Vivarium, Vol. 16, Nos. 1-2, 1978 PDF

Vivarium
Volume
16
1978
Reprintedwiththe permissionofthe originalpublisher

by
Periodicals Service Company
Germantown,NY
2013
23:16:58 PM
Printed
onacid-free
paper.
Thisreprint
wasreproduced
from
the
bestoriginal
edition
copyavailable.
NOTETOTHEREPRINT
EDITION:
Insomecasesfullpageadvertisements
which
do notaddto
thescholarly
valueofthisvolume
havebeenomitted.
Asa result,
somereprinted
volumes
pagination.
mayhaveirregular
23:16:58 PM
/';-=09
)(8*=-0/']
23:16:58 PM
VIVARIUM
editors
advisory
committee
publishers
published
AN INTERNATIONAL JOURNAL FOR THE PHILOSOPHY AND INTELLECTUAL LIFE OF THE MIDDLE
AGES AND RENAISSANCE
vivariumis devotedin particularto theprofane side of
mediaevalphilosophy
and theintellectual
lifeoftheMiddle
Ages and Renaissance.
C. J. de Vogel,(Utrecht)- L. M. de Rijk, (Leyden)- H. A.
G. Braakhuis,(Nijmegen)- F. F. Blok, (Amsterdam)
- J.
IJsewijn,(Louvain).
oftheEditorialBoard: Prof.L. M. de Rijk.
Secretary
All communications,
should
exceptthoseofa businessnature,
be addressedto C. H. Kneepkens,KatholiekeUniversiteit,
Erasmuslaan40, 8.26,Nijmegen,
The Netherlands.
- Tullio Gregory,
Marie-Therse
d'AJverny,
(Paris-Poitiers)
(Rome)- Paul OskarKristeller,
(New York)- JanPinborg,
- AlbertZimmermann,
(Copenhagen)
(Cologne).
E. J. Brill,Leiden,The Netherlands.
Twiceyearly,
ca 160pagesyearly.
MayandNovember;
Contributions
submittedto vivarium should preferably
be writtenin English,Frenchor German.The manuscripts
shouldbe typewritten
and doublespaced,exceptforlong
quotationsand footnotes.
Adequatemargins(ijinch) should
be left at each edge of the sheet. Footnotesshould be
numberedcontinuously
each article.Theymay
throughout
be placedeitherat thefootofthepage or at theendofthe
text.
Contributors
receive25 off-prints
freeof charge.
23:16:58 PM
CONTENTS
OF VOLUME XVI (1978)
Calvin bower
Chapei Hill ,
N.C., U.S.A.
Boethius and Nicomachus: An Essay

Concerningthe Sources of De institutione musica
ELIZABETHkarger
Paris
Consequencesetinconsequencesde la suppositionvide dans la logique d'Ockham.
46
E. p. bos
Leiden
Mental Verbsin TerministLogic {John

Buridan, Albertof Saxony, Marsilius of
Inghen)
56
Olga WEIj ers

Voorburg(N.L.)
Contribution l'histoiredes termes*na'

tura naturans' et natura naturata*jusqu' Spinoza
70
l. M. de rij
Leiden
On Ancient and Mediaeval Semantics

and Metaphysics(2)
81
c. h. KNEEPKENS
Master Guido and his View on Govern

ment: On TwelfthCentury Linguistic
Nijmegen
Thought
108
m. L. fuehrer
Minneapolis,
Minn., U.S.A.
Wisdom and Eloquence in Nicholas of

Cusa's Idiota de sapientia and de
mente
142
book reviews
156
23:16:58 PM
XVI, i (1978)
Vivarium,
Boethius and Nicomachus:
An Essay Concerning theSources of De institutionemusica
CALVIN BOWER
Remarquons
que cettepartiede la sciencemusicale,par suited'un vicede
la nomenclature
a
grecque,aggrevencorepar les auteursdu moyen-ge,
t perduependantdes sicles,mleet confondueavec celledes tonsou
chellesde transposition.
Le chaos,dj impntrable
Boce,au Vie sicle
de notrere,n'a commenc
se dbrouiller
que depuisle milieude XVIIIe.1
sentences of Franois Auguste Gevaert, written a century
Theseago, are representativeof most scholarshipconcerningBoethius'
De institutionemusica during the last one hundred years. While
the prejudice of recensiorergodeteriorhas been forsakenin most areas
of classical and medieval studies, it seems to hang on with dogged
tenacity in critical literature concerning Boethius' musical treatise.
The only muscological study which has expressly challenged Gevaert
is Henri Potiron'sBoce, Thoriciende la Musique grecque(Paris, i960).
But Potiron's study is principallyexpositoryin characterand does not
systematicallydeal with the question of Boethius' sources. The most
thorough study concerning sources of Boethius1 musical treatise,
Ubaldo Pizzani's Studi sulle fonti del "De InstitutioneMusica " di
Boezio,2 adopts Gevaers attitude to the point of even citing the
century-oldwork concerning fundamental theoretical matters. The
attitude leads him to a somewhat distortedpicture of the relationship
between Boethius and his sources.3Hence the presentstudy.
I beginthisinquirywith two underlyingassumptions.First,Boethius
1 FranoisAugusteGevaert,Histoireetthorie
dela musiquedeVantiquit,
Gand
1875-81,vol. I, p. 128.
z In: bacns erudrn,16
5-164.
3 I cannotagreewith.(1965),
ofPotiron'sand Pizzani's
J.de Vogel'sbriefassessment
in: Vivarium,10 (1972),37) works(Boethiana
(notPizzano,corrected
, in: Vivarium,9 (1971)52-53).As willbecomeevidentin thepresentstudy,Pizzanidoes
not suggestthat Boethiusbased his workon a Latin source,but ratheron
variousGreekworks,oneofwhichwas translated
intoLatin.ThusbothPizzani
and PotironagreethatBoethiusis nota homouniuslibri.Potiron'scomments
onBoethius'sourcestaketheformofopinionsexpressed,
whereasPizzanibuilds
and theses.Pizzani's argumentsobviouslyovershadowPotiron's
arguments
opinions.
I
23:17:05 PM
was principally a translatorwhen putting togetherthe De institutione

in
arithmetica
musica. The treatise seems to follow the De institutione
the chronology of Boethius' works,4and the arithmeticaltreatise is
recognized to be a translation of Nicomachus of Gerasa )
piOpjTix).5Cassodorus referredto Boethius' works on the various
artesin the followingterms:
enim tais Pythagorasmusicus,Ptolomaeusastronomus
Translationibus
Euclidesaudiuntur
Nicomachus
arithmeticus,
Itali,
geometricus
leguntur
Ausonii.
Boethius' method of composingin his early works is that of compiling
throughtranslation with some commentary;the arithmeticaltreatise
and the logical works 7 clearlydemonstratethispoint. Thus in inquiring
into Boethius' sources I am tryingto determinewhich Greek treatise
Boethius was translatingwhen he compiled his musical treatise.
My second assumption is that Boethius was a conscientious and
competent translator. Boethius characterized his approach to translating the mathematicalworksas adheringto the strictestlaw of translation, but adding for the sake of elucidation, sometimes condensing
when his source became too diffuse,and supplyingchartsand diagrams
forthe sake of clarity:
sedpaululumliberius
memetipsetranslations
. . . artissima
legeconstringo,
a
insisto.Nam et ea, quae de numeris
nonvestigiis,
evagatusalienoitineri,
Nicomachodiffusius
disputatasunt,moderatabrevitatecollegiet quae
4 See S. Brandt,Entstehungszeit
in:
undzeitliche
FolgederWerkevonBoethius,
testsand thechronology
62 (1903),152-154.A. P. McKinlay,Stylistic
Philologus,
in: HarvardStudiesin ClassicalPhilology,18 (1907b
oftheworksofBoethius,
and musicaltreatises
123-156,challengesBrandt'sviewthat the arithmetical
statistics
and areBoethius'firstworks.But McKinlay'sstylistic
belongtogether
can be usedto proveBrandt'sviewifsimilarsubjectmatterin thetwotreatises
is compared,as I willshowin a forthcoming
study.
5 ForeditionseeIntroductioms
K. Hoche,JLeipzig
LibriII, recensuit
artthmehcae
trans,into
to Arithmetic,
1866. See also Nicomachusof Gerasa, Introduction
arithmetic
Greek
in
with
studies
Luther
Martin
by Frank
D'Ooge
Englishby
EglestonRobbinsand Louis CharlesKarpinski,AnnArbor1938,pp. 132-137.
CassiodorusVariaei, 45,4 (ed. Mommsen,
MGH, Auct.ant.12,p. 40).
7 Concerning
0/noetmus
thelogicalworkssee L. M. de Rij, Un thechronology
2 (1964),1-49,125-162.Recentstudiesofthelogical
onlogic, in: Vivarium,
works
workshave arguedthatBoethiusmayhave translatedevenmorethana prinof glosseson
are translations
cipal source,forin somecases his commentaries
et
in: RevueBelgede philologie
Greektreatises;see J.Bidez,BohceetPorphyre,
A Latincommentary
2 (1923),189-201;L. Minio-Paluello,
( ?translated
d'histoire,
in: JournalofHellenic
and itsGreeksources,
byBoethius)on thePriorAnalytics
'
in:
on Aristotle,
Studies,77 (1957),93-102;JamesShiel,Boethiuscommentaries
Mediaevaland RenaissanceStudies,4 (1958),216-244and L. M. de Rijk,Logica
Modernorum
I, Assen1962,28-39.
2
23:17:05 PM
velociusangustiorem
transcursa
aditummediocri
intellegentiae
praestabant
adiectione
resera
rerumnostris
etiamformulis
vi,utaliquandoad evidentiam
ac descriptionibus
uteremur.
(De inst.arith.,praefatio,
4, 28-5,14) 8
In a briefstudy of the relationshipbetween Boethius' De institutione
arithmeticaand Nicomachus' treatise on arithmetic,Frank Egleston
Robbins comments as follows:
A comparisonof the two bookswill convincethe readerthat Boethius
follows
fromfirstto last,expanding
Nicomachus
hereand condensing
there,
as hesaysinhisprefacethathewilldo,butneveraddinganything
essential,
eitheroriginalor derivedfromothersources,thatdepartsfromhismodel.
Boethius expressed his concern forcarefultranslation again in the In
,10and the logical works furtherattest to
Isagogen Porphyrii commenta
Boethius' skill as a translator. Arthur Patch McKinlay sees the influenceof translatingfromGreekto be the essential elementin forming
Boethius1style, and he characterizesBoethius' style of translationas
"literal".11Since Boethius' knowledgeand understandingof Greek and
his ability as a translator are demonstrated in works for which his
sources are extant, I hold that his abilities did not forsakehim when
he wrote concerningthe art of music. Thus I assume that Boethius in
compiling the De institutionemusica followed the "path" if not the
"footprints"of his Greek source.
Since this essay is somewhat expositoryin nature, its organization
must largely followthat of Boethius' treatise. Seven principal sections
will be designated as follows:
I.
II.
III.
IV.
V.
Pattern of citation in the mathematical works

Books I and II
Book III
Book IV
Unity of Books I-IV
8 Page andlinecitationsfollowing
quotesorcitationsofBoethius'mathematical
worksreferto the editionof G. Friedlein,Anicii Manlii TorquatiBoetiide
institutione
arithmetica
libriduo, de institutione
musica libriquinqu
, accedit
Boetii,Leipzig1867.
geometria
quaefertur
9 D'Ooge, op. cit.,p. 132.
10In IsagogenPorphyriicommenta,
ed. S. Brandt,Leipzig 1906,p. 135,5-10:
Secundushie arreptaeexpositionis
labornostraeseriemtranslationis
expediet,
in qua quidemuereorne subierimfidiinterpretis
culpam,cumuerbumuerbo
reddiderim.
cuiusinceptiratioestquodinhisscriptis
expressm
comparatumque
inquibusrerumcognitio
nonluculentae
orationis
quaeritur,
lepos,sedincorrupta
ueritasexprimenda
est.
11McKinlay,op. cit.,pp. 124,127.
3
23:17:05 PM
VI. Ncomachus and Ptolemy

VII. Book V and the originalscope of De institutionemusica
In that my conclusions concerningthe firstthree books are similar to
those of Pizzani and other writers,these sections may be brief and
concise. Since my treatmentof Book IV stands in sharp contrast to
previous scholarship, that section must be the most detailed and
extended.
I.
Patternof citationin themathematicalworks
An initial step in determiningthe relationship between Boethius

and his sources is a description of a certain pattern of citing other
authors that appears in the De institutionearithmeticaand De institutione musica. The De institutionearithmeticais a translation of the
arithmetical treatise of Nicomachus, yet at no time does Boethius
acknowledge that fact apart from the referenceto expanding and
condensingNicomachus in his prefatio.Nicomachus is not even mentioned during the course of Book I, and he is mentioned only three
times in Book II. The firsttwo citations of Nicomachus are merely
"ut ait Nicomachus" clauses, referringto unusual words or accounts
recorded in Nicomachus' treatise (80, 5 and 114, 17-18). The third
citation of Nicomachus accredits him with having discovered a unique
characteristic of arithmetical proportionality,a characteristic NicoAlmost
machus himselfstates has escaped the notice of otherwriters.12
from
taken
over
other
than
are
to
Nicomachus
all references sources
and thus Boethius' treatisebasically predicates
Nicomachus himself,13
of
no
other
treatise
except that of Nicomachus. The signifiknowledge
cant pattern to note is that Boethius only cites Nicomachus when
some aspect of general mathematical theory or language is unique to
Nicomachus. So long as the text is consistentwith the general arithmetical and philosophicalpositionof Neo-Pythagoreanismwhichdominates Nicomachus' text, Boethius apparentlysaw no necessityto refer
to Nicomachus.
This pattern can be furthersubstantiated using Book V of De institutionemusica, forthe last book of the musical treatiseis clearly based
12Eisagogearithmetica
ii, 23; see D'Ooge, op. cit.,p. 269,n. 3.
13Boethiusadds one importantglossconcerning
categoriesin ue vnsntuttone
firstdistinguished
thePythagorean
towhichArchytas
arithmetica
ii,42,according
andthatPlatoandAristotle
sitambiguum),
thetenpraedicamenta
[licetqubusdam
followedhis distinction
(139,9-21).
4
23:17:05 PM

