viii). Priestley opposed
the socialization and professionalization of chemical knowledge because
he believed that science should give the "greatest scope to true freedom
of thinking and enquiry" ([41) ix). It is symptomatic of the diversity
within the unitary movement of the Enlightenment that Lavoisier found
the centralization of power and authority associated with the dissemination of his theoretical doctrines perfectly compatible with the rational
and liberal principles which led Priestley to oppose him ([16) 124-31).
CONCLUSION: THE NATURE OF SCIENTIFIC CHANGE
Implicit in this study is a challenge to current theories of scientific

change. In particular, it is opposed to the current view that the
Chemical Revolution constituted a fundamental and irrevocable break
322
JOHN G. McEVOY
with previous chemical theory and practice ([24]). In place of the

ideology of incommensurability, which exalts the discontinuous over
the continuous, this account of the Chemical Revolution points to a
more balanced account of the interrelatedness of the moments of
continuity and discontinuity in the process of scientific change ([33]).
Contributing to the discontinuous dimensions of the Chemical Revolution were the radically different opinions that Lavoisier and Priestley
held on the relation between theory and experiment, the function of
hypotheses in the discovery of knowledge, the role of language in
scientific conceptualization, the relative priorities to be given to reason
and experience in the development of chemistry, and the structure and
organization of science as an institution. Beyond the discontinuous level
of opinions and practices, however, there was a realm of shared forms
and presuppositions - necessary for the formulation of these opinions
and practices - constituted by prevalent views of the relation between
science and metaphysics, the centrality of the method of analysis in the
understanding of nature, and the importance of a rationalized language
and nomenclature for the development of science. Viewed in these
terms, the Chemical Revolution was an integral part of the Enlightenment and, as such, was governed by the logic and dynamics of a unitary
movement of thought with diverse manifestations.
NOTE
I This paper was written with support from the Taft Committee of the University of
Cincinnati.
REFERENCES
1. Albury, William R., The Logic of Condillac and the Structure of French Chemical
2.
3.
4.
5.
6.
and Biological Theory, 1780-1801 (Baltimore: The Johns Hopkins University,

Unpublished Ph.D. dissertation, 1972).
Baker, Keith M., Condorcet. From Natural Philosophy to Social Mathematics
(Chicago: Univ. of Chicago Press, 1975).
Brumfitt, J. H., The French Enlightenment (Cambridge, MA: Schenkman, 1972).
Buchdahl, Gerd, The Image of Newton and Locke in the Age of Reason (London
and New York: Sheed and Ward, 1961).
Buchdahl, Gerd, Metaphysics and the Philosophy of Science. The Classical Origins,
Cassirer, Ernst, Substance and Function and Einstein's Theory of Relativity, tr. W.
C. Swabey and M. C. Swabey (New York: Dover, 1953).
323
7. Cassirer, Ernst, The Philosophy of The Enlightenment, tr. F. C. A. Koellin and J.

A. Pettegrove (Boston: Beacon Press, 1955).
8. Cohen, I. Bernard, The eighteenth-century origins of the concept of scientific
revolution', Journal of the History of Ideas 37 (1976),257-88.
9. Cohen, I. Bernard, Revolution in Science (Cambridge MA: Harvard Univ. Press,
1985).
10. Condillac, abbe Etienne Bonnot de, Oeuvres philosophiques, 3 vols, ed. G. LeRoy
(Paris: Presses Univs. de France, 1947-51).
11. Crosland, Maurice, Historical Studies in the Language of Chemistry (London:
Heinemann, 1962).
12. Donovan, Arthur, 'The chemical revolution revisited', in Science and Technology
in the Eighteenth Century, ed. Stephen H. Cutliffe (Bethlehem, PA: Lehigh Univ.
Press, 1984), 1-15.
13. Donovan, Arthur, 'Chemistry and the progress of science in eighteenth-century
thought' (Unpublished paper presented at the Edinburgh IUHPS/BSECS Conferences, August 1986).
14. Eagleton, Terry, Literary Theory. An Introduction (Minneapolis: Univ. of Minnesota Press, 1983).
15. Foucault, Michel, The Order of Things. An Archaeology of the Human Sciences
(London: Tavistock, 1970).
16. Guerlac, Henry, Antoine-Laurent Lavoisier. Chemist and Revolutionary (New
York: Charles Scribner's Sons, 1975).
17. Hankins, Thomas L., Science and the Enlightenment (Cambridge: Cambridge
Univ. Press, 1985).
18. Hannaway, Owen, The Chemists and the Word. The Didactic Origins of Chemistry
(Baltimore: Johns Hopkins Univ. Press, 1975).
19. Hobsbawm, E, J., The Age of Revolution, Europe 1789-1848 (London: Abacus,
1962).
20. Kirwan, Richard, An Essay on Phlogiston and the Constitution of Acids; to which
are added Notes defending and establishing the Antiphlogistic Theory . .. by
Messrs. de Morveau, Lavoisier, de la Place, Monge, Berthollett and de Fourcroy,
translated into English [by W. H. Nicholson]; with additional remarks and replies
by the author (London, 1789).
21. Lavoisier, Antoine-Laurent et al., Methode de nomenclature chimique (Paris,
1787).
22. Lavoisier, Antoine-Laurent, Elements of Chemistry, tr. Robert Kerr (Edinburgh,
1790; reprinted New York: Dover, 1965).
23. Lavoisier, Antoine-Laurent, Oeuvres, 6 vols (Paris, 1864-93).
24. McCann, H. Gilman, Chemistry Transformed: The Paradigmatic Shift from Phlogiston to Oxygen (Norwood NJ: Ablex, 1978).
25. McEvoy, John G. and James E. McGuire, 'God and nature: Priestley's way of
rational dissent', in Historical Studies in the Physical Sciences, sixth annual volume,
ed. Russell McCormmach (Princeton, NJ: Princeton Univ. Press, 1975),325-404.
26. McEvoy, John G., 'Joseph Priestley: "Aerial philosopher": Metaphysics and
methodology in Priestley's chemical thought, from 1772 to 1781, Part 1', Ambix
25 (1978),1-55.
27. McEvoy, John G., 'Joseph Priestley: "Aerial philosopher" ... Part 2', Ambix 25
(1978),93-116.
324
JOHN G. McEVOY
(1978),153-75.
(1979),16-38.
30. McEvoy, John G., 'Electricity, knowledge, and the nature of progress in Priestley's
thought', British Journal for the History of Science 12 (1979), 1-29.
31. McEvoy, John G., 'Enlightenment and dissent in science: Joseph Priestley and the
limits oftheoretical reasoning', Enlightenment and Dissent 2 (1983), 47-67.
32. McEvoy, John G., 'Causes and laws, powers and principles: The metaphysical
foundations of Priestley's concept of phlogiston', in Science, Medicine and Dissent,
Joseph Priestley (1733-1804), eds. Robert Anderson and Christopher Lawrence
(London: The Wellcome Trust and the Science Museum, 1987),55-71.
33. McEvoy, John G., 'The chemical revolution: Continuity and discontinuity in
scientific change', in New Perspectives on the Chemical Revolution. Osiris, volume
4, ed. Arthur Donovan (Philadelphia: The History of Science Society, 1988).
34. Oldroyd, David R., 'An examination of G. E. Stahl's philosophical principles of
universal chemistry', Ambix 20 (1973),36-57.
35. Perrin, C. E., 'Lavoisier's table of the elements: A reappraisal', Ambix 20 (1973),
95-105.
36. Porter, Roy, and Mikulas Teich, The Enlightenment in National Context (Cambridge: Cambridge Univ. Press, 1981).
37. Priestley, Joseph, The History and Present State of Electricity, with Original
Experiments (London, 1767).
38. Priestley, Joseph, Experiments and Observations on Different Kinds of Air, 3 vols
(London, 1774-7).
39. Priestley, Joseph, 'Experiments relating to phlogiston and the seeming conversion
of water into air', Philosophical Transactions 73 (1783), 398-434.
40. Priestley, Joseph, 'Further experiments relating to the decomposition of dephlogisticated and inflammable air', Philosophical Transactions 81 (1791), 213-22.
41. Priestley, Joseph, Heads of Lectures on a Course of Experimental Philosophy
(London, 1794).
42. Priestley, Joseph, Experiments and Observations Relating to the Analysis of
Atmospherical Air. . .. To which are added Considerations on the Doctrine of
Phlogiston, and the Decomposition of Water (London, 1796).
43. Priestley, Joseph, Considerations on the Doctrine of Phlogiston and the Decomposition of Water, Part 11 (London, 1797).
44. Priestley, Joseph, 'Experiments relating to the calces of metals', New York Medical
Repository 2 (1799),249-57.
45. Priestley, Joseph, 'Experiments on the production of air by the freezing of water',
New York Medical Repository 4 (1801), 17-21.
46. Priestley, Joseph, The Doctrine of Phlogiston Established and that of the Composition of Water Refuted (Northumberland P A., 1803).
47. Schaffer, Simon, 'Priestley and the politics of spirit', in Science, Medicine, and
Dissent, Joseph Priestley (1733-1804), eds. Robert Anderson and Christopher
Lawrence (London: The Well come Trust and the Science Museum, 1987),39-53.
48. Siegfried, Robert, and Betty Jo Dobbs, 'Composition: a neglected aspect of the
chemical revolution', Annals of Science 24 (1968), 275-93.
325
49. Taylor, Charles, Hegel (Cambridge: Cambridge Univ. Press, 1975).

50. Thackray, Arnold, Atoms and Powers: An Essay on Newtonian Matter Theory and
the Development of Chemistry (Cambridge MA: Harvard Univ. Press, 1970).
NICHOLAS JARDINE
THE SIGNIFICANCE OF SCHELLING'S "EPOCH OF

A WHOLLY NEW NATURAL HISTORY": AN ESSAY
ON THE REALIZATION OF QUESTIONS 1
Abstract. This essay argues for the fruitfulness of a historiography of the sciences
centred on changes in the horizons of disciplines through realization and dissolution of
questions. The approach is illustrated by an interpretation of certain developments in
natural history in the German lands in the years 1780-1811.
"By this, indeed, such a system finds the surest touchstone of its truth, that it not only provides a ready
solution to problems hitherto insoluble, but actually
generates entirely new problems ..." (Schelling,
System of Transcendental Idealism (1800) ([63[ 3,
330; tr. [64[ 1
1. INTRODUCTION
In the first part of my essay I shall tell a story about certain developments that occurred in the field of natural history in the German states
around the tum of the eighteenth century. My narration will be
deliberately thin and uninterpretive, a bare rehearsal of opinions and
speculations of a series of authors on the issues of development and
differentiation of living beings. In the second part I shall raise questions
about this chronicle of developments in German natural history. How is
the narrative to be read? What is going on in it? What is its plot? How
may it be elaborated, thickened up into a tale worth telling?
2. A TRANSFORMATION IN NATURAL HISTORY, 1780-1811
The story starts with Johann Friedrich Blumenbach, first a curator of

natural history cabinets then Professor of Medicine at the University
of Gottingen, and one of the principal figures in the late eighteenthcentury movement often referred to by historians as "the revival of
vitalism". Our concern is with his postulation of a formative drive Bildungstrieb - immanent in living matter.
Blumenbach's Ueber den Bildungstrieb, which first appeared in 1780,
criticises preformationist accounts of the generation of animals accord327
328
NICHOLAS JARDINE
ing to which the parts of the adult are formed by accretion of material
to parts preformed in the male or female germ. Instead he defends an
epigenetic account according to which parts are formed by concretion
and organization of originally homogeneous living matter. 2 In the much
revised 1789 edition he declares the following conviction based on
experiment and reflection.
That there is no such thing in nature, as pre-existing germs: but that the unorganized
matter of generation, after being duly prepared, and having arrived at its place of
destination takes on a particular action, or nisus, which nisus continues to act through
the whole life of the animal, and that by it the first form of the animal, or plant is not
only determined, but afterwards preserved, and when deranged, is again restored. A
nisus, which seems therefore to depend on the powers of life, but which is distinct from
the other qualities of living bodies (sensibility, irritability, and contractility), as from the
common properties of dead matter: that it is the chief principle of generation, growth,
nutrition, and reproduction, and that to distinguish it from all others, it may be
denominated the Formative Nisus (Bildungstrieb, or Nisus formativus).3
In this work and in his other natural historical and physiological