14The
on Book I of Ptolemy's Apfxovix.
followingcitationsofauthors
fifth
book
:
in
Boethius'
occur
Chapter 3.
4.
5.
7.
8.
9.
10.
11.
Aristoxenus,Pythagoreans, Ptolemy
Aristoxenus,Pythagoreans, Ptolemy
Ptolemy
Pythagoreans
Ptolemy, Pythagoreans
Ptolemy
Ptolemy
Ptolemy
Aristoxenus
13.
14. Ptolemy
16. Aristoxenus
17. Archytas
18. Ptolemy, Archytas,Aristoxenus
Ptolemy is clearly cited more than any other source, yet he is never
identifiedas the author upon which the text is based. Boethius attributes theoriesto specific authors only a) where differencesof opinion
between Pythagoreans and Ptolemy arise, b) where theories of the
Pythagoreans and Ptolemy can be used to refutetheoriesof Aristoxenus, or c) where a certain theory is of a personal nature and not
necessarilygenerallyaccepted. So long as theoryin question is consistent with the basic tenets of Pythagorean-Ptolemaicmusical thought,
Boethius gives no citation whatsoever. Furthermore,the mere citation
of a name does not imply that Boethius used that author's work as a
direct source; his source for theories of Aristoxenus, Archytas, and
even the Pythagoreansin Book V is simply Ptolemy's treatise.
If these general principles can be established as governingthe patternof citationin the arithmeticaltreatiseand last book of the musical
treatise,textsforwhich the Greek source is still extant, the same principles should be equally valid in the firstfour books for which no
complete source is extant.
II.
Books I and II
Cassiodorus' reference to the source of the musical treatise as
14For editionsee IngemarDring,Die Harmonielehre

desKlaudiosPtolemaios
,
HgskolasArsskrift
Gtebprg(Gteborgs
XXXVI) 1930. I shall latinizethis
treatiseas Harmonica.
5
23:17:05 PM
15 is
impossible to interpretas any indicative
"Pythagoras muscus"
reference. Although the referencesto "Nicomachus arithmeticus",
"geometricus Euclides", and "Ptolomaeus astronomus" may be taken
to referto the sources of the treatises on these respective arts, the
referenceto Pythagoras must merely be a rhetorical citation of the
attitude characterized by the musical treatise. The sole distraction
from orthodox Pythagoreanismin the musical work is the theory of
Ptolemy, and although Ptolemy is criticalof Pythagoreansconcerning
several matters, he remains faithfulto the crucial Pythagorean doctrine of expressingintervalsas proportionsas opposed to the Aristoxenian method of using unrelated quantities. Moreover, Ptolemy's
insistence on superparticular proportions throughouthis tetrachord
divisions carries one Pythagorean principle furtherthan traditional
Pythagoreanism.Nicomachus of Gerasa has long been consideredthe
source of the essentiallyPythagorean theoryfoundin Books I and II.1
Citations of authors and works in Books I and II are as follows:
Chapter i. Plato, Cicero, Statius
2. Aristotle
3. Ptolemy
4. Ptolemy, De institutionearithmetica
9. Pythagoreans
12. Albinus
20. Nicomachus
24. Albinus
27. Cicero
30. Plato
31. Nicomachus
32. Nicomachus
Book II. Chapter 2. Pythagoras
3. Pythagoras
4. De institutionearithmetica
*
Book I.
15See above n. 6.
16See e.g.,W. Miekley,De Boethnlibriae musicafontibus,
Jena1898;M.
9
Cappuyns,Bobee,in: Dictionaired'histoireet de gographieecclsiastiques,
and theirGreekSources,
Late Latin Writers
(Paris 1912), col. 364; P. CoTircelle,
Massachusetts
trans.H. E. Wedeck,Cambridge,
1969,p. 278; Pizzani,op. cit.,
pp. 10-66.
6
23:17:05 PM
15- De institutionearithmetica
18. Nicomachus
19. Eubulides, Hippasus
20. Nicomachus
27. Pythagoreans,Nicomachus, Ptolemy
31. Aristoxenus
Nicomachus is cited more than any other author in Books I-IV, and
he appears to predominateparticularlyin the firsttwo. The most cited
source, however,is the De institutionearithmetica,a work which Boethius translatedfromNicomachus; the dependence of the arithmetical
treatisethus furtherpoints to Nicomachus as the source forat least the
firsttwo books.
The only extant musical work of Nicomachus is the 'Apjxovixv
a brief work which can be considered a minimal introYx^tpStov,17
duction to Pythagorean musical thought. Definitions by Boethius of
such terms as sound, interval, consonance, dissonance, and types of
voice seem to be taken quite literally from this brief treatise by
Nicomachus.18On the other hand, theoriesattributed to Nicomachus
in Book I, 20, 31, and 32, as well as Book II, 20 and 27, cannot be found
in any extant work of Nicomachus.
Nicomachus' Enchiridionis a workaddressed to a noble lady, written
on her request (JanS. 237, 15), and in the introductorysentencesNicomachus acknowledges that this exposition of music is limited (JanS.
238, 6-7). Furthermore,he promisesa more complete musical treatise,
an ],as soon as he has the time to compose it (JanS. 238, 6ff.).
This promise is repeated throughoutthe work in conjunction with
specificaspects of musical theory: in chapter 3 Nicomachus promises
1) more concerningthe harmony of the spheres (JanS. 242, uff.) ; at
the end of chapter 9 he promises 2) more concerningthe addition of
notes, theirinventors,and the times and circumstancesof theirinven17Ed. Karl von Jan,Musici scriptores
graeci,Leipzig 1895,pp. 235-265(referencesto thiseditionwillbe indicatedwith"JanS."andpageandlinenumber).
I shalllatinizethistreatiseas Enchiridion.
For translation
see Flora R. Levin,
Ph.
and Commentary,
Nicomachus
ofGerasa,Manual ofHarmonics:Translation
D. diss.,ColumbiaUniversity,
New York1967.
18For comparisons
ofvariousparallelpassagesbetweenBoethiusand EnchiridionseePizzani,op.cit.,pp. 35-62.Myforthcoming
translation
ofDe institutione
each
musica(Yale University,
MusicTheoryinTranslation
Series)willdocument
suchdefinition
takenfromNicomachus.
7
23:17:05 PM
tions (JanS. 260, i2f.), as well as 3) a division of the monochord

following Pythagorean principles(JanS. 260, I2ff.); chapter 12 promises 4) more concerningmusical proportions(JanS. 261, 18), as well
as 5) furtherdiscussionof the octave, its merit,and that it consists of
five tones and two semitones rather than six tones (JanS. 264, iff.).
Nicomachus' Enchiridion concludes with an apology for the brevity
of the work, and again promisesa much more complete work on music
(JanS. 265).
A comparison of these promises with the chapters of Boethius'
treatise which cite Nicomachus clearly indicates that Boethius must
have had access to the more extended work. Book I, 20 cites Nicomachus concerningthe additions of stringsto the lyre, giving their
inventors and the circumstancesof theirinvention. This discussion is
not foundin the Enchiridion, but it fulfillspromiseno. 2. Book 1, 31 and
32 cite Nicomachus concerningtheory of consonance and the merits
of various consonances, especially the diapason. Again no such discussion is foundin the Enchiridion, but such a presentationwould result
frompromise no. 5. The theoryattributedto Nicomachus in Book II,
18, 20, and 27 is likewisemissingfromtheEnchiridion; but thediscussion
of consonances foundin these chapters continuesthat ofpromiseno. 5.
Promise no. 1 is fulfilledin Book I, 27, with no specificreferenceto
Nicomachus; for this chapter presents a more accurate picture of
celestialharmonythanthat foundin the Enchiridion.19Books I and II as
a whole representthe referenceto more concerningmusical proportions
of promise no. 4. Only the promise of a monochorddivision according
to Pythagorean principles(promiseno. 3) is not found in Books I and
II.
The obvious source for the firstchapters of Book II is Boethius'
De institutionearithmetica.This work is, in fact, such an integralpart
of the De institutionemusica that the musical work appears to be a
direct continuation of the arithmetical treatise. The arithmetical
treatise, however, is nothingmore than a translation of Nicomachus'
treatise on the same subject. Just as Boethius' musical treatise is
(xouoruo)
dependent on his arithmeticalwork,so Nicomachus' Eic<xrfcyy'
Therefore,since
would have been based on his )
apbfxvjTLXT).20
19See R. Bragard,L'harmonie
des sphresselonBolce, in: Speculum,4 (1929),
206-210.
20I prefer
and referto thelargemusical
to maintainNicomachus'terminology
as do Courcelle(op.cit.,
ratherthanIIspl [zouaix7j,
treatiseas )(jlouctlxt)
to translate
theuseofInstitutio
p. 278) and Pizzani(op.cit.,passim).Concerning
Nicomacho
ex
The
cit.
De
see
(ed. Jan,
42.
Excerpta
(n. 7), p.
Rijk,op.
)
8
23:17:05 PM
Book II is such a logical and necessaryoutgrowthof Book I, and since

Nicomachus- through both cited and uncited sources as well as
- is without doubt the prevailing
through his Eisagoge arithmetica
source forthe theorypresentedin Books I and II, it seems inescapable
to conclude that the more extended musical work promised by Nicomachus served as the principal source fromwhich Boethius translated
and compiled at least the firsttwo books of De institutionemusica.
The fact that Nicomachus is cited only where some theorypeculiar to
him is discussed is consistentwith Boethius' general use of citations.
Boethius' source fortheoriesof Eubulides, Hippasus, and Aristoxenus
was the work of Nicomachus; his citations of Cicero, Statius, and
Albinus were additions of a well read Roman. The act of workingfrom
one source is betrayed in Boethius' second citation of Albinus:
Sed nobisin alienooperenoneritinmorandum.
I. 26 (219,1-2)
The meaningis obvious : Let us quit lingeringin anotherwork and get
back to our centralsource, namely Nicomachus.
III.
Book III
Citations of authors and works in Book III are as follows:

Chapter 1. Aristoxenus,De institutionearithmetica
3. Aristoxenus
5. Philolaus
8. Philolaus
9. Archytas
Nicomachus emergesas the source for the firsttwo books fromboth
citations and uncited passages; but Nicomachus is not cited again
throughoutDe institutionemusica, and no single extant source has
been discovered forBooks III and IV. I propose to argue concerning
the sources of these two books based on the followingthree criteria:
1) sources that can be identifiedas possible sources forshort passages
or definitions;2) cross referencesbetween Books I and II and Books
III and IV which demonstrate dependencies between the books; and
3) the generaltone of Books III and IV in relationto othersources and
the remainderof the treatise.
No extant sources can be found fromwhich passages of Book III
extractsfromthelostEisagoge(see Jan,pp. 225-232);
266-282)mayrepresent
the text,contents,
traditionof
and manuscript
yetmanyquestionsconcerning
thesefragments
remainto be answered.
9
23:17:05 PM
were taken. Moreover, the authors cited in the book are unlikely
sources for the whole of Book III. All citations of Aristoxenus are
refutationsof his position that a semitone is half of a tone. The two
referencesto Philolaus, thoughnot refutations,can be consideredlittle
more than interestinginsertions,21for the theory of these references
is m no way central to the basic contents of Book III. Finally, the
citationofArchytasrefersto a basic geometricaxiom whichis necessary
to prove certain argumentspresentedin the treatise; but the axiom is
presentedhere as being inadequate forprovingthe particular point.22
If one compares these referencesto those otherthan to Nicomachus in
the firsttwo books and to those other than to Ptolemy in Book V,
theyare seen to be presentedin exactly the same manner.Thus it must
be concluded that Boethius used no primarysources for the specific
citations occuring in Book III. Rather Book III was based on some
Greek work which also made referenceto these authors.
An examination of the interrelationshipsbetween Book III and the
firsttwo books is the key to determinethe source of the third book.
De institutionearithmeticais cited in the veryfirstchapter of this book
(269, 9-10), and a dependence on the mathematicaltheoryof the arithmetical treatise and Book II remains evident throughout: the six
continuous tones of II, 31 are cited and reexaminedin III, 3 (273, 2223) ; the number containing the comma discussed in II, 31 is recalled
in III, 4 (275, 13-14); and finally,the axioms concerningproportions
presented in II, 9 are cited and used to prove argumentsconcerning
the size of the semitonein III, 12 (288, 9 and 290, 7-9). In short, the
contents of Book III would be incomprehensiblewere it not for the
mathematicalfoundationof Book II. Furthermore,Book III completes
discussions which were promised in earlier passages: I, 16 (202, 18)
promised conclusive arguments that the tone could not be divided
into half, a proofthat is found in III, 1-2; II, 29 (262, 13) anticipated
and cited the prooffoundin III, 1 that the semitonelies between 18 : 17
and 17 : 16 ; and Aristoxenus' concept of six tones completingthe diapason found in III, 3 completes the referenceto this doctrinefoundin
II, 31 (267, 3-5). Thus just as the firsttwo books are necessaryforan
intelligiblereading of Book III, so Book III completes theory only
anticipated in Books I and II.
21Concerning
oftheseextractsfromPhilolaus
and significance
theauthenticity
trans.Edwin
Loreand Sciencein AncientPythagoreanism,
see WalterBurkert,
Massachusetts
L. Minar,Jr.,Cambridge,
1972,pp. 394-400.
22Cf.Burkert,
op. cit.,pp. 442-447.
10
23:17:05 PM
The general tone of Book III is completely Pythagorean, and no

theories presented as valid in this book are exceptional to theories
of the firsttwo books. Since the position of Book III is whollyconsistent with the basic tenets of Pythagorean theory, no citation of its
primarysource was necessary.Given the Pythagorean tone of the book
and its logical dependence on the previous books and the arithmetical
treatise,one must conclude that it is a continuationof the same source
whichwas used forthe firsttwo books, Nicomachus' Eisagoge musica.23
IV.
Book IV
No student of De institutionemusica can deny that Book IV is the

most difficultpart of the treatiseto read and to relate to othersources.
Pizzani has called this book a ''wild forest",24and he sees it as the only
place in the entirework where Boethius tries to use disparate sources
and, in the end, contaminates the basic integrityof the work.25Lawrence Gushee is equally critical of Book IV, arguing that in this book
Boethius' "dependence on sources of differingpoints of view, content,
or age lead the reader (and the author?) into obscurity and contradiction'' 26It seems, however,that with patience, a glance at Boethius'
early works as a group, and closer examination,one can perceive Book
IV as a unified whole and read its theory as logically related to the
remainderof the work. But I must firstexamine Pizzani's thesis concerningthe structureof Book IV.
Pizzani describes the fourthbook in the followingterms: chapters
1-2 are a "faithfultranslation" of Sectio canonist chapters 3-4 are
taken froma Latin source, namely Mutianus' translation of Gaudentius;28chapters 5-12 are derived fromsome unknownsource;29chapter
23Pizzani,op. cit.,pp. 83-87,
agreesthatNicomachuswas thesourceforBook
III.
24Pizzani,op. cit.,p. 9: "Senza dire che,se si esclude
poi
quellaselva selvaggia
che il quartolibro,la novitrispetto
al trattato
aritmetico
sireduce,insostanza,
all'assunzionedi due in luogodi una sola fonteprincipale..."
25Pizzani,op. cit.,p. 87: ". . . nonsi
nelsoloquartolibro
peritidi contaminare
un materialetrattoinequivocabilmente
da fontidiverse".
2e LawrenceA. Gushee,Questions Genrein MedievalTreatiseson Music,in:
of
derMusikin Einzeldarstellungen,
Gedenkschrift
Leo Schrade,Erste
Gattungen
Folge,Bernand Munich1973,pp. 365-433,p. 380.
27Pizzani,op. cit.,p. 88: "I due
primicapitolinonsonoche una fedeletraduzionedei primiottoparagrafidi un'operache non appartienea Nicomaco,la
Sectiocanonis..."
28Pizzani,op. cit.,pp. 89-105.
2e Pizzani,op. cit.,p. 122.
II
23:17:05 PM
13 is based on Nicomachus' lost musical treatise;30and chapters 14-18

are taken fromPtolemy's Harmonica.31
Before examining each of these claims I would examine the broad
implicationsofthe argumentthat Book IV is taken froma wide variety
of sources. Even Pizzani recognizesthis set of eighteenchapters as the
only place in the musical treatise where Boethius uses more than one
principalsource. In fact,this would be the sole example in all the early
works of Boethius, the mathematical works and the logical works, in
whichthe authorset aside his basic techniqueofsystematicallyworking
throughone particular source at a time and embarked upon a course
of coordinatingand reconciling at least five differentsources. Such a
process is simplynot Boethius' moduscomponendi.Boethius does sometimes bringin referencesto works other than the one with which he is
working,but such referencesare consistentlypassages which reinforce
or complementthe principalsource,and the variant source is generally
cited.32 No extended passage in Boethius' early corpus attempts to
coordinate such a varietyof sources as Pizzani and Gushee would have
us believe concerningBook IV of the musical treatise. Thus fromthe
perspectiveof Boethius' generaltechnique of compilingworksthroughout his early career, the thesis that Book IV is a pasticcio of variant
sources seems very unlikely.
The weeknessof this positionis furtherdemonstratedupon examination of its particulars.I will thus examine each part ofthe "disparate
sources" hypothesis in relation to the six sections of Book IV: A.
introduction,B. notation,C. monochorddivision,D. fixedand movable
notes, F. modal theory,G. intervallic tests.
A. Introduction: Chapters 1-2 of Book IV are certainlydrawn from
Sectio canonist yet I would hesitate to describe them as a "faithful
translation". In the firstplace, the end of each axiom, presentedonly
geometricallyin Sectio canonis, is expanded to include an arithmetical
statement of the argument. But much more significantthan these
arithmetical additions are two rather substantial differencesbetween
the text as found in Boethius and Sectio canonis itself.The firstcon30Pizzani,op. cit.,pp. 121-i22.
31 Pizzani,op. cit.,pp. 124-139.
32See e.g.references
musicai, i (185,9"1?)and i27
to Ciceroin De institutions
(219, 19-25).
33For editionsee JanS. 113-166.For corrective
noteson Jan'seditionand a
TransAn Annotated
of the work,see ThomasJ. Matheisen,
clear translation
s Divisionof a Monochord
lationof Euclid*
, in: Journalof MusicTheory,19
(I975) 236-258.
12
23:17:05 PM
cerns the origins of high and low sounds, and the passages are as
follows:
at(xv
8kxtvifjaecov
a Sed omnismotushabetin se tumveloTcv
ecjiv,
TuuxvTepa
Si igi8kpaiTspai,
xalal [xv
vr- citatelatumetiamtarditatem.
7ruxvTEpat,
motusgravior
al 8kpaiTe- tursittardusinpellendo
ro(pdyyoD,
pou7rotoai
- vayxatov
sonus.Namut tarditasprxipat PapUTpou,
[lvur- redditur
x cuxvoTpov
xal ma stationiest,ita gravitascontigua
pousvai,7ret7rep
7uXetvcv
Velox vero motusacuto 8 taciturnitati.
auyxsivrat
xtvrjaecov,
xal tam voculampraestat.
patOTpwv
'
ixpuTepovq
Xaaavcv
oyxetvTai
xtvrjaecov.
areto jvuTpou
Sovro, Praetereaquae gravisest intentione
to8kPapuTpou
TctTetvofzvou
7rpoa- crescitad medium,quae vero acuta,
aetxiVTjaeco
Seovto. remissionedescrescitad medium.
Tuyxveiv
(JanS.148,9-149,6)
(301,17-23) ''
Two aspects ofhis 'translation" requirecomment: 1) the Latin version,
with the exception of one short sentence, is considerably condensed;
the entireocvayxaiov
clause is omittedin the Latin and the finalsentence
(cgts praeterea) is greatly abbreviated. Secondly, the Greek terms
- are rendered with
7tuxvo<;and pat- frequent and intermittent
substantive forms of tardus and velox- slow and fast- rather than
spissus and rarus, and the order of high and low sounds is reversed.
Here one mighttake the easy way out and accuse Boethius of a mistranslation,but in other contexts Boethius clearly shows a comprehension of these terms;34Boethius was not so careless a translator.
Thus one must at least consider the possibility that the Greek from
whichBoethius was workingwas altered,both abbreviatedand changed
from7tt>xvTepo
and patTspo
to some formof tcc/and paSu^.
Yet a more significant' 'infidelity'1 in the transmissionof the Sectio
canonis text is found in the definitionof consonant and dissonant
sounds:
xal to{xvaufxcpcovou
(xtav
xpaivttjv Consonae quidem sunt, quae simul
interse
. pulsaesuavempermixtumque
ejxotv
7CoiouvTa,
to8k8ia<pa>vou<;
(JanS.149,18-20)
coniunguntsonum. Dissonae vero,
suavem
quae simulpulsaenonreddunt
sonum.(302,2-5)
nequepermixtum
The Euclidian text defines consonant sounds simply as two sounds
34In Book I, 3, bothsetsof termsare used to describe
highand low sounds:
Et si tardusquidemfueritac rariormotus,gravesnecesseest sonosefficiipsa
tarditate
etraritate
pellendi.Sinverosintmotusceleresac spissi,acutosnecesse
est reddisonos.(189, 28-190,2). Book V, 19, 16, the wordspissusis used to
translate7ruxv
in an entirely
different
context,namelythatoftetrachords.
I
23:17:05 PM
which mingle,while dissonant sounds axe thosewhichdo not. Boethius'

text expands this definitionby describingconsonant sounds as both
"pleasant and intermingled",and adds the phrase "when struck at
the same time". Again one mightargue that the new words were added
by Boethius, but such an argumentwould give Boethius considerable
theoreticalpowers. The wordingof this definitionis very close to that
found in Book I, 27, where the terms "permixtum. . . et suavem" and
"simul pulsi" likewise appear:
simulquepulsireddunt
Quotensenimduo nerviuno gravioreintenduntur
quodammodoet suavemsonum,duaequevocesin unumquasi
permixtum
coniunctaecoalescunt;tuncfitea, quae diciturconsonantia.(220,2-7)
The effectof the additional words in the Latin text clearly makes the
definitionof consonance presented in Book IV consistent with that
found in Book I. Nicomachus was the source of that book, and Nicomachus' extant definitionof consonance is strikinglysimilar to that
found both in Book I and here in Book IV :
(cyQei
o ireptôvre
6vre,fifia
8i<popoi
(lv,rcetS)
tpOyyoi
a(x<p<ova
& voetS
orto,
XXr)Xoi
) (acttotjxôavTeS
yxpaOoiv
xpouavre
xal olovjiav.(JanS.262,1-5)
iarW
yevatai
<$>a>v")v
rijv
Furthermore,Nicomachus in one extant passage equates high and low
pitch with fast and slow motions:
. . . xal paSiax;(lvvexOvro
apeiav,Taxoi$ etav.(JanS. 243,9-10)
The effectof these departures fromthe earlier text, whether at the
hand of Boethius or not, is to bring the character of the opening of
Book IV into harmonywith the previous books. One cannot help but
at least suspect the hand of Nicomachus in these changes. The notion
that the openingtwo chapters of Book IV are merelya "faithfultranslation" of Sectio canonis must be rejected in any case.
Notation: The thirdchapter of Book IV reviewsthe notes of the
systematelionin preparation forthe monochorddivision and presents
Greek instrumentaland vocal notation; the fourthchapter is nothing
B.
35The languageusedby Nicomachuswithregardto consonanceand highand

tradition.Theonof Smyrna
a generalPythagorean
low soundmay represent
trans,[anded.] J. Dupuis,Paris1892,
Platonicien,
Philosophe
(TheondeSmyrn,
ofhighand lowsoundsto thePythathefollowing
description
p. 84) attributes
goreans:
6 ?,
paSeia? Pap?
Tapetas(i.vnoTeXeitat
of consonance
occursin the definition
Moreover,the phrase (ia xpooavTe
the
to
84).
(also
p.
Pythagoreans
according
14
23:17:05 PM
other than a diagram of the notation. The presence of Latin names

forthe notes in chapter 3 leads Pizzani to argue that theremust have
been a Latin source forIV, 3-4. The facts that 1) the Latin names used
by Boethius are the same as those used by Martianus Capella, and 2)
Boethius did know (and had cited) theoryin Latin throughthe works
of Albinus, are "eloquent coincidences" leading Pizzani to state that
a Latin source for these chapters is an "incontrovertiblegiven".36
Pizzani sees parallels between a passage in Gaudentius and chapter 3 ;37
thus he argues that the Latin translation of Gaudentius by Mutianus
was the source.38There are indeed parallels- albeit loose- between
the two passages cited by Pizzani, but to describe them as "perfect
textual correspondence"39is a misleading exaggeration. The Gaudentius text refersto signs for only 18 sounds, while Boethius' text presents 28 signs. Boethius' text refersto writingthe signs over a metered
verse, a referencewholly absent fromGaudentius. Moreover,Boethius
presentssigns forthe Lydian mode, while Gaudentius begins with the
Hypolydian. It is difficultto maintain that Gaudentius is in any way
a direct source forthe chapters in question.
Rather than leap to speculative conclusions concerningthe third
and fourthchapters of Book IV I would answer the followingfour
questions: 1) Why reintroducethe names of the notesin both Greek
and Latin- in this place in the treatise? 2) Why introduceGreek nota36Pizzani,op. cit.,p. 94: "Di frontea coincidenzetantoeloquentinon pi
legittimo
parlaredi ipotesi:l'influssosu tuttoil branodi una fontelatina si
imponecommedato di fattoincontrovertibile".
37Pizzani,op. cit.,p. 98; foreditionof Gaudentiussee JanS. 317-356.The
:
parallelpassagescitedby Pizzaniare as follows
8 ol 7raXaii
'ExpvjaavTo
[vfiaai]
np Veteresenimmusiciproptercompenv o7)jjL<xaav
t&vxTcoxaSexa
ne integrasemper
TT)
<6>dium scriptionis,
xa Yp^fxaai,
nominanecesseessetapponere,excotoxocXoujxvot
cnrjxeoi
7uepl vvvjTov.
[zouaixo,
tv gitaverentulasquasdam,quibusnerxOeat
(xouatxcv
o7){xetcv
yrfovz izl vorum vocabula notarentur,
easque
t&v <p
<j7]fxeta>aet
, ; r per generamodosquedivisere,simul
xa0' IxaaTovyppoiTo,
xat vi etiamhac brevitatecaptantes,ut, si
v[xaTa
8 <j7jjxe6)
SvaiT
xai quandomelosaliquodmusicusvoluisTtImyivcbaxeiv
7TOO7){xetouo0at
(pyyov.
(JanS.347,11- set adscriberesuperversumrythmica
metri compositionedistentum,has
17)
sonorumntulasadscriberet.
. . . (308,
23-3097).
38A translation
ofGaudentius
: Institutiones
byMutianusis citedbyCassiodorus
ii, 5 (ed. Mynors,
p. 142).
39Pizzani,op. cit.,p. 99: "Pi che di una
genericaaffinitdi contenutoci
sembrasi debba qui parlaredi una perfettacorrispondenza
testualefrai due
autori".
15
23:17:05 PM
tion? 3) What are possible sources of the notation? 4) What is the