writings Blumenbach assigns a central role to the Bildungstrieb in the
explanation of a vast range of phenomena.4 It regulates the generation
and organization of each living being and integrates the activities of
the other vital powers - irritability, sensibility, etc. It governs the
"vegetative processes" of growth, regeneration, nutrition and digestion.
The variety of organic forms results from the variation in direction
(Richtung) of the Bildungstrieb, its direction being affected by the
chemical composition of the generative material, by the action of the
environment, and by diet. Alteration in direction of the Bildungstrieb is
invoked to explain racial differentiation of species; and in later works
such alterations are invoked also to explain the replacements of species
that have accompanied the revolutions in the history of the earth. 5
Blumenbach proposes the formative drive as an heuristic in the
search for empirical laws, analogous in status to Newtonian gravitational attraction:
Obviously for most readers it is a quite superfluous reminder that the word "Bildungstrieb", like the words "attraction" and "gravity", should serve no more and no less than
to designate a force whose constant operation may be recognized from experience, but
whose cause ... is for us a qualitas occulta. Ovid's saying, caussa latet, vis est notissima,
holds for all these forces. The only profit from the study of these forces is to characterize their operations more nearly and to reduce them to general terms. 6
REALIZATION OF QUESTIONS
329
In the sequel Blumenbach illustrates the heuristic power of his postulate
by deriving from observation a number of laws concerning the strength

of the Bildungstrieb and the ways in which deflections of it from its
normal course result in sports, hermaphrodites and monstrous births
([58] 93ff).
To summarize, we have in the Bildungstrieb a regulative principle
governing the production of many of the phenomena of physiology and
natural history, a principle that whilst overtly teleological is conditioned
by and acts through mechanical and chemical causes.
For Kant, as for Blumenbach, the generation and maintenance of
living beings can be understood only in terms of final causality. In his
Critique of Judgement of 1790, he characterizes this mode of causation
as that in which something is both cause and effect of itself - as, for
example, trees both give rise to and originate from trees, and the tree as
a whole both sustains and is sustained by its parts ([31]5, para. 64).
Beings that exhibit such final causality Kant calls "natural ends"
(Naturzwecke).
Natural ends face the student of nature, the Naturforscher, with a
dilemma. On the one hand, "[a]ll production of material things must be
estimated as possible on mere mechanical laws". On the other hand it
follows from the occurrence of natural ends that "[s]ome products of
material nature cannot be estimated as possible on mere mechanical
laws, that is, for estimating them a quite different law of causality is
required, namely that of final causation" ([31] 5, para. 70, p. 387; tr.
[32], part 2, p. 37). No conflict of assertions about nature herself arises,
since these are maxims of reflection, concerning only the ways in which
we are bound by our cognitive constitution to interpret nature. To us
natural ends cannot but appear "accidental" (zufiillig) if considered as
having arisen through the operation of mechanical laws. It is, nevertheless, on the cards that teleology and mechanism are reconciled in the
"inner ground of nature", although that inner ground is inaccessible in
principle to discursive intelligences such as ourselves.7
Although the maxims of mechanism and teleology are not, on Kant's
reckoning, in substantive conflict, they do create an evident practical
dilemma in the study of living beings. In an appendix on "The theory of
the method of applying the teleological judgement" Kant addresses the
issue. His general scheme for a solution is as follows:
So where it is established beyond question that the concept of a natural purpose
applies, as in the case of organized beings, if the student of nature is not to throw his
labour away, he must always in forming an estimate of them accept some original
330
NICHOLAS JARDINE
organization [urspriingliche Organisation [ as fundamental. He must consider that this

organization avails itself of that very mechanism [of phenomena in their particular laws]
for the purpose of producing other organic forms, or for evolving new structures from
those given - such new structures, however, always issuing from and in accordance
with the end in question ([31] 5, para. 80, p. 418; tr. [32] part 2, pp. 77-8).
In a famous passage Kant then ventures what appears to be an extreme

application of his scheme, speculating on the genesis of all types of
living beings from a single original organization, "the womb of mother
earth" ([31] 5, para. 80). He next turns to the crucial question of the
way in which we may envisage the realization of natural ends through
the operation of mechanical laws. Kant briskly dismisses occasionalist
and preformationist accounts of the generation of living beings primarily
on the grounds that they appeal uneconomic ally to "supernatural contrivances"; he then warmly endorses Blumenbach's theory of epigenetic
development through the operation of mechanical laws under the
guidance of a formative drive. s
In the later editions of his Handbook of Natural History Blumenbach
returns the compliment, citing passages from the Critique of Judgement
with approval and writing in an unmistakably Kantian vein about the
way in which the Bildungstrieb connects "merely mechanical" with
purposive modifications in living beings.9
Lenoir argues that this scheme for the subordination of mechanism
to teleology, followed in practice by Blumenbach and made explicit by
Kant, was widely adopted in the German states. It provided the
framework for extensive empirical researches in plant and animal
physiology; and it was an important context for the rich body of early
nineteenth-century speculation about transformation within and between
species ([41], [42], [43], also [38]). It was, however, a scheme that set
strict bounds on the knowledge attainable by the student of nature. In
particular, the scheme excluded as entirely beyond the bounds of
possible experience all questions about the inner nature of the formative drive (and other vital forces) and all questions about the way in
which the original organizations themselves arose.
The rest of our story concerns the way in which Schelling proposed
and Lorenz Oken carried out an extension of natural history into the
territories ruled out of bounds by the Kant/Blumenbach scheme.
Before we consider Schelling's proposal, however, certain more modest
incursions deserve mention.
One who ventured into the forbidden territory was Kant himself. "In
my efforts to utilize if not actually master the Kantian theory, it
331
sometimes seemed to me as if the worthy man were proceeding

roguishly and ironically, at one point appearing to set narrow limits for
our perceptive capacity and at another beckoning us furtively beyond
them" ([24] 232). Goethe had been reading the Critique of Judgement.
One passage to which he alludes is Kant's "adventure of reason"
concerning the production of living beings from the womb of mother
earth, a passage that can all too easily be read as touching on the
illegitimate question of the ultimate origin of living organizations.
Goethe cites also one of the many passages in which Kant introduces
the concept of an "archetypal" non-discursive intellect ([31] 5, para. 77).
With his various "presentiments" about the role of such an intellect as
the inner ground of nature in which teleology and mechanism are
reconciled Kant does indeed point beyond the Kantian ''bounds of
sense". In the Opus posthumum Kant goes much further along these
lines, openly speculating on the creative powers of an animate nature
and on the ground of the analogy between human and natural purposes.!O
Transgressions of a different kind are to be found in the writings of
Karl Friedrich Kielmeyer. In 1793, three years after his return to
Stuttgart from Gottingen where he had studied under Blumenbach, he
delivered an Address on the Relations between Organic Forces (see
[13], [36], [37], [65], [41]158-70). The occasion of the address was the
sixty-fifth birthday of Karl Eugen, Duke of Wiirttemberg, founder and
patron of the ferociously disciplinarian Hohen Karlsschule at which
Kielmeyer was Teacher (Lehrer) of Zoology and Chemistry and Curator
of the natural history cabinets. On the strength of this address Kielmeyer
was to be hailed as a precursor of Naturphilosophie; yet he himself
became a sharp critic of the speCUlative tendencies of the movement.!!
The opening remarks of the Address clearly echo the central tenets
of the BlumenbachlKant programme for seeking within a teleological
framework empirical laws governing the development of living beings
([33] 248-50). On the basis of comparative anatomical and physiological observations Kielmeyer recognises five vital forcesP He then
proposes a series of empirical laws concerning the relations and
interactions of the vital forces. The most general of these laws applies
both to stages in the development of each individual living being and to
the stages in the entire sequence (Stufenfolge) of kinds of living beings.
This is the law of succession of organic forces, according to which
(setting aside some complications) the reproductive (Le., formative)
force at first predominates, then the force of irritability, and finally
332
NICHOLAS JARDINE
the force of sensibility ([34) 261). All this is very much in the
Blumenbachian manner. In notable respects, however, Kielmeyer's
Address transgresses the boundaries set by the BlumenbachlKant
scheme for research. Explicit transgressions include a passing remark
about the way in which one might seek to understand the emergence of
the first organizations on earth ([33) 262), and a tentative speculation
about a Grundkraft of which all forces in nature, organic and inorganic,
are manifestations ([33) 264; cf. [77)70). Implicit violations include the
recurrent reference to organs as "expressions" (Aeusserungen) of organic
forces and the claim that the entire course and condition of living
nature must "arise from" ( hervorgehen ) this balance of forces
([33) 265). Taken literally such language is entirely at variance with the
BlumenbachlKant scheme according to which the vital forces regulate
the organization of living matter and the functions of living organs, but
certainly do not constitute living matter or living organs.
In On the World-Soul of 1798, Schelling's second major essay in
Naturphilosophie, both Kielmeyer's law of succession of organic forces
and his speculations about a Grundkraft are cited with approval. Future
ages, Schelling claims, will recognise Kielmeyer's Address as marking
"the epoch of a wholly new natural history" ([63)2, 565). In his First
Outline of a System of Naturphilosophie of the following year these
themes are much elaborated. The inorganic Kriifte, magnetism, electricity and chemical process, are constructed as successive manifestations of a primordial conflict of forces; the vital Kriifte, sensibility,
irritability and the galvanic or life process are in tum constructed as
their higher analogues; and Kielmeyer's law of succession of vital forces
is derived a priori. Further, the priority of vital forces over vital organs,
adumbrated in Kielmeyer, becomes explicit: "Sensibility is there before
its organs have been formed; brain and nerves rather than being causes
of sensibility are themselves already its product" ([63)3,155).
In this context the programme for the ''wholly new natural history" is
set out. Schelling notes that some have misinterpreted the succession
of organizations (Stufenfolge der Organisationen) as evidence of a
genealogy of types, even supposing all types of living beings to be
progeny of a single ancestral type ([63)3, 62). This is impossible
Schelling claims:
The distinctness of the stages at which we now see the organizations fixed evidently
presupposes a ratio of the original forces peculiar to each one; whence it follows that
nature must have initiated anew each product that appears fixed to us ... ([63) 3,63).
333
Properly understood comparative anatomy and physiology testify not to

a genealogy of species but to development directed towards an original
ideal ([63]3, 64-5; see also [22] 393-5). Given that the various types
of organizations are determined by and expressive of ratios of organic
forces, it should be possible in principle to derive a priori the entire
sequence of types of organizations through which nature strives to
realize the original ideal. Schelling goes on in a striking passage to
contrast such a "history of nature herself" both with the standard
descriptive natural history and with the genealogical natural history that
Kant had proposed:
Natural history has up to now been only the description of nature, as Kant has very
rightly remarked. He himself suggests the name "natural history" for a special branch of
the science of nature, namely knowledge of the gradual alterations that the various
organizations of the earth have undergone through the influence of external nature,
migrations from one climate to another, etc. If only the idea just set out were practicable, the name "natural history" would assume a much higher import, for it would then
actually convey a history of nature herself, namely of how through continual deviations
from a common ideal she gradually brings forth the whole multiplicity of her products
and thus realizes that ideal, not indeed in the individual products, but in the whole. 13
Schelling himself produced no system of natural history. However,

natural history was one of the disciplines in which Naturphilosophie
achieved a substantial following, and the opening decades of the
nineteenth century see the development of a considerable number of
systems of classification of plants and animals that can be read as
attempts to implement Schelling's programme. 14 One of the most
elaborate of these is due to Lorenz Oken, declared follower of
Schelling and at the time of publication of his system Professor of
Natural History and Medicine at the University of JenaY
In an address of 1809 On the Value of Natural History, especially for
the Education of the Germans Oken protests that the cultivation of
natural history only for its practical fruits in medicine and agriculture
leads to a "senseless enumeration, description and naming of animals"
([55]258-9). In place of such an ignoble, "mercenary" (wucherisch)
natural history he pleads for a natural history that will be an integral
part of the new Naturphilosophie ([55]260-3). This noble natural
history will unify the German people with themselves and the world; it
will give them an understanding of their own nature and their relations
to plants and animals; and it will imbue them with manly resignation
334
NICHOLAS JARDINE
when their power falls short of their understanding ([551264 ff). Such
is the natural history elaborated by Oken in his Lehrbuch der
Naturphilosophie, 1809-11.16
Oken's Textbook is ambitious. In it he offers nothing less than a
derivation, from the original zero and its polarization into the positive
and negative series of numbers, of the whole system of "products of
nature", organic and inorganic, together with their principal activities
and functions. The number three plays a central role. From the first,
second and third "potencies" of nature there emanate three histories of
development and differentiation, a history of the heavens, a history of
the earth, and a history of "the organic". The work is set out in 3562
numbered paragraphs, some consisting of single pronouncements, some
containing a cluster of related pronouncements. Many of Oken's dicta
are on a first reading opaque: "The organic must be a vesicle because it is
the image of the planet"; "The animal kingdom is but a dismemberment
of the highest animal, man"; "A fish is a mussel from between whose
shells a monstrous abdomen has grown".
The third volume of the Textbook deals with the development of the
animal kingdom. In its preface Oken boasts that he has "established for
natural history grounds of classification that are entirely new, covering
the whole body and applying to all animals" and that he has "on that
basis arranged all the animals according to their faculties" ([521 3,
iv-v). He goes on, in an aggrieved vein, to insist that he does not dash
off just anything that comes into his head, but that his work is considered and methodical. 17 One of the methods he has followed he calls
"the factual (siichlich) method". It is a constructive method which
"always links the object which follows with the most important of the
preceding ones". His other method, the naturphilosophisch, exploits
correspondences between part and whole in the cosmos. To exemplify
the two methods Oken instances his two derivations of the thesis that
the primordial organism from which all others are composed is a
vesicle (Bliischen). According to the siichlich method "the organic must
become a vesicle, since it is a galvanic process which can take place
only between the elements. The action of air is necessarily an external
one, so it divides the slime inwards into the earthy and the watery, cell
wall and cell content". According to the naturphilosophisch method
"the organic must be a vesicle because it is the image of the planet".
The core of Schelling's technique of Konstruktion is a dialectical
procedure whereby successively "higher" and more specialized natural
REALIZA TION OF QUESTIONS
335
processes and products are derived as the successive partial resolutions