relationshipof these chapters to the whole?
1) The answer to the firstquestion is given in the firstsentence of
chapter 3 :
Restt nuncquoniamsumusnervumsecundumpraedictasconsonantias
per regulamdivisuri,quoniamque necessriossonos tribus generibus
notasapponere,ut,cum
cantilenaeexhibebitistapartitio,musicasinterim
divisamlinemisdemnotulissignaverimus;quod unicuiquenomensit,
facillimepossit agnosci.(308, 18-24)
The treatise has arrived at the point of division of the monochord.
Such a division requires that the notes be named in all three genera.
But no mentionof the names of any notes has occurredsince Book I,
26, forno notes are named in either Books II or III ! It indeed seems
necessary at this point to review the names of the notes, so that they
may be freshin the mind of the reader as he studies the division of the
monochord. Moreover, it would seem necessary for the Latin reader
to have some translation of the Greek names in order to understand
the logic of the system as a whole. Thus the mere presence of Latin
names per se does not argue for a Latin source of the passage.
2) The introductionof Greek notation into this treatise is perhaps
the most perplexingquestion with regardto the work. Extant theoretical works treating notation are all in the Aristoxenian tradition, a
tradition wholly foreignto Boethius' treatise. Treatises of a speculative character such as the Sectio canonis and Ptolemy's Harmonica
seem to avoid the use of notation; and even Aristoxenushimselfcriticizes those theorists who would base their theoreticalconsiderations
on notation.40Yet in the firstcenturiesA.D. the use of notation seems
to have been in the air; with the exceptions of Ptolemy, Clenides,
and Nicomachus' brief Enchiridion, every treatise uses notation, especially in describingthe transpositionsof the modes. As it happens,
notation does become a crucial factor in explicating the modes later
in Book IV, and this seems to be the most acceptable justificationfor
the use of notation in the work. Another reason for using notation is
the very convenience of the system in the elementary monochord
division which followsthese chapters. Thus the principlesof notation
are fittinglyintroduced here at the beginningof Book IV. Yet there
are indications in these chapters on notation which reveal a tradition
10 TheHarmonics
ed. withtrans,by HenryS. Macran,Oxford
ofAristoxenus,
Book
II,
40.
1902:
16
23:17:05 PM
other than that foundin the extant treatises using notation,but these
indications must be discussed under the question of possible sources.
3) Although no single work can be cited as the obvious source of
Boethius' notational discussion, at least three treatises may be cited
as possible texts fromwhich the notational theorywas drawn. Gaudentius' Harmonica introductio,
Bellermann's AnonymusIII, and Alypius'
all
notational
present
Isagoge
theory in the form of sentences describing altered Greek letters representingvarious notes of a specific
ancient mode; moreover, Alypius presents the Lydian mode first,41
and Anonymus III presents a discussion of only the Lydian mode,
very much like Boethius.42The primacy of the Lydian mode in these
Aristoxeniantreatisesis furthertestifiedto by Bachius, who, without
any specificreferenceto the Lydian, uses that mode forall his illustrations of musiceliintervals.43The terminologyforthe lichanoi and paranetai in chapter 3 furtherlinks the descriptions of the notes to the
traditionof these treatises,for,similar to the notational treatises,the
names lichanoi and paranetai are not used at all, and the notes are
merelycited as hypaton enharmonios,hypaton chromatice,or meson
chromatice,meson diatonos, or hyperboleonenharmonios,hyperboleon
chromatice.Finally, the use of the term hyperdorianas equivalent to
mixolydian in the notational charts at the end of Book IV further
reflectsthe terminologyof the notational treatises (343, addenda).
Yet minor differencesin both the descriptions of the symbols and
the symbols themselvesraise serious doubt as to whetherany of the
sources mentionedare the actual source of Boethius' thirdand fourth
chapters.44Perhaps an even more serious question is raised by the fact
that the three genera are integrated into one list in Boethius' text,
while the notational treatisesexplicate each genus as a separate entity.
Thereforeit seems unlikelythat any of the treatises cited served as a
source for the passage in question. They all representa tradition of
musical theoryessentiallyopposed to that found in the De institutione
41For editionsee JanS.357-406.
" The originaleditionoftheGreekanonymi,F. Bellermann,
Anonymi
scriptio
demusica(Berlin1841),has beensuperseded
by DietmarNajock,Dreianonyme
Traktate
berdie Musik,Eine kommentierte
griechische
NeuausgabedesBellermannschen
Musikwissenschaftliche
Anonymus(Gttinger
Arbeiten,Band 2),
Gttingen
1972;foruse ofLydianmode,see pp. 116-110.
43Foreditionsee JanS.283-316;foruse ofnotationsee,e.g.pp. 294-295.
44Forexample,Alypiusdescribesthe
symbol4 as 9)
|eXir)Tixv
xa0eiXxuo(ivov
(JanS. 369,25), whereasBoethiusdescribesit as "ny inversumdeductum"
tritehyperboleon
(311,14). Theinstrumental
accordingto Alypiusis formedIY (314).
(JanS. 369, 16), whereasaccordingto Boethiusthesymbolis written
17
23:17:05 PM
musica, and, given Boethius' manner of citingother writers,Boethius

would have cited any theoristwho was not consistentwith the Pythagorean-Ptolemaic position of the treatise.Thus Boethius' presentation
of notation seems to reflecta slightly differenttradition than that
found in Aristoxenian works, and is thus rooted in a text which is no
longer extant.
4) The question of relationshipof Book IV, 3-4 to the treatise as a
whole has been partially answered in explainingwhy the names of the
notes and the notational system were introduced at this particular
moment in the treatise. These chapters are obviously linked with
both the division of the monochord and the discussion of the modes
through their review of the names of the notes and their exposition
of notation. Yet these chapters seem to be related equally to Book I.
Althoughchapter 3 presentsa systemofnamingnotes which resembles
the notational treatises, it presents the notes of the three genera
integrated into one list. Moreover, chapter 4 returns to the terminology of the notes which was found in Book I, 22, and the notes as
presented here add up to the twenty-eightwhich are specifically
cited in I, 22 :
nominum
et differentiam
In quibuset similitudinem
; ut si nervi
pernotabis
fiantsimul
similesin omnibuscum eis, qui sunt dissimiles,
colligantur,
omnesoctoet viginti.(215,16-19)
notes in Book IV, carried over from
The adherence to twenty-eight
Book I, is a distinctlyPythagorean imprinton the text; for twentyeight is a perfectnumberin Pythagoreanarithmetic.45
Therefore,although a certain disparate element is introduced into
De institutionemusica in the discussion of Greek notation, that disparate element is presentedin a way consistentwith the treatise as a
whole rather than with sources of notational theory. The question
remains whetherthe hand of Boethius changed this material to make
it consistentwith the remainderof his work, or whetherthe material
46 Concerning
i, 16, and
Eisagogearithmetica
perfectnumberssee Nicomachus
arithmetica
BoethiusDe institutione
i, 20 (41-42).D'Ooge, op. cit.,p. 209,transwithitselfthe
lates Nicomachusas follows:"Now whena number,comparing
of all the factorswhosepresenceit will admit,neither
sum and combination
exceedsthemin multitudenor is exceededby them,thensuch a numberis
as onewhichis equalto itsownparts.Suchnumbers
saidto be perfect,
properly
has thefactors
fourteenth,
axe6 and 28. . . . Twenty-eight
seventh,
half,fourth,
make28,
whichare 14,7, 4, 2, and 1; theseadded together
and twenty-eighth,
and so neitherare thepartsgreaterthanthewholenorthewholegreaterthan
is in equality,whichis thepeculiarqualityofthe
theparts,buttheircomparison
perfectnumber".
18
23:17:05 PM
was already changed in Ms Greek source. The immediate response to

this question is that Boethius does not play such a stronghand in his
early works.
C. Monochord division: The division of the monochord formsthe
heart of Book IV, and indeed if the opening sentences of Book IV are
to be taken literally, the monochord division forms something of a
climax to the treatise as a whole; it is "the division of the rule toward
which our whole effortis directed":
Etsi omnia,quae demonstranda
libritractatione
erant,superioris
digessimus, non paenitettamenrursuseadem brevitermemoriaerecolligenda
tractatus,ut his rursusad memoriam
praestarecum quadem diversitate
redeuntibus
ad regulaedivisionem
veniamus.(301,
quo totatenditintentio,
7-12)
Moreover, the reader must assume that this rule is that same rule
introduced in Book I, 11, the rule discovered by Pythagoras, "concerningwhich we will speak later, by which we measure the sizes and
sound of notes" :
Itaque invenitregulam,de qua posteriusloquemur. . . , perquam magnitudinescordarumsonumquemetimur.
(198,23-26)
In the Enchiridion (JanS 260, I2ff.) Nicomachus had likewisepromised
a division of the monochord to be included in his Eisagoge, and this
divisionis likewise associated with the name of Pythagoras. Therefore,
if the principlesdeterminingthe division of Book IV can demonstrably
be shown to be those of Nicomachus, then we may conclude that
Nicomachus' Eisagoge is the source forthe heart of Book IV.
In his principal passage concerning the monochord, Nicomachus
promises a division of the canon "rigorouslyfashioned in accordance
with the design of this master Pythagoras, not as Eratosthenes or
Thrasyllus misinterpretedit, but as the Locrian Timaeus understood
it, whom Plato followed,up to the twenty-seventhmultiple":46
xal 7Tpoaex7)a{xc0a
tvtou
xavvo
Xeyopivou
xaTocrofrjv
xp^co
xalxart ouX7)fxa
$touSiSaaxXou
auvrTeXeajziv7)v,
% 'EpaTOcGvrj
XX*
a> AoxpTâio, xal IIXTcov
7J0pauXXo,
7cap^xouaev
7rapY]xoXo'jIctou717.
07jaev,
(JanS.260,12-17)
Yet another passage from the Enchiridion mentions the monochord
division to come, and one principleof Pythagoras' approach is stated
46Translation
:
byFloraRoseLevin,Nicomachus
ofGerasa ManualofHarmonics
andCommentary
Translation
(diss.ColumbiaUniversity,
1967),p. 47.
*9
23:17:05 PM
clearly: Ptyhagoras firstdeterminedthe division of the diatonic genus,

and from this division he determinedthat of the chromatic and enharmonicgenera:
tivIcpuaixfl
<*7u
apuTOCTOu
xaT
m toctov
dcvdyxfl
Ty)v8kTup0aotv
toutoTStaTOVtxv
xal vapfxvtov
yvo;epiaxe.(t yp xp<|aTixv
xtotouSieTpvcooev
a07TOTS
cbcTat7tot
SeaLcroi.(JanS.249,2-6)
yvo
The process of derivingthe chromaticand enharmonicgenera fromthe
diatonic is partially explained in the Enchiridion. The thirdto lowest
note (lichanos or paranete) of each tetrachordin the chromaticgenus
is altered in relation to the diatonic genus, but the second to lowest
note (parhypate or trite) of each tetrachordremains the same in the
diatonic and chromatic genera; that same pitch (second to lowest of
diatonic and chromatic genera) becomes the third to lowest note
(lichanos or paranete) in the enharmonicgenus. The departure from
diatonic proceeds throughthe chromatic to the enharmonic:47
8k8erepo
v{iivo>v
TpTo
^P^ t Skxtovov,
7)XX*f*)
<00
v8
aT
StccTovtxc
8k vapjxovou
jxetvev,
(jLOTOvet
(.
[xv
t vapol Soiia01r)XXY7)aav
bar'vTtxetaOat
t StTOVOv
7rp
evapfxovc*)
' H*pv
8' aTcov
t xpwM-aTtxv
Y<*p
fxcov
[xvLov
7rpxetv
SiocTvp,
aitSiocTovtxou.
^)jjLtTvtov,
(JanS.263,3-10)
,v(xvov
The mathematics for exact placement of the notes of the enharmonic
and chromaticgenera is not found in the Enchiridion.The enharmonic
dieses are described as halves of semitoneswhich add up to a complete
semitone:
7)(jlltovou
xal rcXiv
Seot,auvajxTspat
8e<n,
*)(xtT0v<j)
XXyj
^{xtoru,
taon.(JanS.262,22-23)
Such language obviously does not referto a geometricdivision of the
semitone.48The intervals of the chromatic genus are defined only to
47Nicomachus'derivation
of generafromthe diatonicis easilycomprehended
in a figuresuchas thefollowing:
Diatonic:
semitone
tone
:
Chromatic
semitone
semitone
tone
:
diesis diesis
Enharmonic
48"Half" (especially
meanhalf
theprefixes and semi-) doesnotnecessarily
musicaiii, 1-2 (271-273).A
musicaltheory;see De institutione
in Pythagorean
are equal, and
"geometric"divisionwouldbe one in whichboth proportions
couldbe expressedthus: taketheintervalofa semitoneand divideit intotwo
equal, but when
parts,however,are not necessarily
equal parts.Nicomachus*
20
23:17:05 PM
the extent that they all add up to equal two tones and a semitone:
xal yjjxitovou
ovtaccexov<pav7]TaL
el xai [ir]XSuotvcov
vuxpeaxiv,XX*
Suaitvol
T 8iaarj[xaTa
xal rj(jLLxovq>.
(JanS.262, 18-21)
Beforecomparingthese principlesof division fromNicomachus with
the division found in Boethius, the criticismsof Gushee and Pizzani
should be examined. Gushee suggests that the omission of the letter
"g" in the monochordalphabet of IV, 11 mightpoint to a Latin source
for the division.49Gushee furtherimplies that Boethius derived his
discussionof the monochordfromPtolemy,but that he did not "clearly
or unequivocally adhere" to the sophisticated instrumentspresented
in Ptolemy.50I fail to see how the presenceor absence of the letter"g"
can imply a Latin source unless one can find a source which similarly
omits "g". Furthermore,I can find no trace of Ptolemy in Boethius'
division of the monochord; such a division of the entire system is a
concern wholly foreignto Ptolemy. One must not let the similarity
between the last chapter of Book IV and a passage fromPtolemy lead
one to assume that the heart of Book IV is influencedby, much less
borrowedfrom,Ptolemy.51
Pizzani considered the possibilitythat Boethius' division might be
based on that whichNicomachus promisedforhis Eisagoge. But Pizzani
sees three difficultieswith this thesis:
1. The mathematical inconsistencyof Boethius' chromatic genus;
2. The absence of any trace of the divisions of Thrasyllus or Eratosthenes;
The
3.
neglect, in Boethius' division, of the number 27 mentioned
at EnchiridionJanS. 261, 17.
Such problemslead Pizzani to conclude that even a hypotheticalderiaddedtogether,
A semitoneaccordingto thePythatheymakeup a semitone.
whichwillnotadmita geometric
divisionusing
goreansis 256:243,a proportion
wholenumbers.
49Gushee,op. cit. (n. 26), p. 380, n. 47. Thereis a certainsymmetry
in the
omitted" g's", fortheybothoccuroversecond-to-highest
notesin enharmonic
the firstin the hypatontetrachord
and the secondin the hypertetrachords,
boleontetrachord
(334 addenda).Everyothercomparablenotein enharmonic
has a letterassignedthereto,
tetrachords
whereasthesetwonotes,one "g" and
theother"gg" aremissing.Suchan inconsistency
wouldmostprobablyarisein
withtextualtransmission.
conjunction
50Gushee,op. cit.,p. 379.
51Concerning
the last chapterof Book IV, see sectionF. Intervallictests,
PP-37-38.
21
23:17:05 PM
vation of Boethius' division fromNicomachus is impossible.52Pizzani

is perceptive in centeringon these three questions, yet he does not
carry them far enough; for ultimately these traits may be used as
evidence to argue for Nicomachus as source for the division.
Nicomachus had set forthone overridingPythagorean principlefor
the divisionofthe canon in the Enchiridion: the primacy of the diatonic
genus. One does not have to read very farinto theoreticalliteratureto
sense the polemic tone in Nicomachus' statementof Pythagoreanprinciples. A monochord division begun from the Pythagorean diatonic
genus is truly exceptional in ancient musical theory. As early as
Aristoxenus, the Harmonists- Eratocles in particular- are criticized
forbeing obsessed with the smallest of intervals and the enharmonic
genus to the exclusion of the chromaticand diatonic genera.53Although
Aristoxenusand his followersdo presenttwo shades of a diatonicgenus,
theiremphasis remainson variationsof the enharmonicand chromatic,
and their mathematical principles are unacceptable to any Pythagorean.54Yet the obsession with the pyknon, the lowest segment of
the enharmonic and chromatic tetrachords,was not the franchiseof
Aristoxenianmusical thought; fortheoristsusing Pythagorean mathematical principles- i.e. proportions- equally emphasized the enharmonic and chromaticgenerato the exclusionof the diatonic genus or to
the compromiseof its Pythagorean integrity.Archytas and Didymus
had derived divisions of the threegenera which,although expressed in
proportions,wholly forsookthe Pythagorean tetrachordconsistingof
9:8/9:8/2 56 : 243.55 Ptolemy does include the Pythagorean scale as
one of his five shades of diatonic genus, but the "diatonic diatonic"
shade is included almost like an afterthoughtand plays no centralrole
in his derivation of genera and shades. Thus we come to consider the
52Pizzani,op. cit.,pp. 115-121
a farciescludere
concorre
; p. 121: "Tutopertanto
dellasectioboezianada Nicomaco".
ancheuna ipoteticaderivazione
53Aristoxenus,
Elementai, 2-3,6; ii, 35-36.
54For Aristoxenus'
shadesofthegenera,see e.g.Elementa
ii,48-52.For Pythasee De inst. mus.iii, 3 (273-274).
goreancriticism,
55The divisionofArchytasand Didymusare knownthroughPtolemyHarmonica:
Archytas(Ptolemyi, 13):
28: 27
Diatonic:
9:8
8:7
Chromatic:
32:27
28:27
243:224
Enharmonic: 28:27
36:35
5:4
Didymus(Ptolemyii, 14)
Diatonic:
9:8
10:9
16:15
Chromatic:
6:5
25:24
16:15
Enharmonic: 32:31
31:30
5:4
22
23:17:05 PM
two divisions mentionedby Nicomachus in the Enchiridion,those of

Eratosthenes and Thrasyllus. The diatonic genus of Eratosthenes did
consistof the Pythagoreandiatonic,but his chromaticand enharmonic
were by no means related to or derived fromthe diatonic, but rather
were related to each other and the enharmonicwas derived fromthe
chromatic.5Our knowledge of Thrasyllus' division, known only
through Theon of Smyrna,57is extremelysketchy. Nevertheless we
can affirmthat Thrasyllus based his division on the diatonic genus,
and that the chromaticand enharmonicgenera were derived fromthe
diatonic. Thrasyllus' division,in fact,resemblesthat foundin Boethius
more than any other extant division,and Thrasyllus' derivationof the
chromaticlichanos might even seem more consistent than that found
in Boethius.68One can only speculate concerningNicomachus' reasons
forrejectingThrasyllus' division; based on the versiongiven by Theon,
Thrasyllus' mathematicswould present difficultiesto one workingout
details, and Thrasyllus'divisionofthe chromaticwas probably unsatisfactoryto Nicomachus.59
Now the broad context of Nicomachus' citations of Eratosthenes
and Thrasyllus can be broughtinto focus: These two theoristsare at
least Pythagoreans,they followedthe master ( ouXrjâtoSs
at least to the extent of using the Pythagorean diatonic
SiSatrxXou)
but
Nicomachus
genus,
they somehow misunderstood (7tap)xou<rcv).
does not even mentiontheoristswho did not adhere to the Pythagorean
diatonic genus, and thus his criticismof Eratosthenes and Thrasyllus
is a formof praise compared to the silent condemnation of those who
would forsakethe 9 : 8 tone and the 256: 243 semitone. In preciselythis
context Nicomachus comes to cite Plato's account of the section of the
world soul in the Timaeus. The principal import of this citation is
again to state the primacy of the diatonic genus, for the tuning presented in the Timaeus is the authoritative Pythagorean tuning with
2:1 octave, 3:2 fifth,4:3 fourth,9:8 tone, and 256:243 limma or
58Eratosthenes'
divisionis likewiseknownthrough
Ptolemy(ii, 14):
Diatonic:
9:8
9:8
256:243
Chromatic:
20:19
19:18
6:5
Enharmonic: 40:39
39:38
19:15
67TheonofSmyrnaii, 35-36(ed. Dupuispp. 143-153).
58Thrasylluscomputedthe chromaticsynemmenon
9:8 (i.e. the
by figuring
diatonictone) fromthe mese (Theonii, 36, 150, 10-13).The enharmonic
is
computedby dividingthediatonicsemitonein half(Theonii, 36, 152,1-2),but
no definite
is givenforthiscomputation.
proportion
59Theon(152)gives384as
thenumberofthenetehyperboleon
and 10,368that
oftheproslambanomenos.
Thesetwonotesshouldbe related4:1.
23
23:17:05 PM
semitone. The division of the Timaeus was followed by Plato up to

the twenty-seventhmultiple (eoa 7rocxaieixo0i7tX<xcrou)
; that is to
say, Plato carried the progressionto the cube of 2 (the firsteven
number), i.e. to 8, and to the cube of 3 (the firstodd number),i.e. to
80
27. The number 27 itselfhas nothingto do with the division except
to establish, along with the multiplications of 2, the mathematical
context within which the division takes place. Thus the answer to
Pizzani's inquiry concerningthe meaning of Nicomachus' expression
be answered precisely:61it establishes
Ic eTtTocxaieixocn-jrXaoriou
the Pythagorean diatonic mathematics as the basis of Nicomachus'
monochord division.
The principles set forthby Nicomachus must now be compared to
the monochord division found in Boethius' fourthbook. The Pythagorean diatonic genus is clearly the basis of the division. Chapter 5
sets forththe linear principlesof the diatonic genus noting each pitch
with a Greek notational symbol.2But such linear demonstrationis
not sufficient,and each pitch of the systematelionis translatedinto a
discrete number in chapters 6 through 12. The diatonic division adheres to the Pythagorean diatonic of 9 : 8 tones and 256 : 243 semitones.
The chromatic division is constructedin such a way that the second
note (parhypate or trite) of each tetrachordin the diatonic and chromatic genera are the same, just as set forthin Nicomachus' Enchiridion.
Moreover,the same note becomes the thirdnote (lichanos or paranete)
of the enharmonicgenus, again followingthe principlesspecified by
Nicomachus. Boethius' division rigorouslyfollowseach and everyprinciple of monochorddivision set forthin the Enchiridion.
Yet as pointed out by Pizzani and Gushee,63a certainmathematical
inconsistencymay appear in the computation of the thirdnote (lichanos or paranete) of the chromaticgenus. Brieflystated, Boethius come0 Timaeus35 b-c.
41Pizzani,op. cit.,p. 120: "Ma comeinterpretare
alloral'espressione
di Nicomaco 1<tourtTotxateixooiitXaotou
? Una precisarisposta,allo statodellenostre
conoscenze,sembraimpossible".
SJThe divisionof chapter5 is incomplete,
foronlyelevennotes(of 16) are
andsystema
synemcomputed.Onlythefixednotesofthesystema
diezeugmenon
menonand thehighertonesofeachtetrachord
axedetermined;
thelowertone- ofeachtetrachord
and thustheremaining
semitone
is missing.
Sincetheselast
fivestepswouldlogicallyfollowthosewhichare presentin the chapter,one
in thetransmission
ofthistext,either
arguethatsomewhere
mightconvincingly
the finalsteps of this divisionwereomittedthrough
pre-or post-Boethian,
scribalerror.
,s Pizzani,op. cit.,pp. 117-118;Gushee,op. cit.,p. 380.
24
23:17:05 PM
putes the highestinterval of the chromaticgenus by taking the arithmetic mean of the highest diatonic 9 : 8 tone, i.e. 18 : 17, and adds that
proportionto the tone. But this division is inconsistentwith the remainderof the treatise only in so far as it calls the arithmeticmean of
a tone a "semitone", while the proportion256 : 243 has repeatedlybeen
demonstratedto hold the proportionof the semitone. Pizzani similarly
views the division of 256:243 by an arithmeticmean in the computation of the enharmonicgenus to be a mathematical erroranalogous to
that ofthe computationof the chromaticlichanos.64To judge Boethius'
chromaticand enharmonicdivisions inconsistentis to judge them by
the rules Boethius has set forthforthe diatonicgenus. The onlyinterval
the three genera share in common is the 256:243 semitone, the 9:8
tone as a discreteintervalhas no place in the chromaticor enharmonic
genus. The properquestion to ask at this point is, given the limitations
of derivingthe chromaticand enharmonicgenera fromthe diatonic as
specifiedby Nicomachus, what means should one use to compute the
one note needed in each of these genera? The answer lies in the question: means. Plato's Timaeus describes the process of completingthe
monochord division as one of using harmonic and arithmeticmeans
to divide the products of multiplying2 and .65 Similarly Boethius,
translatingfromNicomachus, describesthe process of using arithmetic,
geometric,or harmonic means to divide a monochordin Book II, 50,
of De institutionearithmetica
:
Nos autempraestaredebemusquatenus,quemadmodum
dato calamoextremis
foraminibus
manentibus
musicismosest,ut mediumforamen
permutantesatque alios aperientesalios digitisoccludentesdiversosemittant
sonos,vel cum duabus altrinsecusprotensischordismediinervisonum
musicusvel adstringendo
tenuatvel remittendo
gravatiita quoque datis
duobusnumerisnuncquidemarithmeticam
nunc
nuncverogeometricam
autemarmonicam
medietatem
utrectumpropriumque
inserere,
experiamur
medietatis
nomensit,quod manentibus
hueatqueillucferri
extremitatibus
videatur.(160,8-18,cf.Nicomachus,
Eis. arith.ii, 27)
permutarique
Given the Pythagorean traditionof the three types of means and the
diatonic arithmeticfromwhich to derive the chromaticand enharmonic genera, a consistentPythagorean would be required to employ the
theoryofmeans to divide these genera. Moreover,given the limitations
of diatonic division, the arithmeticmean is the only one which could
be used to compute the chromaticlichanos (or paranete) or the enhar64Pizzani,op. cit.,p. 118.
5 Timaeus36 a-b.See on thewhole
Plato's
passage(35 b - 36 b) F. M. Cornford,
London21948,pp. 66-72.
Cosmology,
25
23:17:05 PM
monic parhypate (or trite).But mathematicalnecessityshould not lead

us to consider the "arithmetic" computationof the chromaticlichanoi
and the enharmonic tritai in this division mere acts of expediency.
Nicomachus, the arithmetician
, considered arithmeticproportionality
prior by nature to other types of proportionality,and his discussion
of unique characteristicsof this class of proportionsapproaches the
mood of a paean. 68 Considered in the context of Pythagorean arithmetic, the computations of the enharmonicand chromatic genera in
Boethius' division, rather than being inconsistencies,furtherunite
Boethius' text with the traditionof Nicomachus. The absence of theory
of means in Thrasyllus' computation of the chromatic genus could
well be Nicomachus1groundforconsideringit wayward to the "will of
the master".
The degree of consistencybetween Boethius and Nicomachus presents clear circumstantialevidence to suggest that Boethius' division
of the monochordwas taken fromNicomachus. The final step in proving this relationshipmay be taken in examiningchapter 13.
D. Fixed and Movable Notes: The theoryof notes bounding tetrachords as fixed and notes withinthe outside notes as movable through
the genera is a common denominator of all classical Greek musical
theory. But Nicomachus presents this theory in a way that is found
in no other author of antiquity. He definessome notes as fixed, some
notes as movable, and some notes as both movable and immovable.
The thirdcategoryis the elementin Nicomachus' theorythat is unique
to him and he definesnotes of this third category as those which do
not move between the diatonic and chromatic but move in the enharmonic:
ou yp {171ol (xvovtoGTETpaxpSou
xpoiat&te
XyovTat,
vye vapjzovcp.
vS
t&vyevcvo 8 [jtaoi
TouaivvoSevi
xivofievoi
7cp<;
xat ou xtvofxevo
SeTepo
(xvypt Skxtovov
xivojxev
<(1
8 t vap(jLviov
o fXTa7U7rcet,
(JanS.263, 11-17)
(.
7cp
This descriptionclearly followshis earlierdescriptionof the chromatic
genus in which the second note of the chromatic tetrachordwas the
same as the second note in the diatonic tetrachord.87The only other
source of Greek musical theory containingthe theory of notes which
sometimes move and sometimes remain the same is Boethius' De
Nicomachus,Eisagogearithmetica
ii, 22-23; cf. Boethius,Be inst.arith.ii,
42-4(139-144).
87See above n. 47.
26
23:17:05 PM
institutionemusica.**Even Pizzan must admit that this chapter of