or syntheses of the primordial strife between expansive and contractive
forces. 18 There are evident affinities with Oken's siichlich method. In
carrying out his constructions of natural processes and bodies Schelling
has frequent recourse to analogies between natural products that
are generated at corresponding moments of the successive phases of
construction. Such are the analogies between, on the one hand, the
inorganic forces of magnetism, electricity, and chemical process and, on
the other hand, the "higher" organic forces of sensibility, irritability
and galvanism. Similarly, a substantial role is played in Schelling's
cosmogony by the doctrine of Affinitiitsphiiren, according to which
analogous patterns of differentiation arise in the successive phases of
construction of the cosmos.1 9 Oken's naturphilosophisch method has
much in common with Schelling's constructive use of analogy. (Oken's
methodological debt to Schelling should not, however, be exaggerated.
Underlying Schelling's technique of construction is a substantial body
of metaphysical doctrine - the identity of the real and the ideal, the
correspondence between the epochs of production of nature and the
epochs of development of Geist, Konstruktion as the re-enactment by
the philosopher of the original intuition productive of nature. Though
Oken subscribes to many of the central doctrines of Schelling's
Naturphilosophie, it is far from clear from his writings how far he
understood or endorsed Schelling's metaphysical system. Further, there
are certain aspects of Oken's procedure that have little precedent in
Schelling. Thus there is nothing in Schelling's writings directly comparable to the "mathesis" of the first volume of the Textbook according
to which the entire sequence of products of nature is a copy Nachbild - of an ideal progression of numbers and figures emanating
from God, the original zero; and though the number three plays a not
insignificant role in the architectonic of Schelling's systems, there is
little precedent in Schelling for theosophical reasoning of the kind
Oken uses to ground his theory of the four elements in the Trinity. 20)
The first stage in Oken's developmental history of living beings is the
construction of the vesicles of slime mentioned above. In isolation these
occur in water as Infusoria; variously combined they constitute other
types of organisms. The second stage is the construction of the plant
kingdom. The third stage, which synthesizes the other two, is the
construction of the animal kingdom, culminating in man, the complete
and perfect realization of God. The original ideal, for Schelling
336
NICHOLAS JARDINE
unrealizable in any particular finite product of nature, is for Oken fully

realized in man - or rather, to be precise, in the German warriorhero. 21 The basis of the construction is provided by a ranking of the
organic processes associated with each of the four elements. From this
is derived a partition of "the animal body" into tissues, organs and
organ-systems, themselves ranked according to the rank of the process
or processes that they enact. The system of types of animals is constructed by addition and reduplication of successively higher-ranking
organs culminating in man who possesses all organs in their highest
forms: hence Oken's pronouncement that the animal kingdom is a
dismemberment of the highest animal, man. In the demarcation and
ranking of types it is sense organs, the noblest organs, that provide the
primary criterion.
The principal ranking of types in Oken's system is constituted by the
sequence of seventeen classes, from infusorians to mammals, shown in
Table I. The criteria used to determine and rank the classes of animals
TABLE I
Oken's construction of the series of classes of animals (based on 152, 1843 ed.1 484-5.
In the earlier editions only four circles are recognized).
DOMINANT DOMINANT DOMINANT
ELEMENT SENSE
ORGAN-SYSTEM
Gastric
Alimentary Intestinal Protozoa
Absorbent
Earth
Water
Tactile
Air
Fire
I
I
I
CIRCLES
Taste
Smell
Hearing
Sight
Venous
Vascular Arterial
Cardiac
Conchozoa
Cutaneous
Respiratory Branchial Ancyliozoa
Tracheal
Osseous
Muscular
Sarcozoa
Nervous
Sensory
Aesthesiozoa
CLASSES
I
I
I
I
Infusorians
Polyps
Jellyfishes
Shellfishes
Snails and Slugs
Squids
Worms
Crustaceans
Insects
Fishes
Reptiles
Birds
Mammals
337
are iteratively applied within each class to determine and rank orders,
families, genera and species. The resultant scheme is of extraordinary
complexity. In it there are correspondences between types of different
categories in the hierarchy: such is the parallel between classes of
animals and families of birds and fishes shown in Table II. There are
also correspondences between "lower" and "higher" types of the same
category: such are the parallels between orders of fishes and orders of
birds and between families of fishes and families of birds shown in
Table II. Finally, there are correspondences between stages in the
development of the individual organism and the stages of development
of the entire animal kingdom. In the course of its development an
organism of a given type passes in tum through stages representative of
each of the classes that rank below the class to which it belongs.
All these correspondences are reflected in "analogies" of anatomical
structure; and it is in connection with such analogies that Oken makes
some of his most bizarre-sounding claims - that the limbs of insects
are the ribs of mammals, that the fish is a mussel from between whose
shells a monstrous abdomen has grown, and so-on.
TABLE II
Parallelisms in Oken's system (based on [52, 1831 ed.[ 40; in the first edition only
parallelisms of orders of animals are worked out).
CIRCLES
CLASSES
ORDERS
FAMILIES ORDERS FAMILIES
OF ANIMALS OF ANIMALS OF FISHES OF FISHES OF BIRDS OF BIRDS
Protozoa
Conchozoa
Ancyliozoa
Sarcozoa
Aesthesiozoa
!
!
Infusorians
Polyps
Jellyfishes
II
Worms
Crustaceans
Insects
III
Fishes
Reptiles
Birds
Mammals
IV
V
Eels
Haddock
Gobies
II
Tunnies
Bream
Perch
III
Herrings
Salmon
Pike
IV
Sharks
I
I
!
!
Tree-creepers
Woodpeckers
Cuckoos
Lampreys
Pipefishes
Shad
Shellfishes
Snails and Slugs
Squids
I
I
!
!
Sparrows
Crows
Parrots
I
I
Song-birds
Fly-catchers
Hawks
Geese
Herons
Fowl
Bustards
338
NICHOLAS JARDINE
3. READINGS AND ELABORATIONS
What kind of story has just been told? What is the significance of this
chronicle of opinions on the scope, content and methods of natural
history?
One type of answer can be ruled out with confidence. This can
hardly be read as a story of scientific progress. From the standpoint of
our biology the natural historical concerns of Blumenbach, Kant,
Kielmeyer, Schelling and Oken are entirely alien. This alienation does
not arise from their having given what are, from the standpoint of our
biology, largely false answers to genuine questions. Nor does it arise
from their having addressed what are, from the standpoint of our
biology, genuine but eccentric or uninteresting questions. Rather, I
suggest, the alienation is engendered by their having addressed what are
for the most part, from our standpoint, unreal questions, questions that
do not have true direct answers. If pressed to pass judgement on the
developments retailed here, we may well conclude that in the relevant
period German natural history was not by present-day standards "on
the secure path of a science". But there is little scope for further
evaluation - too few of the protagonists' beliefs were, by our lights,
even candidates for truth.
More profitable perhaps are readings that would cast our story into
one of the various genres of "history of ideas"; for these are genres that
can be pursued without undue regard to the reality of questions or the
truth of answers. In accordance with the traditional type of history of
ideas one may read the story in terms of motifs, themes and topoi, their
sources and their fortunes. To thicken our story into an interesting
specimen of this genre we would have to undertake some prosopography: Who taught whom? Who read whom? Who corresponded with
whom? etc. We would have to embark on some philology, investigating
in detail the roles and connotations of key terms - Organisation,
Stufenfolge, Dijferenzierung, etc. Above all, we would have greatly to
refine (and doubtless correct and qualify) our account of the various
authors' natural historical beliefs, as well as attempting to relate them to
their entire systems of beliefs. In fact, there is a substantial body of
distinguished work in which parts of our story are elaborated along
such lines. I have particularly in mind Reinhard Low's work on Kant's
treatment of living beings, Dorothea Kuhn's studies of Kielmeyer's ideas
about vital forces, the accounts of Kuno Fischer and Dietrich von
Engelhardt of aspects of Schelling's conception of the living world, and
339
Gunther Busse's study of Oken's natural system ([46], [36], [37], [22],
[21], [11 ]).
In accordance with the currently more fashionable "collectivised"
mode of history of ideas our story would be read as a sketch for an
account of the development of certain research programmes (schools,
traditions) in natural history. To thicken the narrative into such an
account the primary task would be to show that the beliefs retailed are
suitably representative of consensuses in communities of inquirers in
natural history and cognate fields. Timothy Lenoir, to whose work this
essay is greatly indebted, has documented the existence of a "Gottingen
School", whose natural historical and physiological researches were
prosecuted under the aegis of the Blumenbach/Kant scheme ([43]). It
is my impression that a school of comparable extent pursued natural
historical activities under the aegis of Schelling's Naturphilosophie in
the German states in the first three decades of the nineteenth century.
Of the many objections that may be levelled against "history of
ideas" approaches to the historiography of science, I believe two to be
of special cogency. First there is the now standard objection that such
approaches are too "intellectualist", failing as they do to take account of
the socially-embodied practices that engender and sustain scientific
beliefs. A second objection is that such approaches skip too blithely
over the dark abyss of time: with their emphasis on continuity, they
familiarize and assimilate too facilely past beliefs that are, properly
understood, entirely alien. The two weaknesses are intimately related,
or so I shall suggest.
Consider yet again some of Oken's wondrous pronouncements: ''The
organic must be a vesicle because it is the image of the planet"; "The
animal kingdom is but a dismemberment of the highest animal, man";
"A fish is a mussel from between whose shells a monstrous abdomen
has grown". I hope my story has shown that the way in which we are
alienated from Oken's beliefs does not have to do primarily with
understanding. We come to understand the questions Oken addressed
by seeing how his methods of inquiry enable him to get to grips with
them. But such understanding does not make the questions any the more
real for us. Nor is the problem one of interest, salience or centrality. The
unreality of Oken's questions for us does not result from our finding
them dull, obscure or marginal. The problem is rather that from our
standpoint the questions Oken addressed simply do not have true direct
answers.
The reading of the story that I advocate construes it as having the
340
NICHOLAS JARDINE
makings of an account of changes in the scope of the discipline of

natural history through validation and invalidation of questions.
It would be a major undertaking to give a proper exposition and
defence of "history of questions" as a historiographical genre and of the
philosophical account of the reality of questions that underpins it. Here
only the barest sketch will be attempted. (These topics are treated more
fully in [30).)
The starting point for an account of reality of questions is the claim
that for a question to be real for members of a community of inquirers
is for there to be evidential considerations that they would acknowledge
as relevant to its resolution. More precise explication faces a series of
difficulties on the issues of the types of evidential considerations that
may be involved. Further technical problems arise in dealing with
questions that have formal presuppositions (e.g., "How does the
Bildungstrieb regulate the activities of the other vital forces?" - a
question which presupposes that there is a Bildungstrieb, that it is a
vital force, that there are other vital forces, that the Bildungstrieb
regulates their activities, etc.) More formidable philosophical problems
arise when we ask how an absolute notion of reality of questions is to
be articulated and, in particular, how such a notion would relate to the
notion of reality of questions relative to a community of inquirers.
Setting aside these problems, we may note that on this account of
reality of questions the methods in which members of a community
acquiesce and the ways in which they apply those methods will be
crucial in determining the range of questions real for them.
What does our story look like when read as a tale of validation and
invalidation of questions? In particular, on such a construal what is the
significance of Schelling's "epoch of a wholly new natural history"?
First let us consider again the BlumenbachlKant programme for
seeking laws governing development and differentiation of living beings
within the teleological framework established by the postulation of
original organizations and a formative drive. This methodological
scheme can be seen as having validated a substantial range of new
questions. Types of physiological and natural historical questions
validated include: questions about the laws governing the operations of
the formative drive and its interactions with other vital forces; questions
concerning the nature and number of the original organizations from
which existing living forms have derived; and questions concerning the
mechanical and chemical states and processes which affect the direction
341
of the formative drive and mediate its operations. On the other hand a
considerable body of fundamental questions is explicitly invalidated,
being declared to be in principle beyond the range of evidential
considerations. Types of questions thus invalidated include: questions
about the underlying reasons for the apparent harmony between
mechanism and purposive formation in living beings; questions about
the inner nature of the formative drive and its immanence in living
matter; and questions about the way in which the original organizations
themselves came into being.
Kielmeyer's Address appears on this reading as in the most literal
sense an original work. His metaphors and speculations render vivid
and challenging questions about the genesis of living organizations and
the inner nature of the vital forces, questions that the Blumenbach/Kant
scheme had ruled out.
In Schelling's early works there is elaborated a rich and powerful
methodology designed to engage with questions of types that the
BlumenbachlKant scheme had placed beyond "the bounds of sense".
Here we find fully articulated and validated questions about the
underlying grounds of purposive development and differentiation in
nature; questions about the origin and inner nature of the life forces;
questions about the ways in which the life forces are materialized as
functioning organs; questions about the genesis of the entire sequence
of living beings. Schelling leaves us in no doubt that his breaching of the
barriers erected by the BlumenbachlKant scheme is a deliberate one:
for example, he remarks bluntly that the answer to those who assume
that it is impossible to "go beyond" the Bildungstrieb are best answered
by doing so ([63) 2, 529).
Oken's Textbook of Naturphilosophie appears on this account as one
of the many attempts by Schelling's followers to answer the vast range
of questions about the origin and differentiation of living beings
validated by the Schellingian methodology.
How should we proceed in fattening up our slender narrative into a
proper essay on change in the horizon of a discipline? An obvious
requirement here, as in the case of the "collectivised" version of the
"history of ideas", is the identification of communities of inquirers
whose researches were guided and informed by the beliefs we have
retailed. As mentioned above, a "Gottingen School" of natural history
and physiology has been well documented; but much further research
on naturphilosophisch natural history and physiology is needed to
342
NICHOLAS JARDINE
establish the existence of a comparable community (or communities) in

the opening decades of the nineteenth century. Further, there is evident
need for detailed study of methodological commitments and practices
in the relevant communities of natural historians. Finally, if the story is
to acquire general historical interest it will have to address itself not
only to the description, but also to the explanation of the changes in
precept and practice that occasioned changes in the ranges of questions
valid for natural historians.
There are two obvious lines along which we might seek a deeper
understanding of the methodology of natural history in the period. One
is to scrutinize more closely the links between the methodologies of
natural history that we have reviewed and the general epistemological
systems of Kant, Schelling and other philosophers of the period. This
line is ably pursued in the work of, for example, Reinhard Low, Kuno
Fischer and Xavier Tilliette ([461, [221, [75]). The other line is to try to
establish connections between these general methodological precepts
and the particular day-to-day practices of natural historians: practices
of classification, practices of anatomical description and illustration,
of dissection, of arrangement of museum specimens (Blumenbach,
Kielmeyer and Oken were all, at some stage in their careers, responsible for natural history cabinets). I know of relatively little work along
these lines. 22 Yet it is, I suggest, study of the embodiment of methodological precepts in specific practices that is needed if detailed
understanding is to be achieved of changes in the horizons of disciplines through validation and invalidation of questions.
On the score of locus of change in disciplinary horizons the story I
have told is potentially misleading. It appears to be a story in which the
ranges of questions valid in a practical discipline - natural history were largely determined by very general and abstract methodological
considerations originated by philosophers. I am fairly sure that this is
not the true situation. I have, in fact, already indicated how Kant's
"methodology of the teleological judgement" many be considered, at
least in part, as an articulation and legitimation of the Blumenbachian
practice of natural history and physiology. Schelling's and Oken's
methodologies are more resistant to such an interpretation. However, it
is my impression that even here abstraction from existing disciplinary
practices plays a substantial role. Schelling's occasional references to
comparative anatomy are significant in this connection. His method of
Konstruktion and Oken's siichlich method as applied to types of living
343
beings may be seen as abstractions from a practice that became

widespread in anatomical treatises at the end of the eighteenth century
- the practice of ordering types from the anatomically simple and
homogeneous to the anatomically complex and differentiated (see [16]
ch. 3; [3] 53-7). More tentatively, I conjecture that Schelling's and
Oken's brands of argument from "structural" analogy are to be seen in
part as abstractions from another practice that gains ground in comparative anatomy in the latter part of the eighteenth century - the
practice of representing superficially diverse organisms as realizations
of the same "ideal" morphological type (see, for example, [3]161-6).
The above remarks suggest strategies that may yield explanations of
change in the scope of natural history. One such strategy is to look at
the ways in which changing exigencies and resources have bought about
changes in disciplinary practices. At this level one might profitably
study the way in which natural historical practice changed in response
to the shifting demands of medicine and agriculture (subjects to which
natural history was, as Oken so bitterly complained, subordinated in the
curriculum); or one might study the impact on classificatory and
curatorial practices of the deluge of new and exotic specimens culled in
this "age of expeditions"; or one might study the response of natural
historical and anatomical practices to changes in techniques and technologies of illustration. Another strategy is to examine more closely the role
of philosophy and general methodology as means of legitimation and
securement of acquiescence in new natural historical methods. Yet
another is to seek practices and methods in cognate disciplines cosmogony, anatomy, physiology, etc. - that provided models for
methodological innovations in natural history.
A further line of investigation that would, I believe, yield considerable explanatory insights in this instance concerns the role of
historiography in "fixing" new methods by securing acquiescence and
competence in their use. Acquiescence in new methods may be secured
by general historical claims about their power and fruitfulness. Thus in
Romantic histories of philosophy, medicine and the sciences the
transition from "merely empirical" methods to the methods of Naturphilosophie was frequently presented as the inauguration of a proper
balance between speculation and experiment and as holding promise of
a restoration of the lost harmony between man and natureP The
retailing of exemplary successes and failures in historical accounts may
serve at once to promote new methods and to educate readers in their
344
NICHOLAS JARDINE
use. 24 In naturphilosophisch works we find this double role of promotion

and didactic instantiation of methods played by such exemplary
successes as Goethe's discovery of the ideal type of the plant and Oken's
(hotly contested) dicovery of the vertebral "analogies" of the skull.
4. CONCLUSION. THE AFTERLIFE OF

ROMANTIC HISTORIOGRAPHY
Schelling and his followers were opposed to merely descriptive and

didactic approaches not only in natural history but also in human
historiography including the historiography of medicine and the
sciences.25 Such "outer" historiography they attributed to "that absence
of ideas that dares to call itself Enlightenment". In its place there was
advanced by Hans Christian Oersted, Dietrich Georg Kieser and
Henrik Steffens, to name but a few of the prime movers of Romantic
historiography of the sciences, a new "spiritual" and "inner" historiography ([48], [35], [72]). At the metaphysical level the history of the
sciences was construed as the striving of Geist to restore the lost unity
of man and nature, as a record of the way in which nature comes to
self-consciousness through man and man to self-knowledge through the
contemplation and study of nature (see [19] pt. 3). At the level of plot
and motif the history of the sciences was conceived organically, in
terms of cycles and epochs of evolution and differentiation of ideas,
and of growth, metamorphosis and decay of disciplines. 26 At the level
of specific interests there is a concern to excavate the original insights
into nature hidden in the myths of antiquity and preserved in folkbeliefs; there is the concern to place developments in the arts and
sciences in a general pattern of cultural and spiritual development; and
there is the concern with changes in the scopes of disciplines and with
the inner connections between different stages of their development.
The "judgemental" and "history of ideas" brands of historiography
mentioned above are descendents of the "outer" historiography of
science inveighed against by Schelling, Steffens, Kieser and others of
their circle. In the ubiquitous and insidious organic metaphors of
historical narrative - growth, decay, development, differentiation - we
have a seemingly ineradicable legacy of Romantic historiography. But
Romantic "inner" historiography of the sciences is extinct. It passed
345
away with the collapse of German idealism and Naturphilosophie in the

second half of the nineteenth century. For better or for worse our
historiography of science, along with our sciences, is now firmly
grounded in what Steffens called "the rubbish dump of sensory reflection". Revival of the metaphysical bases of Romantic historiography of
the sciences would be a Herculean labour. However, I believe it to be
both feasible and fruitful to rescuscitate the practical concern of
Romantic historiography with the inner transformations of scientific
disciplines. In that belief I advocate a historiography of science centered on changes in the horizons of disciplines through the realization
and dissolution of questions.
NOTES
I I warmly thank Gerd Buchdahl for advice on Kantian matters, Sue Morgan for
guidance in Naturphilosophie, and Andrew Cunningham for pouncing on anachronisms.
2 Blumenbach regarded as crucial evidence for epigenesis the capacity of hydras to
regenerate perfect, but smaller, replicas of amputated parts. On eighteenth-century
accounts of the generation of animals see [23), [60).
J [5, 2nd ed.) 24-5; tr. [6120, from the 3rd ed., in which the passage is unchanged.
4 On Blumenbach's natural history and physiology see 140J; 143J ch. 1; [47J; [17J.
5 On replacement of species see especially [8) 20-3.
(, 15, 2nd ed.) 25-6. There are many precedents for appeal to the example of Newton's
postulation of gravitational attraction to justify postulation of forces active in living
beings but of unknown inner nature; for instance, Albrecht von Haller had used the
analogy in support of his postulation of irritability as a force inherent in muscle: see
[60J 33,96-9.
7 [31) 5, paras. 70-8. It has sometimes been thought that since the principle of
causality is a principle of determinate judgement (understanding) Kant is not entitled to
resolve the antinomy by treating the principle of mechanism as a merely heuristic
maxim of reflective judgement. However, as emphasised by [10) 655ff, Kant's principle
of causality does not entail the principle of mechanism since it is merely a precondition
for the determination of empirical laws.
8 [31) 5, para. 81, pp. 412-14. The account of the generation of animals that Kant
himself had presented in his papers on racial differentiation is less unequivocally
epigenesist. He there postulates the presence in the germ of rudiments and predispositions (Keime and Anlagen) responsible for organs and their arrangements.
9 On Blumenbach in relation to Kant see 167), [40).
10 On these aspects of the Opus posthumum see, e.g., [39J, 126), [46).
II For a selection of contemporary appreciations of Kielmeyer see [34) 282-302. For
Kielmeyer's later opposition to the speculative tendencies of Naturphilosophie see
especially his letters of 1804 to Windischmann and 1807 to Cuvier ([34) 203-10,
235-54).
346
NICHOLAS JARDINE
12 Kielmeyer's emphasis on the role of comparison in the derivation of laws is

especially evident in the plan of a course of lectures given at the Hohen Karlsschule in
1790-3 and published in 1814 ([34[ 13-29). A similar emphasis on comparison and
induction is found in the work of Jakob Friedrich Abel, his philosophy teacher at the
Karlsschule (see [1 ]).
13 [63J 3, 68. Kant's remarks to which Schelling refers are in his essay of 1777 "Von
der Verschiedenheit der Racen iiberhaupt" ([31]2, 434n). "The history of nature
[Naturgeschichtel, which we still almost entirely lack, would teach us the alteration of
the form of the earth; and likewise the alteration which the earth's creatures (plants and
animals) have undergone through natural migrations; and the deviations from the
original form of the stem-kind [Urbild der StammgattungJ that have thence arisen. It
would presumably reduce a great array of apparently distinct species to races of the
same kind and would convert the scholastic system [Schulsysteml of description of
nature [NaturbeschreibungJ into a physical system for the understanding."
14 I know of no extended treatment of naturphilosophisch natural history; but see [61J
chs. 4-8; [76J; [58]; 159].
15 On Oken's career see, e.g., [18], [73], [56]. The only detailed recent study known to
me of Oken's system of Naturphilosophie is [11], to which the present sketch is much
indebted. On his systematics of the plant and animal kingdoms see also [69]230-330;
[25]40-5; [58].
16 [52]. Oken's Lehrbuch der Naturgeschichte (1813-26) presents natural history
"merely narrated, without grounds" ([54]1, 1).
17 [52J vi-vii. Oken may well have been stung into this by a review of the first volume
in the Heidelbergische Jahrbiicher der Literatur flir Theologie, Philosophie und Piidagogik for 1810, anonymous but attributed by Pfannenstiel ([57]) to Oken's fellow
Naturphilosoph Joseph Gorres. The review charges Oken with hasty and unsystematic
presentation.
18 The way in which Oken applies his siichlich method owes much to Henrik Steffens'
writings on the history of the earth and of the animal, vegetable and mineral kingdoms:
see, e.g., [70], [71].
19 On AjJinitiitsphiiren see [22] 401 ff.
20 These aspects of Oken's system owe much to the writing of Franz von Baader,
especially [2].
21 As noted by [3]55, Oken's treatment of man as the morphological ideal has much in
common with Herder's treatment of man as the Hauptform of which all other types of
animals are simplifications and degenerations ([28]1, part 3).
zz Information on practical aspects of classification in natural history and anatomy is to
be gleaned from such works as [16], [76J, and [3]; further there is a considerable
literature on the collection of specimens and the production and illustration of natural
history books (many papers on these topics are to be found in Journal of the Society for
the Bibliography of Natural History). Of particular interest in the present context is
Stephen Cross's paper ([15]) on John Hunter, in which the arrangement of the
Hunterian collections is related to general methodological issues.
23 See [10] ch. 3 on Romantic historiography of philosophy; [66J and [27] on Romantic
historiography of medicine; and the important studies of [44J and [19J on Romantic
historiography of the sciences.
347
Histories of natural history which promote the naturphilosophisch approach include