Book IV is taken directlyfromNicomachus." But the threefolddivision of notes is found in Boethius' Book 1, 27 (2x9, 25-28), and at that
place the discussion of Book IV, 13 is promised. Thus the threefold
division of notes serves to unifyBook IV with the firstbook of Boethius' treatise.
The placement of the passage concerningfixed and movable notes
immediatelyfollowingthe monochorddivision is crucial,forthe mobility or immobilityof notes is determinedin a division of the monochord.
By viewing the monochorddivision fromthe perspectiveof the threefold classificationof notes we come to see that the division found in
Boethius is as unique as the classificationgiven by Nicomachus. No
other division of the monochord extant from antiquity is consistent
with Nicomachus' classificationof notes and mathematical principles.
The division of Didymus is constructedin such a way that the parhypate or trite is the same proportionfromthe lowest note in both the
chromatic and diatonic genera, but that proportion is 16:15 rather
than 256:243. Similarly in the division of Archytas the second notes
are the same in both diatonic and chromaticgenera,but the proportion
is 28 : 27 ; and the proportionis the same forall three genera, not just
the diatonic and chromatic.70The divisionof Thrasyllus,althoughonly
knownin part, comes closest to that ofBoethius, but it presentsmathematical difficultiesand Nicomachus expresslyrejected it. Therefore,I
can only conclude that the monochord division of Boethius' fourth
book- chapters5-12- and the discussionof fixedand movable noteschapter 13- are cut fromthe same piece of cloth, and that the fabric
of this theoryis drawn fromthe lost musical treatise of Nicomachus
of Gerasa.
E. Modal theory: No section of Boethius' De institutionemusica has
attracted more attention from musicologists than the four chapters
dealing with the theoryof Greek modes. Any reader approaching these
chaptersforthe firsttime feelsconfusedif not bewilderedby the complexity and seeming contradictionsfound therein. But the confusion
of these chapters is nothingcompared to the contradictionsfound in
secondary literature treating them. It is difficultto examine these
88BoethiustranslatesNicomachus'
xa o xivo|xevoi
as interimxivofxsvoi
mobiles
in i, 27 (219,25-26) and neein totum
inmobiles
in totum
mobilesque
mobiles
in iv, 13 (335,10).
" Pizzani,op. cit.,pp. 121-122.
70See above. 55.
27
23:17:05 PM
chapters withoutexploringeach theoreticaldetail, but here I will only

examine theoreticaldetail insofaras it relates directlyto the question
of Boethius' sources. The much needed freshstudy of the theorycontained in these chaptersmust await a futureessay.
The most carefulexamination of Boethius' modal theoryis a fortydes Boethius.11Kunz
year-old study by Lucas Kunz, Die Tonartenlehre
examines each aspect of Boethius' theory in comparison with other
ancient sources, points out the aspects they share in common and the
aspects unique to the treatmentfoundin Boethius, and concludes that,
althoughthereis evidence of Ptolemaic thought,Boethius did not draw
directlyon Ptolemy forthese chapters. He argues ratherthat Boethius
based these chapters on two non-Ptolemaic sources. Kunz views
with transBoethius as a somewhatconfusedscholar,having difficulty
lation from Greek to Latin, juggling two sources with conflicting
aspects, composingneverthelessan importantsource forthe historyof
ancient modal theory.
Without refutingthe arguments of Kunz, Pizzani concludes that
Ptolemy is the "essential source, if not to say the unique source" of
Book IV, 14-18.72 He bases this assertion on loose parallels between
Ptolemy and Boethius which we will examine below. Pizzani suggests
that most of Boethius' errorsarise fromhis attempt to simplifythe
theory he found in Ptolemy and coordinate it with the notational
charts which, like chapters 3 and 4 of Book IV, he borrowed from
Gaudentius. Pizzani considers the last three chapters treating the
modes as a "deformed and vilified exposition" of theory of tropes.73
He accuses Boethius of confusingoctave species with trope,an accusation carefullydemolishedin the study of Kunz.74
The one opinion that Kunz and Pizzani- and most other musicologists who have commented on Boethius' modal chapters share in
common is that Boethius was inept at reading Greek and rather
muddleheaded. Yet this opinion is never stated as a hypothesis,but
rather advanced insidiously as a means of explaining elements in
Boethius' text which do not fit their argumentsof the moment. The
71 In: Kirchenmusikalisches
31 (1936),5-24.
Jahrbuch,
72Pizzani,op. cit.,p. 136: "Da tuttoci risultacon estremaevidenzache,pur
i travisamenti,
le suture,Tolemeo la fonteessenziale,
le omissioni,
attraverso
musica".
pernondireunica,dituttaquestasezione(IV, 14-18)delDe institutions
73Pizzani,op. cit.,p. 128: "I rimanenti
quattrocapitolidel quartolibro(15-18)
la tantodenigratae vilipesaesposizionedella dottrinadei tropi
contengono
scale di trasposizione".
74 Kunz,op. cit.,pp. 14-18.
28
23:17:05 PM
corpus of Boethius' writingsrefutesthis opinion; formost of his works

are built upon Greek sources, and he repeatedly proves himselfable
both to understand and to translate the most difficultmathematical,
logical, and philosophical concepts. This is not to say that Boethius
was beyond makingsome minorslip of understandingor translating;75
it is rather to argue that he had a basic comprehensionof the Greek
source or sources fromwhich he was workingand that his Latin text
faithfullytransmits the basic import and structure of the Greek.
Having restated my estimate of Boethius' ability, I will i) examine
the relationshipof Boethius' modal chapters with Ptolemy, 2) show
aspects of these chapters which are unique in relation to all other
extant sources, and 3) make several limited suggestions concerning
possible sources.
1) Boethius and Ptolemy: The relation of Boethius' text to that of
Ptolemy becomes apparent in the firstdefinitionpresentedin chapter
14, that of species:
orci
rcoi0ait&vxa0' Species autem est quaedam positio
eI8o[lvTovov
IxacTOv
lv toioxclol propriamhabens formamsecundum
yvoISiavTcov
SpoiXycov.
(49,9-10)
unumquodque
genusin uniuscuiusque
proportionisconsonantiamfacientis
terminis
constituta.(337,22-25)
An explication of the species followsthis definition,and Boethius uses
an alphabet to representnotes exactly the same as Ptolemy:
Boethius:7*abcdef
lmnxo
ghk
Ptolemy: ABTAEZH0KAMNSO
ii.3.50)
(Harmonica
75See listofminordifficulties
and variantsbetweenBoethiusand Nicomachus
preparedby Robbinsin D'Ooge, op. cit.(n. 5), pp. 134-136,whichconcludesas
follows:"It is easy to see fromthe unimportant
characterofthesedeviations
and fromtheircomparative
fewness
in suchan extendedtextthatBoethiusdid
notdo violenceto hismodel".
7e Friedlein's
themajority
thisalphaofmanuscripts,
edition,
following
presents
bet as ending... h i 1 m n (338, 4-9). Numerousmanuscripts,
however,
employthe Ptolemaicalphabet,or the Ptolemaicalphabetchangedinto the
Latin alphabet:amongthe latterare: Brussels,Bibl. Royale,Ms 5444-6and
Ms 358; Florence,Bibl. MediceaLauren10114-6;Einsiedeln,Stiftsbibliothek,
Ms52; Munich,Bayerische
Staatsbibl.
ziana,Ashb.1051; Ivrea,Bibl.Capitolare,
Clm367 and Clm 18480; Paris,Bibl. Nat., Ms Lat. 7202; Prague,University
Ms 1717;Vienna,Oest.Nat. Bibl.,Ms51.5and55.2.SincethePtolemaic
Library,
alphabetis the modelforthispassage,and sincemanuscripts
containingthe
Ptolemaicalphabetgenerally
theoldesttextualtradition
represent
(10thc), the
Ptolemaicalphabetin all likelihoodrepresentsthe one originallyused by
Boethius.
29
23:17:05 PM
But Ptolemy considers alpha to be the highest note while Boethius

assigns 'a' to the hypate hypaton, the lowest note considered in this
passage. Boethius does not, however, merely transliterate Ptolemy
when he explicates the species, in fact his species are considerably
differentfromPtolemy's; thereforealthoughthis passage exhibits the
Ptolemaic element of the alphabet, it clearly is not based on Ptolemy
in workingout details.77
Boethius' definitionof system (constitutio
) in chapter 15 appears to
be related to that of Ptolemy Harmonica ii. 4:
xaXeTai
cruy- Constitutio
est plenumvelutimodu. . . cr<JT7)(xa
(xvCLizk&,
x auficpcim&v,
xaOdbtsplationis corpus ex consonantiarum
(ysOo
xefjievov
r avpceiievov
consistensquale est vel
jjl- coniunctione
1iyeQo
au^tpcvia
xal onv
au[jL<pc>va
oujjl- diapason,vel diapasonet diatessaron,
jxeXetcv,
vel bisdiapason.(341, 22-25)
t acraqfjux.
(50,12-15)
cpam&v
But little more can be said concerningthese two definitionsexcept to
say that they are related. The basic idea of structure is the same:
system is made fromconjunction of consonances. But Ptolemy's abstract fjiyeGobecomes the more concreteplenum modulationiscorpus
in Boethius, and Ptolemy gives no specific examples of systems as
does the definitionof Boethius. Ptolemy's text, as a matter of fact,
proceeds fromthis point to demonstratethat only the bisdiapason is a
valid system.78
Of these three parallels, only one, the definitionof species, can be
said to show exact correspondenceto Ptolemy. But thereare numerous
elements in the modal theoryfound in Boethius that run contraryto
the thoughtof Ptolemy. Boethius uses the word modusin his presentation of the Greek "keys", a word Boethius used to translate ;,
whereas Ptolemy seems to avoid the term;[Harmonica ii, 7-11).
Moreover, Boethius gives a clear definitionof modus equating the
concept with moving a system up or down in pitch:
in totisvocumordinibus
vel gravitatevel
Suntautemtropiconstitutiones
acuminediffrentes.
(341,21-22)
Ptolemy gives no such definition,and, in fact, Boethius' definition
runs contraryto Ptolemy's concept ofnon-modulatingsystem(miarla
77Pizzani,op. cit.,p. 126,agreesthat Boethiusis consciousof thisdeparture
coscientedella modificaapportataal
fromPtolemy:" Boezio perfettamente
suo modello. . .
78Pizzani,op. cit.,p. 131,makesPtolemylookmorelike Boethiusby leaving
out phrasesandonewholesentencefromPtolemy'sdefinition
ii, 4,
(Harmonica
violenceto Ptolemy'soriginal
50, 12-20); but in so doinghe does considerable
text.
23:17:05 PM
Boethius includes the diapason system and synemmenon

fXETpoXov).79
his
in
presentation of modes, while Ptolemy rejects these
system
because
they will not contain all species of consonances (Harsystems
monica ii. 4). Boethius determinesthe number of species a consonance
may have in the followingway: the species of a given consonance will
number one less than the number of notes in that consonance (338,
16-17: semperque una minusspecies erit,quam fuerintvoces) ; Ptolemy
determinesthe numberof species as follows: a consonance will have as
many species as there are intervals in that consonance (Harmonica ii,
3, 49). Boethius describes the spacing of the modes using emmelic
intervals (342, 346-347), while Ptolemy insists that the spacing of the
modes must be determinedwith only consonances (Harmonica ii, 10,
63-64).
But these differencesbetween Boethius and Ptolemy merelyenumerate specific points of conflictbetween the two authors, they do not
take account of concepts essential to Ptolemy's thought which are
omitted from Boethius' text. The following Ptolemaic concepts are
totally absent from Boethius' text: the thetic and dynamic placing
of notes (ii, 4), the rationale for the character or "ethos" of the tonoi
(ii, 7), the argumentsthat the numberof tonoi should not be extended
by using the semitoneand that the eighth tonos is qualitativelyidentical to the first(ii, 8), and the notion of a middle octave- comfortable
for the voice- around which the tonoi function (ii, 6). Moreover,
Ptolemy's thought is qualitatively differentfrom that found in
Boethius: Ptolemy argues in the manner of a geometrician,always
abstractly, repeatedly using alphabetical symbols to prove a point
before relating the principle in question to concrete musical terms,
whereas Boethius, repeatedly using specific names of notes and
notational symbols, argues from a more concrete and practical
standpoint.
Given 1) the looseness of correspondences between Boethius and
Ptolemy, 2) the theory contained in Boethius that is contrary to
Ptolemy, and 3) the absence in Boethius of concepts essential to the
Ptolemaic concept of the tonoi, it seems improbable that Ptolemy's
Harmonica was even a direct source, much less the essential source,
79See especiallyHarmonicaii, 5, wherethenotesas names
( Gaiv)become
a circle,as it were,offunctions
the
; I. Dringclearlyillustrates
(xar Svaîv)
and commentary
on Harmonica
undPorprocessin his translation
, Ptolemaios
phyriosberdie Musik, Gteborg(GteborgsHgskolasrsskrift
XL) 1934,
to Boethius'
p. 228. Such a conceptis uniqueto Ptolemyand whollyforeign
transposed
systems.
31
23:17:05 PM
of Boethius' discussion of the modes. An argument that Ptolemy is

the direct source of this passage must be based on Boethius' incompetence, a premisewhich remains to be proven. Boethius was obviously
workingfroma source other than Ptolemy, but a source which either
was acquainted with Ptolemy's theoryor with the source upon which
Ptolemy based his thought.
A final perplexingproblem concerningthe relation to Ptolemy and
the addition of the eighth mode must be approached. I must examine
the context and passage of this problemin detail since it leads Kunz
to argue that Boethius used two sources for the modal theory80 and
Pizzani to attribute a flagrant error to Boethius.81 In chapter 14
Boethius had shown that there were seven species of diapason using
the alphabet "a . . . m n x o", consistingof fourteenletters.In chapter
15 Boethius explained that fromthese seven species were derivedseven
modes:
Ex diapasonigiturconsonantiaespecibusexistunt,qui appellanturmodi
. . . (341,19-20)
But in the charts found at the end of chapters 15 and 16, one finds
eight modes. Thus chapter 17 begins with a qualification concerning
the eighthmode:
Septemquidemesse praediximusmodos,sed nihilvideaturincongruum,
quod octavussuperadnexusest. Huius enimadiectionisrationempaulo
posteriuseloquemur.(343,17-20)
Boethius proceeds to explain the rationale of the modal chart,then by
using the mesai he traces the intervalsbetween the modes. He explains
carefullyto the reader that the proslambanomenoi,the lowest pitches,
are to the reader's left and the netai, the highest pitches, are to the
reader's right.Then he takes the mese of each of the firstseven modes
and shows its intervallic relationship to its neighbors and others.
Finally, he comes to the justification for the eighth mode. Given
Boethius' previous rationale formodes, two aspects of the mode must
be explained: its species of diapason, and its position relative to the
other modes. In order to explain the diapason of the eighth mode,
Boethius introducesa new alphabet, describedsimplyas a bisdiapason:
Sit bisdiapason consonantia haec: (347, 19). The new alphabet proceeds
A ... N P, and contains fifteenletters.Boethius then takes the diapason A : H and says it is what "we called the firstspecies of diapason".
80Kunz,op. cit.,pp. 19-20.
81Pizzani,op. cit.,p. 134.
32
23:17:05 PM
Primamigiturdiximusessespeciemdiapasonearn,quae est AH.