1681 and 1121. Natural history is also dealt with in certain Romantic histories of medicine,
e.g., 1451, [29[.
Controversies provide further contexts for historical treatment of specific natural
historical questions. Notable instances are the dispute between Oken, Goethe and
others about priority in discovery of the vertebral "analogies" of the skull, and the
famous confrontation between Cuvier and Geoffroy St Hilaire on unity of morphological type.
24 For reflections on the role of retrospective attribution of discoveries in the legitimation of techniques and practices see [141 and [621.
25 On parallels between natural and human history in Romantic historiography see,
e.g., [74], 144], and 120].
26 On the motifs of Romantic historiography see [44].
REFERENCES
1. Abel, J. F., Einleitung in die Seelenlehre (Stuttgart, 1796).
2. von Baader, B. F. X., Ueber das pythagoraische Quadrat in der Natur, oder die vier
Weftgegenden (No place, 1798).
3. Balan, D., L 'Ordre et Ie temps. L 'Anatomie comparee et l'histoire des vivants au
XIXe siecle (Paris: Vrin, 1979).
4. Blumenbach, J. F., 'Ueber den Bildungstrieb (Nisus formativus) und seinen Einfluss
auf die Generation und Reproduktion', Gottingisches Magazin der Wissenschaften
und Literatur 1(1780),247-66.
5. Blumenbach, 1. F., Ueber den Bildungstrieb und das Zeugungsgeschafte (Gottingen,
1781,1789,1791).
6. Blumenbach, 1. F., An Essay on Generation, tr. Sir Alexander Crichton (London,
1792) (= 15, 3rd ed.]).
7. Blumenbach, 1. F., Handbuch der Naturgeschichte (Gottingen, 1779, 1782, 1788,
1791,1797, 1799, etc.).
8. Blumenbach, J. F., Beytrage zur Naturgeschichte, Part 1 (Gottingen, 1790).
9. Braun, L., Histoire de l'histoire de fa phiiosophie (Paris: Ophrys, 1973).
10. Buchdahl, G., Metaphysics and the Philosophy of Science. The Classical Origins,
11. Busse, G., Philosophische und geistesgeschichtliche Grundziige der Lehre Lorenz
Okens (unpublished D. Phil Thesis, Freiburg i. Br., 1950).
12. Carus, C. G., 'Unterschied zwischen descriptiver, geschichtlicher, vergleichender
und philosophischer Anatomie, von der Entwickelung dieser einzelnen Methoden
in verschiedenen Zeitaltern, und von der zweckmiissigsten Aufeinanderfo1ge im
Studium dieser verschiedenen Methoden', Literarische Annalen der gesammten
Heilkunde 4 (1826),1-30.
13. Coleman, W., 'Kielmeyer, Carl Friedrich', in Dictionary of Scientific Biography, 16
vols, ed. C. C. Gillispie (New York: Scribner, 1970-80) 7,366-9.
348
NICHOLAS JARDINE
14. Collins, H. M., Changing Order. Replication and Induction in Scientific Practice
(Beverly Hills: Sage, 1985).
15. Cross, S. J., 'John Hunter, the animal oeconomy, and late eighteenth-century
physiological discourse', Studies in the History of Biology 5 (1981),1-110.
16. Daudin, H., Les Methodes de la classification et l'idee de serie en botanique et en
zoologie de Linne Ii Lamark (Paris: Alcan, [1926]).
17. Duchesneau, F., 'Vitalism in late eighteenth-century physiology: the cases of
Barthez, Blumenbach and John Hunter', in William Hunter and the EighteenthCentury Medical World, eds. W. F. Bynum and R. Porter (Cambridge: Cambridge
Univ. Press, 1985),259-95.
18. Ecker, A., Lorenz Oken (Stuttgart, 1880).
19. von Engelhardt, D., Historisches Bewusstsein in der Naturwissenschaft von der
Aujkliirung bis zum Positivismus (Freiburg and Munich: Alber, 1979).
20. von Engelhardt, D., 'Geschichte und Gesellschaft in der Naturforschung der
Romantik', in Gesellschaftliche Bewegung und Naturprozess, eds. M. Hahn and H.
J. Sandkiihler (Cologne: Pahl-Rugenstein, 1981),83-93.
21. von Engelhardt, D., 'Die organische Natur und die Lebenswissenschaften in
Schellings Naturphilosophie', in Natur und Subjektivitiit. Zur Auseinandersetzung
mit der Naturphilosophie des jungen Schelling, eds. R. Heckmann et at. (Stuttgart
and Bad Cannstatt: Frommann-Holzboog, 1985), 39-57.
22. Fischer, K., Schellings Leben, Werke und Lehre (Geschichte der neuern Philosophie, VII) (2nd ed., Heidelherg, 1899).
23. Gasking, E., Investigations into Generation, 1651-1828 (London: Hutchinson,
1967).
24. von Goethe, J. W., Goethe's Botanical Writings, tr. B. Mueller (Honolulu: Univ. of
Hawaii Press, 1952).
25. Gould, S. 1., Ontogeny and Phylogeny (Cambridge MA: Harvard Univ. Press, 1977).
26. Heimsoeth, H., 'Kants Philo sophie des Organischen in den letzten Systementwurfen', Bliitter flir deutsche Philosophie 14 (1940-1),81-108.
27. Heischkel, E., 'Die deutsche Medizingeschichtsschreibung in der ersten Hiilfte des
19. Jahrhunderts', Klinische Wochenschrift 12 (1933), 714-17.
28. Herder, J. G., Ideen zur Philosophie der Geschichte der Menscheit, 4 vols (Riga,
1784-91).
29. Isensee, E., Die Geschichte der Medicin und ihrer Hilfswissenschaften (Berlin,
1840-5).
30. Jardine, N., Horizons of Inquiry (forthcoming).
Preussische Akadcmie der Wissenschaften (Berlin: Reimer and de Gruyter, 190283).
32. Kant, Immanuel, Critique of Judgement, tr. J. C. Meredith (Oxford: Clarendon
Press, 1952).
33. Kielmeyer, K. F., Ueber die Verhiiltnisse der organischen Kriifte unter einander in
der Reihe der verschiedenen Organisationen, die Geseze und Folgen dieser Verhiiltnisse (Stuttgart, 1793); ed. H. Balss, Sudhoffs Archiv 23 (1930), 247-65.
34. Kielmeyer, K. F., Gesammelte Schriften, ed. F. H. Holler (Berlin: Keiper, 1938).
349
35. Kieser, D. G., Entwurf einer philosophischen Geschichte der Medicin (Halle,
1817).
36. Kuhn, D., 'Uhrwerk oder Organismus. Karl Friedrich Kielmeyers System der
organischen Kriifte', Nova Acta Leopoldina n.F, 36 (1970), 157-67.
37. Kuhn, D., 'Der naturwissenschaftliche Unterricht an der Hohen Karlsschule',
Medizinhistorisches lournalll (1976), 319-34.
38. Larson, J. L., 'Vital forces: Regulative principles or constitutive agents? A strategy
in German physiology, 1786-1802', Isis 70 (1979), 235-49.
39. Lehmann, G., Kants Nachlasswerke und die Kritik der Urteilskraft (Berlin: Junker
u. Diinnhaupt, 1939).
40. Lenoir, T., 'Kant, Blumenbach, and vital materialism in German biology " Isis 71
(1980),77-107.
41. Lenoir, T., 'The Gottingen School and the development of transcendental Naturphilosophie in the Romantic era', Studies in the History of Biology 5 (1981), 111205.
42. Lenoir, T., 'Teleology without regrets. The transformation of physiology in
Germany: 1790-1847', Studies in History and Philosophy of Science 12 (1981),
293-354.
43. Lenoir, T., The Strategy ofLife (Dordrecht: Reidel, 1982).
44. Lepenies, W., Das Ende der Naturgeschichte. Wandel kultureller Selbstverstiindlichkeiten in den Wissenschaften des 18. und 19. lahrhunderts (Munich: Hanser, 1976).
45. Leupoldt, J. M., Allgemeine Geschichte der Heilkunde (Erlangen, 1825).
46. Low, R., Philosophie des Lebendigen. Der BegrijJ des Organischen bei Kant, sein
Grund und seine Aktualitiit (Frankfurt: Suhrkamp, 1980).
47. McLaughlin, P., 'Blumenbach und der Bildungstrieb. Zum Verhiiltnis von epigenetischer Embryologie und typologischen Artbegriff', Medizinhistorisches 10urnal17
(1982),357-72.
48. Oersted, H. c., 'Betrachtungen iiber die Geschichte der Chemie', lournal for die
Chemie und Physik 3 (1807), 194-231; reprinted in Der Geist in der Natur
(Leipzig, 1854), vol 1.
49. Oken, L., Obersicht des Grundrisses des Systems der Naturphilosophie und der
damit entstehenden Theorie der Sinne (Frankfurt, 1804).
50. Oken, L., Abriss der Naturphi/osophie bestimmt zur Grundlage seiner Vorlesungen
fiber Biologie (Gottingen, 1805).
51. Oken, L., Ueber den Wert der Naturgeschichte, besonders for die Bi/dung der
Deutschen (Jena, 1809).
52. Oken, L., Lehrbuch der Naturphilosophie (Jena, 1809-11; lena, 1831; Zurich,
1843).
53. Oken, L., Principles of Physiophilosophy, tf. A. Tulk (London, 1847) (Translation
of 152, 3rd ed.1)
54. Oken, L., Lehrbuch der Naturgeschichte (Leipzig, 1813; lena, 1826).
55. Oken, L., Gesammelte Schriften, ed. l. Schuster (Berlin: Keiper, 1939).
56. Pfannenstiel, M., 'Lorenz Oken', Berichte der Naturforschenden Gesellschaft zu
Freiburgi. Br.41 (1951),7-20.
57. Pfannenstiel, M., 'Schriften und varia iiber Lorenz Oken von 1806 bis 1951',
Berichte der Naturforschenden Gesel/schaft zu Freiburg i. Br. 41 (1951), 101-18.
350
NICHOLAS JARDINE
58. Querner, H., 'Ordnungsprinzipien und Ordnungsmethoden in der Naturgeschichte

der Romantik', in Romantik in Deutschland, ed. R. Brinkmann (Stuttgart: Metzler,
1978),214-25.
59. Rehbock, P. F., The Philosophical Naturalists. Themes in Nineteenth-Century
British Biology (Madison: Univ. of Wisconsin Press, 1983).
60. Roe, S. A., Matter, Life and Generation: Eighteenth-Century Embryology and the
Haller- Wolff Debate (Cambridge: Cambridge Univ. Press, 1981).
61. Russell, E. S., Form and Function. A Contribution to the History of Animal
Morphology (London: John Murray, 1916).
62. Schaffer, S., 'Scientific discoveries and the end of natural philosophy', Social
Studies of Science 16 (1986), 387-420.
63. Schelling, F. W. J., Siimmtliche Werke, ed. K. F. A. von Schelling (Stuttgart and
Augsburg,1856-61).
64. Schelling, F. W. J., System of Transcendental idealism, tf. P. Heath (Charlottesville:
Univ. of Virginia Press, 1978).
65. Schumacher, I., 'Karl Friedrich Kielmeyer, ein Wegbereiter neuer Ideen', Medizinhistorisches Journal 14 (1979), 81-99.
66. von Seemen, H., Kenntnis der Medizinhistorie in der deutschen Romantik (Zurich:
Orell Fiissli, 1926).
67. Sloan, P. R., 'Buffon, German biology, and the historical interpretation of biological
species', The British Journal for the History of Science 12 (1979), 109-53.
68. Spix, J., Geschichte und Beurteilung aller Systeme in der Zoologie nach ihrer
Entwicklungsfolge von A ristoteles bis aUf die gegenwiirtige Zeit (Nuremberg, 1811).
69. Stallo, J. B. General Principles of the Philosophy of Nature (Boston, 1848).
70. Steffens, H., Beytriige zur innern Naturgeschichte der Erde (Freiberg, 1801).
71. Steffens, H., Grundziige der philosophischen Naturwissenschaft (Berlin, 1806).
72. Steffens, H., Zur Geschichte der heutigen Physik (Breslau, 1829).
73. Strohl, J., Lorenz Oken und Georg Biichner. Zwei Gestalten aus der Obergangzeit
von Naturphilosophie zu Naturwissenschaft (Munich: Corona, 1936).
74. Temkin, 0., 'German concepts of ontogeny and history around 1800', Bulletin of
the History of Medicine 24 (1950), 227-46.
75. Tilliette, X., Schelling. Une Philosophie en devenir (Paris: Vrin, 1970).
76. Winsor, M. P., Starfish, Jellyfish, and the Order of Life (New Haven: Yale Univ.
Press, 1976).
77. Wolff, C. F., Von der eigenthiimlichen und wesentlichen Kraft der vegetabilischen
sowohl als allch der animalischen Substanz (St. Petersburg, 1789).
NOTES ON THE CONTRIBUTORS
DWIGHT BARNABY studied German literature, history and philosophy at Harvard, M.LT. and Yale before doing research at Cambridge in
the history and philosophy of science under Gerd Buchdahl's guidance.
He has more recently been a student of natural science at the Universities of Berlin and Dusseldorf.
ROBERT BUTTS studied at Syracuse University and the University of
Pennsylvania. He has taught at a number of universities in the United
States of America and in Canada where he is now at the University of
Western Ontario. He is editor of Philosophy of Science, and has written
Kant and the Double Government Methodology (1984), and edited
William Whewell's Theory of Scientific Method (1968), The Methodological Heritage of Newton (1970), and Kant's Philosophy of Physical
Science (1986).
STEPHEN GAUKROGER studied philosophy at Birkbeck College,
London, and then at Cambridge where his PhD. (1977) work on
scientific explanation was supervised by Gerd Buchdahl. He has been a
Research Fellow at Clare Hall, Cambridge and at the University of
Melbourne. Since 1981 he has taught at the University of Sydney. He
has written Explanatory Structures (1978) and edited Descartes (1980).
PETER HARMAN was a colleague of Gerd Buchdahl at Cambridge in
the early 1970s. He is now in the Department of History at the
University of Lancaster. He has written Energy, Force, and Matter. The
Development of Nineteenth-Century Physics (1982) and Metaphysics
and Natural Philosophy (1982), and edited Wranglers and Physicists.
Studies on Cambridge Physics in the Nineteenth Century (1985).
MARY HESSE studied mathematics at Imperial College, London and
history and philosophy of science at University College, London. She
has lectured in mathematics at London and Leeds Universities and
in philosophy of science at London and at Cambridge. From 1960
351
352
NOTES ON CONTRIBUTORS
she worked closely with Gerd Buchdahl in the development of the