At this point Kunz argues that Boethius must be using a different
source fromchapter 14, for in chapter 14 "o-g" was called the first
82
species (339, 5-6). Moreover,Kunz argues that Boethius was not even
aware of the discrepancy created by using two alphabets.83But two
characteristicsof these alphabets must be observed: 1) Each alphabet
fits only in the context of its use. The first alphabet was used to
explicate the seven species of diapason, and it representedthe notes
fromthe hypate hypaton to the nete hyperboleon,i.e. only 14 notes
of the 15 in the systematelion.The firstalphabet contains only seven
diapason consonances, the seven species, for one needs 15 notes to
arriveat the reliqua diapason.84Moreover,the firstalphabet cannot be
altered to add the necessary note, for the note to be added is the
proslambanomenos,a note which precedes the hypate hypaton, and
the hypate hypaton is already A. Therefore Boethius needs a new
alphabet. 2) The two alphabets are completely different,for the first
with 14 letters ends in ... m n x 0, while the second with 15 letters
ends in M N P (and uses I). Finally, Boethius does not say: AH est
ea, quae est prima species diapason; but rather: diximus primam
speciam diapason esse earn,quae est AH. The differencebetween these
expressions demonstrates that Boethius was aware he was working
witha new alphabet as well as a new and largersystem.Since Boethius
had traced all species fromhigher to lower pitch, one must assume
that A is the higher pitch in the new context. Therefore,A is nete
hyperboleonin the alphabet concerningthe eighth mode, and indeed
AG of this passage is identical with OG of chapter 14.
An important but overlooked bit of palographie evidence further
proves this point. Two manuscripts85 contain the second alphabet
written from right to left, as follows:
PONMLKIHGFEDCBA
Given the facts that this alphabet is explaining the chart of the modes,
that Boethius has just describedthe chart as representinglow notes on
82Kunz,op. cit., 19-20.
83Kunz, op. cit.,pp.
dass seine
p. 19: "Er ist sich nichtdarberklar geworden,
neueQuelleeineganz andereZhlweisederOktavspezieskennt".
84Boethiusnevermakesthemistakeof
callingthe eighthdiapasonan "eighth
writes:"Reliquiturigiturextra-HP-,quae ut totusordo
species".He carefully
adiectaest". (348,1-2)ThusI referto theeighthdiapason,following
inpleretur,
Boethius,as reliquadiapason'.consonantia'.
86Munich,Bayerischetaatsbibl.Clm
6361and Clm 18478.
33
23:17:05 PM
the left and high notes on the right (345, 20-22), and that Boethius
consistentlytraces species fromhigh to low pitch, the alphabet from
right to left perfectlyfulfillsthe need of the argument that a bisdiapason system (A-P) will contain one diapason beyond those of the
seven species, and that this reliqua diapason is that found between the
mese and the proslambanomenos (HP). 86
Thus it is not necessary to postulate a new source to explain the
logic of the eighth mode in Boethius' text, rather it is obvious that
Boethius- or, more properly,his source- was required to postulate a
new alphabet in order to justify the reliqua diapason. Although it is
awkward to use two differentseries of letters, each is unique to its
context and each is consistentwith the other.
But the alphabet merelyexplains the reliqua diapason, not its position in relation to the other modes. The passage explicating the possibility of eight modes concludes with the followingsentences:
adiectaest.
Relinquiturigiturextra HP, quae ut totusordoimpleretur,
Atquehisest octavusmodus,quemPtolomaeussuperadnexuit.(348,2-3).
If we recall the two necessaryelementsin Boethius' rationale formodes
- a species of octave and a position in relation to other modes- the
concluding sentences of chapter 17 make perfectsense. The "species"
of the eighthmode is HP, the reliqua diapason added so that the whole
series mightbe filledout; Ptolemy joined it (positioned it) at the top.
The referenceto Ptolemy clearly refersto the followingpassage :
ovtcc
7vlt tvoctv
. . . tvo)tvSi raxacov
7rpo<rr)Ypeuaav
afxevov
...
rcaufjLe7)x0TO<;,
tvjjuoXSiov
a>rcp
6lX7)((ivov
u7uep(iioX8tov
(63, 5-7)
In this passage {Harmonica ii, 10, 62-64) Ptolemy is describinga traditional school of Greek theory which espoused eight modes. Ptolemy
is criticalof two aspects of this school, the way they use emmelicinter86Despitethefactthattheoverwhelming
containthe
majorityofmanuscripts
fromleftto right,I wouldarguethattherightto leftreading
alphabetwritten
thepassagewiththeright-left
is the original.It is mucheasierto understand
the alphabetto read noralphabet,and it is easy to imaginescribesreversing
is
intoright-left
mally.On theotherhand,a reversein theorderfromleft-right
inordinate
of
scribes
would
such
a
to
difficult
require
explain;
change
very
eruditionand understanding.
Finally,thisalphabetis a pointof considerable
omitthe
see Kunz,op. cit.,p. 21. Some manuscripts
variationin manuscripts;
seriesof letters,leavingan emptyspace followedby the text explainingthe
missingalphabet;amongtheseare: Ivrea, Bibi. Capitolare,Ms 52; Munich,
Staatsbibl.,Clm367; St. Gall,VadianischeBibl.,Ms 296; and StuttBayerische
Landesbibl.,Ms HB XI 33.
gart,Wrttembergische
34
23:17:05 PM
vals to space the modes and their use of an eighth mode identical in
species to the first.Thus although Ptolemy describes the eighthmode
"
as joined at the top" (and Ptolemy's is the only extant passage concerning the position of the eighth mode), in the end he rejects the
validity of the hypermixolydianmode. Both Pizzani and Kunz mistranslate Boethius' text in saying Boethius attributed to Ptolemy the
"addition" of an eighthmode.87There are clearly two complementary
verbs in the last two sentences of chapter 17: adiecta est and super
adnexuit. The mode is added because of the extra diapason in the
bisdiapason, and Ptolemy, ratherthan adding it, joined it at the top.
Boethius does not attributeto Ptolemy the addition of an eighthmode.
If there is a flaw in this passage, it is that BoetKius- or his sourcedoes not state that "according to Ptolemy (or, Ptolemy said) the mode
was added at the top, but that in the end Ptolemy rejected it".88
2) Unique aspects: Several elementsare foundin Boethius' chapters
concerningthe modes that are foundin no other extant treatise,either
Greek or Latin. Before making suggestions concerningthe source of
these chapters, these unique aspects must be examined.
Boethius' explicationof the species of consonances is unique in three
ways: 1) he traces consonances fromhigh pitch to low pitch whereas
all other theorists trace from low to high.892) Boethius develops a
systemof numberingthe species whichis differentfromothertheorists,
all of whom share the same numberingsystem.903) While Boethius
uses the newly developed numberingsystemto trace out species, when
explaining which species are found between fixed notes he uses the
traditionalsystem; moreoverhe is expresslyaware that he is using two
systemsof numbering.91
87Kunz,op. cit.,p. 13: "Eigentmlich
ist es, dass Boethiusden achtenModus
... als den Modusbezeichnet,
der von Ptolemushinzugefgt
sei". Pizzani,
latinoattribuisce
op. cit.,p. 134: "Per giuntaalla finedel cap. 17 lo scrittore
proprioa Tolemeotale aggiunta:hicestoctavusmodus,quemPtolomaeus
superadnexuit
".
88Cf.Kunz,op. cit.,p. 13: Wahrscheinlich
hat derAutor,aufdenBoethiushier
den Hypermixolydius
zwar den achtenModusdes Ptolemusgezurckgeht,
nannt,aber in dem Sinne,dass er von Ptolemuserwhnt wordensei, was
ist".
richtig
89Boethius'awarenessofdirection
in tracingconsonances
is shownin hisintroremarkto tracingspecies:Utenima meseceterasordiamur,
diatessaron
ductory
consonantiae
speciessunttreshocmodo(338,17-18).
90Boethius'systemof numbering
is explainedveryclearlyby Kunz, op. cit.,
pp. 6-11.
91E.g. iv, 14 (339,12-15):Nam si ab
hypatehypatonincipiam;erit-AD-,id
estab hypatehypatonin mesonhypaten,ea, quae est in hoc ordineprima.
35
23:17:05 PM
Yet the most unusual aspect of Boethius' modal chapters is the use
of notation to explain an eight-foldmodal system.The extant treatises
employing notation are in the Aristoxenian tradition and as such
espouse a system of thirteenor fifteenmodes, or, to be more exact,
keys.92Evidence of the direct background of the Aristoxeniansystem
is present in Boethius' charts where the mixolydian mode is listed as
mixolydius vel hyperdorius.9ZThe treatmentof modes in the Aristoxenian treatisescasts each mode in threegenera,and merelylistsmodes
one after the other with no attempt to representintervals between
notes or relative position between modes. Boethius' chart,on the other
hand, contains eight modes, in only the diatonic genus, representedin
such a way that tones and semitones are clearly'discernable and the
relative position of each mode is immediatelyobvious. This chart,said
to be handed down fromantiquity,94and Boethius' text itselfare the
only extant documents which predicate the eight-foldmodal system.
The only otherwitnessforthis systemis Ptolemy,who findsthe eighth
mode otiose since it is harmonicallyidentical with the first,and thus
he argues forlimitingthe number of modes to seven.95
3) Possible sources: When raising questions concerning possible
sources for Boethius' modal chapters one is immediately impressed
with two rather paradoxical forcesin the text. On the one hand the
text is eclectic; for it draws fromthe notational traditionon the one
side and Ptolemy on the other,it presentsa new and unique numbering
system on the one side and uses the traditional on the other. But, at
the same time, the text is basically conservative; foralthoughit draws
fromvarious traditions,it neverthelessargues for a tonal system forsaken by the Aristoxenianson the one hand and rejected by Ptolemy
on the other, and it presents that system only in the diatonic genus.
It seems inconceivable that Boethius, the sixth centuryRoman patrician, could have writtenthis text or even compiled it fromthe various
92See e.g. Clenides,Isagoge(ed. JanS.167-207),12 (JanS.203-205);Aristides
De musica(ed. R. P. Winnington-Ingram,
Leipzig1963),i, 10,20,
Quintilianus,
10-21,4.
93Concerning
the termkey, and the late Aristoxenians,
hyper-nomenclature,
Modein AncientGreekMusic, Cambridge1936,
see R. P. Winnington-Ingram,
pp. 19-21.
94I, 15 (342,27-343,1): ". . . ab antiquistraditamusicisdescnptio. . .
95See Harmonicaii, 9-10; yet we may exaggeratePtolemys rejectionof an
eighthmode,forhe concludeshis discussionof spacingthe modeswitha pro:
visionforan eighthmodeas follows
xai otc
Tiva St7raac5v
xocv
StjXovtl
r A, Tovtaav
Xp<(xev
. . . (64,7-9).
tvx(ievov
u7repoxv)v
tt)v7rp
36
23:17:05 PM
threads representedtherein.In the firstplace Boethius bases his early

workson prexistentGreek sources, and in the second place, Boethius
is not so original and creative a thinker.For obvious reasons I reject
Pizzani's thesis that the modal chapters were based essentially on
Ptolemy; furthermoreI have shown that Kunz's requirementof two
sources does not necessarilyfollow. I would argue that these chapters
were translated froma single Greek source writtenduringthe lifetime
of Ptolemy or later, by a writerbasically conservative in his attitude
toward musical theory,yet acquainted with a broad spectrumof musical thought.
F. Intervallic tests: The last chapter of Book IV bears such obvious
resemblance to Ptolemy's Harmonica i, 8, 18-19, that Pizzani and
Gushee have stated that it is taken directlyfromPtolemy.96One suspects that the close resemblance of this chapter to Ptolemy may even
have lead them to see Ptolemy as a source for the precedingchapters
where there is not such obvious resemblance. The similaritybetween
the chapter of Boethius and that of Ptolemy begins even in the titles:
tc5vau(j.9mc5v
o Xyot
8i
StarxTox;
Ptolemyi, 8: TtvaTprcov
SetxOvjaovTai
toufxovoxpSou
xavvo.
(16,32)
Boethiusiv, 18: Quemadmodum
indubitanter
musicaeconsonantiaeaure
diiudicarepossint.(248,4-5)
Althoughthe firsthalf of Ptolemy's chapter (16, 32-17, 26) is not found
in Boethius' text, the last half of Ptolemy's text shows such substantial
concordances with Boethius chapter 18 that scholars have correctly
posited it as an obvious source of Boethius' text. But all too obvious
between
parallels have led these scholars to overlook crucial differences
the texts of Ptolemy and Boethius, differenceswhich not only suggest
that the passage was not translated directlyfromPtolemy, but which
aid in identifyingthe probable source. Boethius' text follows that of
Ptolemy until the actual testing of intervals, but at that point the
similarityceases. Ptolemy is concise and abstract:
TOioTCv
jEK StaaTaeoTeaapcv,
ouveotvy)KH
(lvypXafxPavofjtvTj
oi *xdcTepov
auTvcpoyyot
Tptcv,
7rot7j<Touai
tt)v8t Tsaapcov
auficpovav
Stfv7UTptTov
Xyov.
(19, 1-4)
Boethius, on the other hand, expands the text to describe the "interval" of the diatessaron, known by strikingalternately,and the "consonance" of the diatessaron,known by strikingboth parts at the same
time:
9e Pizzani,op. cit.,pp. 137-138;Gushee, cit.
op.
(n. 26), p. 379.
37
23:17:05 PM
vicissim.EK. et .KF. plectroadhibitopellantur,

. . . atque alterutra
diasin verosimulutrasquepercussero,
tessarondistantiaconsonabit,
diatessaronconsonantiam
nosco.(348,24-349,1)
And again concerningthe diapason:
Item si diapasonconsonantiam
temptarevoluero,totamtribuspartibus
seco atque in unamduasquedistribuens
easdemsimulvel alterutram
pulsane,quid consonetvel quiddissonetutraque,cognosco.(349,5-8)
The distinction between interval and consonance is absent from
Ptolemy, as is the notion of strikingsimultaneouslyand alternately.
Moreover, the similaritybetween the "simul percussero" and "simul
pulsans" found in the last chapter of Book IV and the "simul pulsae"
foundin the firstchapterofthe same book sounds more like intentional
harmony than coincidental echo. But this is not the only imprintleft
on the text by a hand otherthan Ptolemy. Ptolemy tests the following
intervals in his text: 4:3, 3:2, 2.i, 8:3, 3:1, and 4:1 (19,1-15).
Boethius carries his tests throughthe triple proportion,but the test
of 8:3, the diapason and diatessaron, is omitted. The Pythagoreans
had rejected the diapason and diatessaronas a consonance because the
proportion8 : was neithera multiplenorsuperparticularproportion.
Ptolemy argued that such rigor on the part of Pythagoreans went
beyond reason {Harmonica i, 6), and thus he pointedlylisted it among
consonances. Boethius records in Book II, 27, that Nicomachus had
said much against consideringthe diapason and diatessaron a consonance, but that he himselfhad abbreviated Nicomachus' arguments:
nostarnen,qua potuimusbreviSed quamquamde hismultaNicomachus,
tate partimea ipsa, quae Pythagoriciaffirmant,
partimex
promentes,
isdemquaedam consequentiaargumentantes
probavimus,si diatessaron
ex his coniunginon posse.
consonantiaediapasonaddatur,consonantiam
(260,12-16)
One might again attempt to argue that Boethius is responsibleforthe
"editing" of Ptolemy in this last chapter of Book IV, but such doctrinaire creativity is not typical of Boethius. The editing of Ptolemy's
text much more probably took place at the hand of a Greek Pythagorean who had already used the concept of "simul pulsae" in defining
consonances and rejectedthe diapason and diatessaronas a consonance,
namely Nicomachus of Gerasa.
Book IV, Nicomachus, and theunityofBooks I -IV
Viewed fromthe perspectiveof the above analysis, Book IV can be
perceived as a unified whole. If one concerns oneself with only the
V.
38
23:17:05 PM
individual parts, one might,like Pizzani, view this book as a "wild