Cambridge Department of History and Philosophy of Science, from
which, as Professor of History and Philosophy of Science, she retired in
1985. Her main publications are Forces and Fields (1961), Models and
Analogies in Science (1963), The Structure of Scientific Inference
(1974), Revolutions and Reconstructions in the Philosophy of Science
(1980), and The Construction of Reality (with M. A. Arbib, 1987).
NICHOLAS JARDINE studied natural sciences at Cambridge. Lectures
by Gerd Buchdahl on William Whewell inspired an interest in philosophical aspects of taxonomy, which resulted in his Ph.D. and work as a
Royal Society Research Fellow. Since 1974 he has been in the
Cambridge Department of History and Philosophy of Science. He
co-edits Studies in History and Philosophy of Science and has published
Mathematical Taxonomy (with R. Sibson, 1972), The Birth of History
and Philosophy of Science (1984), The Fortunes of Inquiry (1986), and
Horizons of Inquiry (forthcoming).
OLIVER LEAMAN graduated from Oxford. With constant inspiration
from Gerd Buchdahl he worked on Kant for his Cambridge Ph.D. From
1977-9 he lectured in philosophy at the University of Khartoum and is
now Reader in Philosophy at Liverpool Polytechnic. He has published
An Introduction to Medieval Islamic Philosophy (1985) and Averroes
and his Philosophy (1988).
GEORGE MACDONALD ROSS was a postgraduate student under
Gerd Buchdahl at Cambridge from 1967-9, working on Leibniz's
philosophy of science. He has taught at the Universities of Birmingham
and, since 1973, of Leeds, where he first went as Research Fellow in
History and Philosophy of Science. Besides many articles on Leibniz he
has written Leibniz (1984) and is President of the British Leibniz
Association.
JOHN McEVOY did a degree in chemistry at the University of
Liverpool, after which he studied for the Certificate in History and
Philosophy of Science at Cambridge (1965) where Gerd Buchdahl was
one of his teachers. He has a Ph.D. from the University of Pittsburgh
and since 1972 has taught at the University of Cincinnati. He has
published many papers on the history of chemistry.
353
JURGEN MITTELSTRASS studied at the Universities of Erlangen,

Hamburg, Bonn, and Oxford. Since 1970 he has been Professor of
Philosophy at the University of Constance. Together with Gerd
Buchdahl he is a lifelong student of Leibniz and Kant. Amongst his
many publications are Die Rettung der Phiinomene (1962), Neuzeit und
Aufkliirung (1974), and Wissenschaft als Lebensform: Redern uber
philosophische Orientierungen in Wissenschaft und Universitiit (1982).
MICHAEL POWER read Philosophy, Politics and Economics at
Oxford and was supervised for his Ph.D. (1984) in History and
Philosophy of Science by Gerd Buchdahl at Cambridge. Also qualified
as a chartered accountant he is a Lecturer in Accounting and Finance
at the London School of Economics and Political Science.
WILLIAM SHEA was educated at the University of Ottowa, the
Gregorian University in Rome, and the University of Cambridge where
he was supervised by Gerd Buchdahl for his Ph.D. (1968). Besides
holding research fellowships in America and France he has taught
at the University of Ottowa and is now Professor of History and
Philosophy of Science at McGill. He has written Galileo's Intellectual
Revolution (1972), and edited Reason, Experiment and Mysticism in
the Scientific Revolution (1975), Basic Issues in the Philosophy of
Science (1976), Rutherford and Physics at the turn of the Century
(1979), Otto Hahn and the Rise of Nuclear Physics (1983), and Nature
Mathematized (1983).
RICHARD WESTFALL was educated as an historian at Yale. He has
been in the Department of History and Philosophy of Science at
Indiana University for a quarter of a century, concentrating primarily
on the physical sciences during the seventeenth century. He first met
Gerd Buchdahl in 1961 in Cambridge, when he went there to study the
Newton manuscripts. Amongst his publications are Science and Religion
in Seventeenth-Century England (1958), The Construction of Modern
Science: Mechanisms and Mechanics (1977), Never at Rest: A Biography
of Isaac Newton (1980).
CATHERINE WILSON studied at the Universities of Yale, Oxford,
and Princeton. She has had research fellowships at the University of
Constance and from the American National Endowment for the
354
Humanities. Her contact with Gerd Buchdahl began in 1980 when she
attended his Kant lectures. Since 1978 she has taught at the University
of Oregon. She has written articles on Descartes, and is working on a
book on Leibniz's natural philosophy.
ROGER WOOLHOUSE graduated in philosophy at University College, London, and then went to Cambridge where he worked on
Locke's philosophy of science for a PhD. (1968) under the supervision
of Gerd Buchdahl. He has taught at University College, Cardiff, and,
since 1968, at the University of York. He has written Locke's Philosophy of Science and Knowledge (1971), Locke (1983), The Empiricists
(1988), and edited Leibniz: Metaphysics and Philosophy of Science
(1981).
INDEX
Aarsleff, H. 210
Abel, J. 346, 347
Abrams, M. 210
Adam 79,186,188,189,199
Aepinus 287
Aggiunti, N. 54, 70
Agricola, R. 108, 109
Agrippa, H. 22,78,79,94
Aiton, E. 224
Albury, W. 322
Aldobrandini, Cardinal 63
Alsted, J. 108
Anselm 256
Apel, K-O. 210
Apelles 54, 67
Apuleius, L. 34, 39
Aquinas, T. 34,39,109,112-14,123,
Baier, K 2
Baillet, A. 73,75-77,81,86,88,97,
98
Baker, K 322
Balan, D. 347
Barberini, F. 50,51
Bardi, G. 71
Barnaby, D. 351
Barnouw,J. 210
Baron, W. 39
Bavaria, Duke of 49
Bayle,P. 136,147,169-71,178,180-
82,243
Bayley,S. 174,182
de Beauval, B. 169
Beck, J. 294,297,302,303
Beck,L. 12
Beekman, I. 78,79,94,97
Bellarmine, R. 57,60,62,63
Bennett, J. 12
Bentivogii, O. 47,68
Bergin, T. 192, 208
Berkeley, G. 12,219,249,272,275
Berlin, I. 210
Bernegger, M. 71
Bernoulli, J. 152, 154, 181, 213-24,
124,129
Arbib, M. 210,352
Archimedes 52,69
d'Argensola 70
Aristotle 2, 17-23, 28, 29, 32-5, 39,
58,75,83,91,96,98,105-11,11315,11~121,122,13~15~19~ 194,
197,201,207,209
Arnauld, A. 115, 152, 165, 167-72,
176,177,180,181
Arnold,P. 98
Astrabel 97
Augustine 20,23,24, 39,249
de l'Aunay, G. 95
Ausonius 86
Averroes 19,39,112
Aversa, R. 114
Avicenna 39, 112
238,243
Bird,G. 12
Blanch, J. 114
Blumenbach, J. 327-32,338-42,345,
347
Boethius 107, 108
B6hme,J. 22
Bolgar, R. 39
Bonaventura G. 34,39
Boole, G. 127
Borghese, Cardinal 66
Borgia, Cardinal 63
Borromeo, Cardinal 63,70
Borsche, T. 39
von Baader, B. 346, 347

Bacon, F. 23,32,33,38,185-7,191,
192,197,205,207-9,319
Bacon, R. 96, 98
355
356
INDEX
Bos, H. 224
des Bosses, B. 137,158
Boyle, R. 17,32,33,38,39,209,234
Bramer, B. 97
Brastberger, M. 303
Braun, L. 347
Brengger,1. 37
Brittan, G. 12, 13,275-77,279,303
Broad, C. 182
Brown, S. 182
Brumfitt,l. 322
Bruno, G. 22,25, 39
Bryk, O. 303
Buchdahl, E. 1
Buchdahl, G. 1-7,9-16,93,98,128,
130, 157, 163, 185, 213, 214, 224,
225, 265, 266, 269-74, 276-80,
303,322,345,347,351-4
Buchdahl, M. 1
Buonamici, F. 68
Burke, P. 210
Burman, F. 95,120
Burton, R. 96, 98
Buscherus 108
Busse, F. 288,303
Busse, G. 339, 347
Butterfield, H. 2
Butts, R. 12, 351
Cabero, C. 114
Caccini, T. 55,56,64
Caesarius 109
Calcidius 25, 34, 36, 39
Calvin, 1. 68
Campanella, T. 22,24,27,35,39,84
de Candalle, F. 77
Cantelli, G. 206,210
Capella, M. 36, 39
Cardano,l. 22,23,88
de Careil, F. 74
Carnap, R. 267
Carugo, A. 71
Carus, C. 347
Casilius, A. 114
Cassirer, E. 274,279,322,323
Castelli, B. 50-6,60,68,70
Caton, H. 130
Cesarini, V. 64,65.70
Cesi, F. 49,63,64,68
Chanut, H.-P. 74
Chastel, A. 98
Chiaramonti, S. 54
Christina, Grand Duchess of Lorraine
54,55
Ciampoli, G. 70,71
Cicero, M. 97,98,107,109,206
Cigoli, L. 55,68,70
Clagett, M. 39
Clarke, D. 98, 130, 150
Clarke, S. 135,143,144,151,174,182,
214,223,224,243,275
Clavius, C. 36,37,58,60,67,70
Clerselier, C. 74,95
Cohen, I. 39, 323
Colanna, G. 71
Coleman, W. 347
Coleridge, S. 300,303
Collingwood, R. 188
Collins, H. 348
dele Colombe, L. 51-3, 69
Compotista, G. 108
Conant,J. 2
de Condillac, E. 308, 312, 314. 315,
317,320,323
Conti, Cardinal 55,70
Copernicus, N. 24, 26, 40, 46, 54, 55,
60-67,69,70,299,301
Cornmann,1. 182
Costabel, P. 97,98
Cotes, R. 39
Couturat, L. 150
Cratippus 97
Crombie, A. 71
Crosland, M. 323
Cross, S. 346, 348
Cunningham, A. 345
Curti us, E. 130
Cusanus, N. 17,21,22,24-27,35,40
Cuvier 336,345,347
de Dainville, F. 71
Daudin, H. 348
Davidson, D. 268
Debus, A. 98
INDEX
Dee, J. 77,78,94
De Mas, E. 210
De Mauro, T. 210
Derrida, J. 267
Descartes, R. 12, 29, 30, 33, 37, 40,
66, 71, 73-98, 101-30, 141, 144,
165, 166, 172-79, 181, 182, 188,
197, 201, 207, 209, 229, 234, 26567,269,271,313,353
Desmaizeaux, P. 95
Dewey, J. 267,268
Dicaearchus 97
Diderot, D. 30,40
Diemer, A. 40
Dieteriei, F. 40
Digby, K. 95
Dijksterhuis, E. 40
Dilthey, W. 209,283, 303
Dini,P. 70
Diodati, E. 68,71
Dobbs, B. 324
Donovan, A. 323
Dortles, G. 203,210
Drake, S. 40
Ducasse, C. 182
Duchesneau, F. 348
Dummett, M. 130
Duns Scotus, J. 124
Eagleton, T. 323
Eberhard 238,242
von Eberstein, W. 303
Ecker, A. 348
Eckhart, J. 34,40
von Eckhartshausen, K. 303
Edwards, P. 2
d'Elci, A. 52
d'Elci, O. 70
Elizabeth, Princess 88, 97
Ellis, B. 128, 130
Emerson, W. 38
von Engelhardt, D. 303,338, 348
Eriugena,1. 19,34
Eschenmayer, A. 297-300,303
d'Este, Cardinal 63,70
Euclid 77
Eugen, K. 331
357
Euler, L. 221
Faber, B. 36
Faber, J. 49,68
Fabri, H. 139
Faulhaber, J. 77,85,97
Ferrier,1. 78
Fichte, J. 295,299, 300, 303
Fieino, M. 23,25,40
Fillipi 181
Fiorontino, F. 40
Fisch, M. 192, 195, 208, 210
Fischer 288
Fischer, K. 338,342, 348
Fludd, R. 81
Fonesca, P. 114
de Fontenelle, B. 137
Forster, E. 37
Foscarini, P. 62,65
Foucault,M.317,323
Frankel, M. 208, 210
Frascatoro, H. 22, 96, 99
de Foix de Candalle, F. 77
Foucher, S. 165, 180
Frege,G. 105,127,128,130
de Fresnoy, L. 93
Friedman, M. 13
Fries, J. 303
Funkenstein, A. 211
Gabrieli, G. 71
Gadamer, H.-G. 211
Gale, G. 182
Galen 23,75,113
Galilei, G. 23, 25, 26, 28-30, 36, 40,
45-72, 82, 90, 94, 125, 187, 195,
209,218
Garasse, F. 76,77,94
Garber, D. 174, 182
Gardiner, P. 190, 211
Gasking, E. 348
Gassendi, P. 49,68,82,139,141
Gaukroger, S. 130, 351
Gebhard, H. 149
Gentzen, G. 117
Ghisilieri, F. 64,65
Gibson, B. 2
358
Giggi 70
Gilbert, L. 304
Gillies, S. 33
Gilson, E. 40,82,99,129,130
Girtanner 302
Giuliani, A. 211
Goclenius, R. 40
von Goethe, J. 331,344,347,348
Goldbeck, J. 304
Giirres, J. 346
Gouhier, H. 84, 85, 88, 99
Gould, S. 348
Griiffe, J. 289,304
Grassi, E. 211
Grassi, O. 60,61, 65-8, 70
di Grazia, V. 68
Gregory XIII, Pope 57
Gregory of Nyssa 36
Grienberger, C. 60,67
Grohmann, J. 304
Grosseteste, R. 26, 30, 36, 40
Griinder, K. 42
Gualdo, P. 48,68
Guerlac, H. 323
Guicciardini, P. 70, 71
Guidicci, M. 45,46,67,68
Habermas, J. 211
Hacking, I. 103-5, 128, 131
Hall, A. R. 3,225
von Haller, A. 345
Hankins, T. 323
Hannaway, O. 323
Hanson,N. 3
Hardy, G. 103
Harman, P. 225, 351
Harney, M. 71
Harre, R. 38, 40, 304
Harris, J. 31
Hartley, D. 314
Harvey, E. 131
Hedwig, K. 40
Hegel, G. 33,40,266,302
Heidegger, M. 267,272,302
Heimann, see Harman
Heimsoeth, H. 348
Heischkel, E. 348
INDEX
van Helmont, F. 22, 35
van Helmont, J. 22,35,88,99,141
Hentisbury, J. 77
Herbert, E. 117
Herder, J. 346,348
Herz, M. 232
Hesse,M.21O,211,274,279,351
von Hessen-Rheinfels, E. 151
Hevelius,1. 26
Heydon, C. 37
Heyne, C. 302
Hintikka,J. 12,128
Hobbes, T. 139, 142, 144, 148, 18689,192,195,205,209,211,266
Hobsbawn, E. 323
Hiiffding, H. 182
von Hohenburg, H. 27,37
Hollis, C. 71
Homer 195
Hooke, R. 38,40
Horvarth, J. 37
Howell, W. 129,131,211
Hiibner, K. 41
Huet,D. 81
Hugh of St. Victor 36
von Humboldt, W. 304
Hume, D. 10, 12, 267, 275, 291,
293
Hungerland, I. 209
Hunter, J. 346
Husserl, E. 41,267,270
Huygens, Christiaan 38,41,172,219,
222,223,225
Huygens, Constantijn 81,95,96
Ingoli, F. 70
Isensee, E. 348
Jackson, A. 2
Jakob, L. 288, 302
Jammer, M. 41
Jaquelot, I. 165, 180
Jardine, L. 131
Jardine, N. 348,352
Joad, C. 2
Kambartel, F. 41
INDEX
Kant, L 1, 9, 11-16, 199, 213, 221,
225, 227-304, 329-33, 338-42,
345,346,348,352-54
Kapp, E. 107,131
Karsten, W. 304
Keckermann 108
Kemp Smith, N. 99,245
Kepler, J. 27-30,33,37,41,125
Kielmeyer, K. 331, 332, 338, 341, 342,
345,346,348
Kieser, D. 344, 349
Kirwan, R. 317,323
Kitcher, P. 13
von Kleist, H. 302
Kliigel 288
Kneale,M. 131
Kneale, W. 131
Knowlson,l. 99
Kollner,l. 297,304
Korner, S. 245,264
Krafft, F. 41
Kristeller, P. 41
Kriinitz,1. 304
Kuhn, D. 338,349
Kuhn, T. 299
Laird,1. 2
Lambert of Auxerre 109
Land, S. 211
Larmore, C. 128,129,131
Larson, J. 349
Laudan, L. 131
Lavine, T. 304
Lavoisier, A-L. 307-24
Leaman, O. 352
Lehmann, G. 349
Lehrer, K. 182
Leibniz, G. 12, 13, 15, 35, 37, 38,41,
74,78,84,86,91,99,101,104,12428, 130, 131, 133-62, 165-83, 186,
213-25,227-64,270,352-54
Lenoir, T. 330,339,349
Lepenies, W. 349
Leupoldt, J. 349
Le Vasseur, N. 93
Livarius, A 108
Lichtenberg, G. 287, 302, 304
359
de Lille, A 36, 41
Linceus 48
Lipstorp, D. 85,97
Locke, 1. 12, 106, 153, 191,230,234,
249, 264, 272, 275, 308, 310-12,
314,318,320
Lorini, N. 55, 56, 60-70
Low, R. 338, 342, 349
Loyola, L 56,57,60
Lucretius Carus, T. 25,41
Lukasiewicz, 1. 106, 131
Lully, R. 79, 94
Luther, M. 68,77
McCann, H. 323
MacDonald Ross, G. 352
McEvoy,1. 323, 324, 352
McGuire, J. 323
Mach,E.2,33,41
MacIntyre, A 266,279
McLaughlin, P. 349
MacLaurin, C. 213-15,219-25
MacLean, J. 320
McMullin, E. 197, 211
McRae, R. 12
Macraelius, 1. 41
Magalotti, F. 45,46
Maier, A 41
Maimon, S. 304
Mako,P. 37
Malebranche, N. 165-68, 176, 178,
180
Manzoni, A 46
Marion,l.-L. 99,123,124,129,131
Martin, G. 12
Martin, R. 180
Marzimedici, A 68
Maull, N. 131
de' Medici, C. 63
de Medici, C. 54,71
Melanchthon, P. 109
Mellin, G. 293,305
Mendelssohn, M. 302, 305
Menne, A. 41
Mersenne, M. 38, 66, 71, 82, 89, 90,
94-99,103,104,117,125,129
Mesland, D. 104
360
INDEX
Metzger, J. 288, 305

Micanzio, F. 68,71
Migne,J. 41
Mill,J. 127
Mitchell, D. 71
Mittelstrass,J. 41,353
Moller, R. 305
Monantholius, H. 36, 37
del Monte, G. 28
da Monte Lupo, R. 69
de Montmort, R. 165, 180
Mooney,M.206,211,212
Morgan, S. 345
Mosner 139
Muraro, L. 99
Muth, L. 305
Muti, C. 70,71
Naude, G. 68,76,99
Nazari, G. 88,99
Newcastle, Marquis of 95
Newton, I. 10,25,28, 31-3, 38, 39, 42,
159, 200, 213-16, 218-25, 234,
275, 277, 287, 290, 310, 313, 315,
328,345,353
Nielson, K. 279
Nietzsche, F. 230,243,244,267
Nobis, H. 36,42
Noze, K. 305
Nozzolini, T. 68
Oersted, H. 290, 302, 305, 344, 349
Oken, L. 330, 333-9, 341-4, 346,
347,349
Oldenburg, W. 130
Oldroyd, D. 324
Ong,W.IIO,131
Oresme, N. 27,42
Orsini, A. 66, 71
Ovid 148,328
Padley, G. 211
Paracelsus 75,97
Parsons, C. 12
Paton, H. 243, 244
Pat rizzi, F. 22,23
Patzig,G. 121,131
Paul V, Pope 63
Paul of Venice 109
Paul,G. 2
Paulson, F. 183
Pauly, A. 107
Peghaire, J. 131
Peiresc, N. 49,65,68,71
Pemberton, H. 38
Perez 137
Perrin, C. 324
du Perron 94
Peter of Spain 108, 109
Pfannensteil, M. 346, 349
Pfeiffer, F. 42
Phillip IV, King of Spain 49-51
Phillips, E. 25
Philo of Alexandria 20,42
Philoponus, J. 20,35,42
Picchena, C. 70, 71
Piccolomini, A. 37
Pico della Mirandola, G. 23-27,42
Picot, C. 81, 96
Pigeaud,J. 131
Plass, P. 12
Plato 2, 17-25,27, 31, 36, 37, 110,
113, 129, 141, 142, 153, 194,209,
230,233,235,237,238,265
Plempius, V.-F. 94
Plitt, G. L. 305
Plotinus 20,34,42,227, 242, 244
Poisson, N. 81
Polybius 84, 85
Pompa, L. 197,207,209,211
della Porta, J. 78,83,94,96,99
Porter, R. 324
Power, M. 353
Priestly,J. 307-24
Protagoras 119
Ptolemy 69
Putnam, H. 274
de Puy 94
Pythagoras 22,87,94,96,97
Querengo, A. 63,70
Querner, H. 350
Quine, W. 276,277
Quintillian 110
INDEX
de Raconis, C. 26
Ramus, P. 108-11,113-15,119,120,
122,129,201
Randall, J. 42
Rapp, F. 42
Rehbock, P. 350
Reichenbach, H. 277
Reimarus 302
Reinhardt, L. 128
Reinhold, C. 305
Remond,N. 151,181
Reuss, M. 305
Riccardi, N. 45,46
Rinuccini,G. 61,70,71
Risse,W.113,129,132
Ritter, J. 42
Rocco, A. 69
Rodis-Lewis, G. 99
Roe, S. 350
Rorty, R. 211,265-70,272-74,276,
279-80
Rossi, P. 94,99, 132,211
Roth,P. 97
Ruschio 52
Russell, B. 150, 175, 178, 179, 183.
249,264
Russell, E. 350
Saarinen, E. 132
de Sacrobosco, J. 26, 36, 42
Sakellariadis, S. 99
di Santillana, G. 71
Sarsi, L. 69,70
del Sarto, A. 68
de Saumaise, C. 94
Scaliger, J. 22
Schafer, L. 12
Schaffer, S. 324, 350
Schapiro, B. 211
Schegel, F. 305
Scheiner, C. 54, 66, 67
Schelling, F. 281,289,297,298,30002,305,327, 330, 332-35, 338-44,
346
Schenkel, L. 94
Scherer, A. 305
Schickhardt 68
361
Schlick, M. 267
Schmid, C. 295-97,305
Schmidt-Biggemann, W. 42
Schmitt, J. 288
Schook, M. 95
Schultz, J. 294,305
Schulze, G. 305
Schumacher, I. 350
Schuster, J. 99, 132
Schwab, J. 288, 306
Scot,M. 19
Scott-Taggart, M. 12
von Seemen, H. 350
Sellars, R. 183
Selle, C. 302, 306
Sextus Empiricus 97, 99
Shea, W. 353
Sibson, R. 352
Siegfried, R. 324
Sieyes 302
Simmel, G. 283,301,306
Slaughter, M. 99,211
Sloan, P. 350
Socrates 22,23,89,209
Sophie, Electress of Hanover 141
Sophie Charlotte, Queen of Prussia 243
Spinoza, B. 165,175,178-80,183
Spix,J. 350
Stahl, D. 138, 140, 144
Stahl, G. 313
StalIo, J. 350
Staurophorus, R. 87
von der Stein, A. 42
St Hilaire, G. 347
Steffens, H. 289,302, 344-46, 350
Steno, N. 230
Stichweh, R. 306
Strawson, P. 12,283
Strohl, J. 350
Strong, C. 181,183
Stump, E. 132
Sturm, C. 171
Suarez, F. 57, 114, 123, 124, 149,
152
Swammerdam, J. 230
Swedenborg,E. 240,242,244
Swineshead, R. 77
362
INDEX
Tagliacozzo, G. 211, 212

Talon, O. 109
Tanckius, J. 37
Tarde,J. 59
Tasso, T. 195
Taylor, C. 325
Taylor,D. 2
Teich, M. 324
Telesio, B. 22,23,35
Temkin, O. 350
Tentzel, W. 135
Thackray, A. 325
Theil, C. 43
Theiler, W. 42
Themistius 108
Theophrastus 75
Thorndike, L. 43
Tiberius 97
Tieftrunk, J. 306
Tilliette, X. 306,342,350
Timpler, C. 108
Toletus, F. 114
Toulmin, S. 2
Treviranus, G. 306
Trevisani, F. 88, 99
Trismegistus, H. 77,97
Trommsdorf, J. 288,289,306
Turgot 320
Valla, L. 109
Varignon,P. 137
Vasari, G. 69,71
Vasoli, C. 99
Vaughan, T. 93
Verene 195,197,198,200,207,209,
Vitruvius Pollio 36, 43

de Voider, B. 134,143,158,237,243
Vuillemin, J. 12
Wagner, J.-M. 99
Waismann, F. 252,264
Wallace, W. 43,72
Wallis, J. 1 72
Watts Cunningham, G. 183
Weber, J.-P. 132
WeiHer, K. 306
Weishaupt, A. 306
Weiss, C. 302,306
Weld,C. 43
Welser, M. 48
Werner, C. 302,306
Westfall, R. 353
Whewell, W. 9,352
White, H. 203,211,212
Whitehead, A. 1,265
Will, G. 306
Wilson, C. 353
Winch, P. 209,212
Windischmann, K. 306, 345
Winsor, M. P. 350
Wissowa, G. 107
Wittgenstein, L. 2,204,245, 249, 260-
65,267,268,271,273,282,283
Wolff, C. 247,256,350
Wolters, G. 43
Woolhouse, R. 354
Wren, C. 172
Yates, F. 94,99,132,212
Yolton,J. 128
211,212
Vick,G. 209
Vico,G. 31,43,185-210
Villoslada, R. 72
Vincent of Beauvais 43
Vinta, B. 38,68
Virgil 148
Zeller, E. 43
Zeno 194
Zizka,J. 96
von Zollem, Cardinal 49
Zuniga, D. 62,65
THE UNIVERSITY OF WESTERN ONTARlO

SERIES IN pmLOSOPHY OF SCIENCE
A Series of Books in Philosophy of Science, Methodology, Epistemology,
Logic, History of Science, and Related Fields
Managing Editor:
ROBERT E. BurrS
Editorial Board:
J. BUB, L. 1. COHEN, W. DEMOPOULOS, W. HARPER, 1. HINTIKKA,
C. A. HOOKER, H. E. KYBURG, Jr., A. MARRAS, 1. MITTELSTRASS,
J. M. NICHOLAS, G. A. PEARCE, B. C. VAN FRAAS SEN
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
J. Leach, R. Butts, and G. Pearche (eds.), Science, Decision and Value. lY73,
vii +219 pp.
C. A. Hooker (ed), Contemporary Research in the Foundo.tions and
Philosophy of Quantum Theory. 1973, xx + 385 pp.
J. Bub, The Interpretation of Quantum Mechanics. 1974, ix + 155 pp.
D. Hockney, W. Harper, and B. Freed (eds.), Contemporary Research in
Philosophical Logic and Linguistic Semantics. 1975, vii + 332 pp.
C. A. Hooker (ed.), The Logico-Algebraic Approach to Qunatum Mechanics.
1975, xv + 607 pp.
W. L. H;uper and C. A. Hooker (eds.), Foundations of Probability Theory,
Statistical Inference, and Statistical Theories of Science. 3 Volumes. Vol. I:
Foundations and Philosophy 0/ Epistemic Applications of Probability Theory.

1976, xi + 308 pp. Vol. IT: Foundo.tions and Philosophy of Statistical Inference.
1976, xi + 455 pp. Vol. ill: Foundo.tions and Philosophy of Statistical Theories
in the Physical Sciences. 1976, xii + 241 pp.
C. A. Hooker (ed.), Physical Theory as Logico- Operational Structure. 1979,
xvii + 334 pp.

1. M. Nicholas (ed.), Images, Perception, and Knowledge. 1977, ix + 309 pp.
R. E. Butts and 1. Hintikka (eds.), Logic, Foundations of Mathematics, and
Computability Theory. 1977, x + 406 pp.
R. E. Butts and 1. Hintikka (eds.), Foundo.tiona/ Problems in the Special
Sciences. 1977, x +427 pp.
R. E. Butts and 1. Hintikka (eds.), Basic Problems in Methodology and
Linguistics. 1977, x + 321 pp.
R. E. Butts and 1. Hintikka (eds.), Historical and Philosophical Dimensions 0/
Logic, Methodology and Philosophy o/Science. 1977, x + 336 pp.
C. A. Hooker (ed.), Foundations and Applications of Decision Theory. 2
volumes. Vol. I: Theoretical Foundations. 1978, xxiii +442 pp. Vol. IT:
Epistemic and Social Applications. 1978, xxiii + 206 pp.
R. E. Butts and 1. C. Pitt (eds.), New Perspectives on Ga/ileo. 1978, xvi + 262
pp.
W. L. H;uper, R. Stalrnaker, and G. Pearce (eds.), Ifs. Conditionals, Belief,
Decision, Chance, and Time. 1980, ix + 345 pp.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
J. C. Pin (ed.), Philosophy in Ec01JOT1Ucs. 1981, vii + 210 pp.

Michael Ruse, Is Science Sexist? 1981, xix + 299 pp.
Nicholas Rescher, Leibniz's Metaphysics ofNoJure. 1981, xiv + 126 pp.
Larry Laudan, Science and Hypothesis. 1981, x + 258 pp.
William R. Shea, NoJure MoJhe17UJtized. 1983, xiii + 325 pp.
Michael Ruse, Nature Ani17UJted. 1983, xiv + 266 pp.
William R. Shea (ed.), Otto Hahn and the Rise of Nuclear Physics. 1983,
x + 252pp.
H. F. Cohen, Quantifying Music. 1984, xvii + 308 pp.
Robert E. Butts, Kant and the Double Government Methodology. 1984,
xvi + 339 pp.
James Robert Brown (ed.), Scientific Rationality: The Sociological Turn. 1984,
xiii + 330 pp.
Fred Wilson, Explanation, Causation and Deduction. 1985, xviii + 385 pp.
Joseph C. Pitt (ed.), Change and Progress in Modern Science. 1985, viii + 398
pp.
Henry B. Hollinger and Michael John Zenzen, The NoJure of Irreversibility.
1985, xi + 340 pp.
Kathleen Okruhlik and James Robert Brown (eds.), The Natural Philosophy of
Leibniz. 1985, viii + 342 pp.
Graham Oddie, Likeness to Truth. 1986, xv + 218 pp.
Fred Wilson, Laws and Other Worlds. 1986, xv + 328 pp.
John Eannan, A Primer on Determinism. 1986, xiv + 273 pp.
Robert E. Butts (ed.), Kant's Philosophy of Physical Science. 1986, xii + 363
pp.
Ian B. MacNeill and Gary J. Umphrey (eds.), Advances in the StoJistical
Sciences, Vol. I, Applied Probability, Stochastic Processes, and Sampling
Theory. 1987, xxv + 329 pp.
Ian B. MacNeill and Gary 1. Umphrey (eds.), Advances in the Statistical
Sciences, Vol. n, Foundations of Statistical Inference. 1987, xvii + 287 pp.
Ian B. MacNeill and Gary J. Umphrey (eds.), Advances in the StoJistical
Sciences, Vol. m, Time Series and Econometric Modelling. 1987, xix + 394
pp.
Ian B. MacNeill and Gary J. Umphrey (eds.), Advances in the Statistical
Sciences, Vol. IV, Stochastic Hydrology. 1987, xv + 225 pp.
Sciences, Vol. V, Biostatistics. 1987, xvi + 283 pp.
Sciences, Vol. VI, Actuarial Science. 1987, xvi + 250 pp.
Nicholas Rescher, ScientifIC Realism. 1987, xiii + 169 pp.
Brian Skyrms and William L. Harper (eds.), Causation, Chance, and Credence.
1988, xii + 284 pp.
William L. Harper and Brian Skyrms (eds.), Causation in Decision, Belief
Change and Statistics. 1988, xix + 252 pp.
R. S. Woolhouse (ed.), Metaphysics and Philosophy of Science in the Seventeenth and Eighteenth Centuries. 1988, x + 362 pp.

Woolhouse - Metaphysics and Philosophy of Science 1988

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Woolhouse - Metaphysics and Philosophy of Science 1988

Hochgeladen von

Copyright:

Verfügbare Formate

METAPHYSICS AND PHILOSOPHY OF SCIENCE IN THE

SEVENTEENTH AND EIGHTEENTH CENTURIES

THE UNIVERSITY OF WESTERN ONTARIO

Dept. of Philosophy, University of Western Ontario, Canada

KLUWER ACADEMIC PUBLISHERS

Library of Congress Cataloging-in-Publication Data

MetaphysIcs and phIlosophy of scIence In the seventeenth and

1. Metaphyslcs--Hlstory--17th century. 2. Metaphyslcs-Hlstory--18th century. 3. Sclence--Phllosophy--Hlstory--17th

Published by Kluwer Academic Publishers,

All Rights Reserved

Gerd Buchdahl: Biographical and Bibliographical

NOTES ON THE CONTRIBUTORS

GERD BUCHDAHL: BIOGRAPHICAL

Gerd Buchdahl was born on 12 August 1914 in Mainz, Germany of

Eighteenth Centuries, 1-7.

1988 by Kluwer Academic Publishers.

him - C. E. M. Joad's Guide to Philosophy. As one of the "table

BIOGRAPHICAL AND BIBLIOGRAPHICAL

Lectureship in Philosophy of Science at Cambridge when N. R. Hanson

'Logic and history: An assessment of R. G. Collingwood's Idea of History', Australian

'Inductive process and inductive inference', Australian Journal of Philosophy 34 (1956),

BIOGRAPHICAL AND BIBLIOGRAPHICAL

BIOGRAPHICAL AND BIBLIOGRAPHICAL

GERD BUCHDAHL: A TRIBUTE

Those persons who have been his students and colleagues in

GERD BUCHDAHL: A TRIBUTE

direction in several areas. I learned something from Gerd about

scholars prior to 1969 (the date of publication of Metaphysics and the

GERD BUCHDAHL: A TRIBUTE

and experience, take on a significance hardly even noticed in earlier

GERD BUCHDAHL: A TRIBUTE

interesting and contentious theses of Buchdahl's reading of Kant is what

NATURE AND SCIENCE IN THE RENAISSANCE 1

When we think of nature we usually think of either the green world

both within and outside science. A particular example is found in the

In other words: within the concept of poiesis, nature and human

NATURE AND SCIENCE IN THE RENAISSANCE

activity form a structural unity. In accordance with this conception the

(Tim. 28bff). In Plato, too, describing the genesis of existing things,

NATURE AND SCIENCE IN THE RENAISSANCE

2. NATURE AND ART

Before we turn to a history which, in the Renaissance, eventually leads

the Aristotelian concept of poietic nature and used in substituting the

On the whole, however, Aristotelianism, with the exception of Paduan

NATURE AND SCIENCE IN THE RENAISSANCE

history of science and philosophy is mainly a result of the departure

Giovanni Pico della Mirandola makes even more extensive use of

creates the "cosmic system" in accordance with an archetypal pattern

NATURE AND SCIENCE IN THE RENAISSANCE

Even within this speculative enlargement of the language of natural

NATURE AND SCIENCE IN THE RENAISSANCE

are not presented in the form of textbooks of astronomy (or a textbook

Here Kepler, by picking up a comparison already found in Nicholas of

This is the same metaphor we found in Cusanus, Pico, and Campanella,

By subsuming the Keplerian laws of planetary motion under laws

And this understanding of mechanics did not change as long as

NATURE AND SCIENCE IN THE RENAISSANCE

Surprisingly, the "discovery" of mechanics as a natural science does

intelligibility of nature, which, in its former sense, is understood both as

What is meant here is that the propositions of mechanics are based

NATURE AND SCIENCE IN THE RENAISSANCE

man who blows soap-bubbles into a crowd. Mangogul is in the land of

We find it hard today to understand how such a view of a divine