forest"and label it an outright"guazzabulio".97 But if one rises above
the trees, one can view the forest as a whole, a forestwith clearly
definedboundaries and a distinctlyunifiedcolor.
In the firstplace, the book is whollyPythagorean: fromthe opening
passage excerptedfromSectiocanonis,throughthe doctrinairedivision
of the monochord,finallyconcludingwith the testingof only multiple
and superparticularintervals,the Pythagorean boundaries of musical
theory are clearly defined. The only two sections not overtly Pythagorean are the chapters on notation and modality symmetrically
framedby the Pythagoreanparts. Moreover,these two sectionsbetray
sufficientindividualityto make them clearly not part of Aristoxenian
or Ptolemaic territory.The notational chapters, unlike the notational
treatises,integratethe genera into a single list of twenty-eightnotes,
and the chapters on modality reveal a staunchly conservative,albeit
eclectic, character. Furthermore,the emphasis on the diatonic genus
to the exclusion of the enharmonicand chromatic in the modal discussion is thoroughlyconsistentwith the Pythagorean bias of Book
IV.98
But in the second place, the hue of Book IV, like the firstthree
books of the treatise,is distinctlythat of Nicomachus of Gerasa. The
division of the monochord is demonstrably that promised by Nicomachus. The simulpulsae in the definitionsof consonance foundin the
firstand last chapters of the book furtherreveal the imprintof Nicomachus' djj
vxxpouCTvrs.
Furthermore,the arithmeticadditions to the
from
Sectio
canonisfollowthe methodsofNicomachus,
geometricproofs
who tended to express both magnitude and multitude in integers.99
The doctrinaireexclusion of the diapason and diatessaron as consonance is one of the most obvious traces of Nicomachus, for his conservatismon thispoint even made Boethius abbreviate what the fundamentalisthad to say. Such conservatismis consistentwithNicomachus'
arithmeticalthought,forwho else but such a doctrinairePythagorean
could write a mathematical treatise in the second centuryA.D. and
" Pizzani,op. cit., 133.
,8 Onecannothelpp.
butask whether
theeight-fold
modalsystemis nota Pythaand criticizesPythagoreanelementin itself.PtolemyattacksAristoxenians
who
goreans;his attackin Harmonicaii, 9 is clearlydirectedat Aristoxenians
use all minuteintervals
to expandthenumberofmodes(to 13or 15).Is Ptolemy
the Pythagoreans
in Harmonicaii, 10, whenhe rejectsthe eighth
criticizing
mode?
differences
betweenEuclidian and Pythagoreanmethodssee
Concerning
D'Ooge, op. cit.,pp. 46-48.
39
23:17:05 PM
100Thus the conservatism of

completely ignore irrational numbers?
the chapters treatingthe modes is again fittingwith Nicomachus, and
the individualityand eclecticismof these chapters reflectstraitsfound
in Nicomachus' other works.101
The unity of Books I-IV can be demonstrated in three ways: i)
Referencesin the firstthree books to a monochord division following
Pythagorean principles (i, 11), to fixed and movable notes (i, 27), and
to a satisfactory proof that superparticular proportions cannot be
102
clearlypredicate the contentof Book IV. 2) Book V
equally divided
with
the
followingstatement:
begins
divisionem
adiciendaesse arbitrar
Post monochordi
ea, in quibus
regnlaris
. . . (351,15-17)
discordant
veteresmusicaedoctoressententiaediversitate
The inferenceis clear: the preceding representeda unified point of
view, and what followsdeparts therefrom.At this point Boethius put
aside the source that he had used in compiling the firstbooks and
began a text based on Ptolemy. Book IV is obviously essential to
what preceded. 3) The point of view unifyingthese fourbooks is that
of Pythagoreanism; forthey argue deductively forthe properproportions of intervals in the diatonic genus, they rank these intervals by
theirmathematical integrity,they are suspicious of any role the sense
might play in musical judgment; yet in the end they let the ear hear
the sound of mathematical integrityin the division of the monochord.
All evidence points to the fact that the lost ) j.ovaixrof
Nicomachus of Gerasa served as the source of these fourbooks. Definitions of terms in Book I follow those found in Nicomachus' extant
treatise; Nicomachus is cited concerningtheorieswhich are not found
in his extant works; these fourbooks are dependent on and consistent
with the De institutionearithmetica,a translation of Nicomachus'
; the monochord division of Book IV follows
) pi0[AY)Td
Nicomachus' specifications,and imprintsof his thoughtand language
100Concerning
numbersee B. L. van derWaerden,
of irrational
the discovery
in: Mathematische
dergriechischen
ZenounddieGrundlagendkrisis
Mathematik,
Annalen,117 (1940-41),141-161;Kurtvon Fritz,TheDiscovery
ofIncommenin: Annalsof Mathematics,
46 (1945).
by Hippasos of Metapontum,
surability
242-264;Becker,op. cit.(n. 21), pp. 455-465.
101Concerning
Nicomachus'use ofAristoxenian
theorysee FloraR. Levin,The
AmericanClassical
and thePythagorean
Harmonicsof Nicomachus
Tradition,
Studiesno. i, The AmericanPhilological
Association,
1975,pp. 26-27,56-63.
102The proofof Archytas(iii, 11, 285-286)is said to be unsatisfactory
and a
is foundiniv, 2,
validproofis promised.The onlyotherproofoftheproposition
303-3440
23:17:05 PM
are foundin the firstand last chaptersof the fourthbook ; finally,the

contents specified for the )
in the Enchiridion are
fxoixnx)
found in Boethius' firstfour books. De institutionemusica therefore
preserves at least the basic plan and contents of Nicomachus' lost
extended musical treatise. Boethius at least followedin the "path" of
Nicomachus, if not in his "footsteps''.103
The fundamentalimportance of Boethius1treatise, especially Books
I-IV, can now be judged more competently.This document represents
not only the Latin text which served as the essential textbook of
musical theory for over a millennium but it likewise preserves the
major musical treatise of Nicomachus. As such it is the only extant
complete document representingthe Pythagorean school of musical
thoughtapart fromthe Sectio canonis and the all too briefEnchiridion.
It is the record of a work contemporarywith Ptolemy, and thus must
be considered one of the two major treatises between Aristoxenus
(4th . ) and the late flourishingof Aristoxenian theory in the
persons of Clenides (late 2nd or 3rd c. A.D.), Aristides Quintilianus
(3rd c. A.D.), Gaudentius (3rd or 4th c. A.D.), and Bacchius (3rd or
4th c. A.D.).104 In short,the attitude of recentiorergodeteriormust be
forsakenand the work must be given its proper place as a valid and
representativedocument in the historyof Greekmusical theory.Works
such as Winnington-Ingram'smonumental Mode in Ancient Greek
Music ,105which does not even mention the De institutionemusica,
must be reconsidered to include the place of second-centuryPythagorean thoughtin the development of music in antiquity.
VI.
Nicomachus and Ptolemy

If my thesis that Boethius' treatiseis a translationof Nicomachus is
accepted, we may be able to date more exactly the lifetimeof Nicomachus. Previous datings have been dependent on primarily two
givens: 1) Nicomachus' mentioningof Thrasyllus in the Enchiridion
(JanS. 260, 16) makes the year 14 A.D. the terminuspost quem of
Nicomachus' life,forThrasyllus flourishedduringthe reignof Tiberius
(A.D. 14-37); Cassiodorus records106that Apuleius of Madaura (b. ca.
103How Pythagorean
to composetwoworksofapproximately
thesamelength
on arithmetic
and music,but withthe arithmetical
treatisedividedintotwo
books(twois the beginning
of number)and the musicaltreatisedividedinto
four(fourrepresents
the harmonictetractys)
!
104The datingoftheselate
can onlybe approximate,
forno clear
Pythagoreans
documentation
exists.
106See above n. 93.
10eInstitutiones
(ed. Mynors),ii, 4, 7.
41
23:17:05 PM
A.D. 124) translated Nicomachus' arithmetical treatise into Latin,

therebymaking the lifetimeof Apuleius the terminusante quem. The
dates in which Ptolemy flourished,on the other hand, can be rather
preciselyestablished, forastronomicalobservationsfromthe Almagest
fall between A.D. 127 and 141.107Three pieces of evidence that Nicomusica:
machus knew Ptolemy's Harmonicsare foundin De institutione
1) the final chapter of Book IV presents a text clearly Ptolemaic in
originbut with obvious alterationsby Nicomachus. 2) Boethius' statement in Book II, 27 that Nicomachus wrote much against considering
the diapason and diatessaron a consonance in all likelihood betrays
Nicomachus1 knowledge of Ptolemy; for Nicomachus would not have
gone to such great lengths to refuteAristoxenusor his disciples, who
did not even express intervals in terms of numerical proportions.
3) The modal chapters of Book IV show some reflectionof Ptolemaic
language and method, and if Book IV is translated fromNicomachus,
then Nicomachus knew of Ptolemy's modal theory. Finally, the socalled Nicomachus Excerpta contain a direct referenceto Ptolemy,
specifically Ptolemy's doctrine that only the system of fifteennotes
contains all species of diapason (JanS. 275, 7-15).108
Thus the position that Nicomachus lived before Ptolemy since he
did not mention Ptolemy, advanced by Meibom,109and subsequently
repeated by Jan and Robbins,110must be rejected. Levin's placing
Nicomachus' floruit around 100 is likewise too early.111Nicomachus
must have flourishedat the same time as Ptolemy, or even slightly
later. Apuleius' translationcould readily have been writtenas late as
175, a time when Apuleius was around 50 years old. Therefore,Nicomachus' extended arithmetical and musical treatises probably date
fromthe years between 150 and 175. The Enchiridion obviously precedes the musical treatise, and probably antedates the mathematical
work as well since it does not mentionit and since the two works on
music and arithmeticseem to have formed a pair. Nicomachus' ac107Pauly-Wissowa,
derclassischen
23, 2
Altertumswissenschaft
Realencyclopdie
(1959)cols. 1788-89.
108Concerning
the Excerpta,see above n. 20. M. Meibom,AntiquaeMusicae
see p. 58, noteson pag.
Auetores
, Amsterdam
1652(underNicomachus,
Septem
36, v. 8, v. 10) arguesthatthe comments
Ptolemywereadded by
concerning
thecompileroftheextracts.
109Meibom,op. cit.,praefatio
to Aristides
to Nicomachus;But in the prefatio
Bacchius
Gerasenus,
anus,Meibomstates:"Reliquiomnes,Nicomachus
Quintili
Senior,Bothius,Ptolemaeojuniores".
110JanS.221; D'Ooge, op. cit.,p. 71.
111Levin,op. cit.(n. 101),p. 10.
42
23:17:05 PM
quaintance with Ptolemy's works is furtherexplained by his probable

location in Alexandria, a locus presented convincinglyby Robbins for
the flourishingof the scholar and Pythagorean, Nicomachus.112
VII.
Book V and theOriginalScope of De institutionemusica
Having completed the translationof Nicomachus' Eisagoge in Book

IV, Boethius began translatingPtolemy's Harmonica to complete his
Latin record of Greek musical thought.Only nineteen of the proposed
thirtychapters of Book V are extant, but from the titles of the remaining chapters we can follow Boethius in Ptolemy's path through
all of the firstbook of the Harmonica. Those chapters in Ptolemy
which cover material already treated in the firstfour books are condensed by Boethius.113The contentsof chapters coveringnew matrial,
especially ideas unique to Ptolemy, are translated in such a way that
Ptolemy's sequence ofthoughtand basic ideas are clearlyand distinctly
presented. Pizzani, followinghis notion of Boethius' incompetence,
tries as he may to discredit Boethius' translation of Ptolemy, calling
attention to his compressionsof some passages on the one hand and
criticizinghis diffusestyle on the other.114Yet in no case can he produce a Greek passage that Boethius mistranslated or misunderstood
in any significantdetail. Book V thus furtherconfirmsBoethius' ability to read, comprehend,and renderinto Latin a Greek that is highly
technical and even awkward and difficultin style.115
Whether Boethius ever completed the treatise remains an unanswered question, although if he did complete works on geometryand
astronomywhich are now lost,116the loss of parts of the musical work
is easily conceivable. But even if Book V were preservedin a complete
state, the projected plan carefullylaid out in the beginningsof Books
I and V would remain unfinished.The opening passage of Book V
promises another "divisin", using not just one string, but rather
eight, whereby "the complete calculation of the proportionsmay be
discernedrightbeforethe eyes in whatever number of notes might be
necessary".
112D'Ooge, op. cit.,pp. 74-75.
113E.g. v, 13; cf.Ptolemy,Harmonica,
i, 9-10.
114Pizzani,op. cit.,pp. 139-156.
115Concerning
Ptolemy'sstylesee Winnington-Ingram,
op. cit. (n. 93), p. 62:
"... a lucidthinker
ifan ungainly
. . ." ; p. 67: "It is explainedin difficult
writer
withtheinadequaciesofexisting
language,as thoughPtolemywerestruggling
. . .".
terminology
116See Brandt,op. cit.
(n. 4), pp. 152,234-236.
43
23:17:05 PM
Potestenimalia quoqueessedivisio,in qua nonunustantummodo

nervus
verumocto,atque huiusadsumitur,
dividatur,
qui positisproportionibus
modifiatcithara,ut in pluribuset quantinecessarii
sintnervitotaproportionumratio quasi oculis subiectacernatur.. . . Sed de his paulo post
loquemur.(351,19-27)
This referenceclearly relates to the tables mathematicallydescribing
the various genera and modes in Book II of Ptolemy's Harmonica.111
Since this material is not mentioned in the chapter titles of Book V,
theremust have been at least one otherbook projected,ifnot finished,
in Boethius' treatise.
Boethius' famous division of music into musica mundana, musica
humana, and musica in quibusdam instrumentisconstituais found in
the second chapter of Book I (187, 20-22). Aftera*relativelybriefdescription of musica mundana, Boethius promises to discuss it more
thoroughlyat a later time: de quibus posteriusstudiosiusdisputandomi
est (188, 25-26). Likewise aftera survey of musica humana he promises
furtherdiscussion: Sed de hac quoque posterius dicam (189, 5). The
musica, however,never returnsto discuss
extant text of De institutione
these two types of music. But if Boethius had finishedtranslating
Ptolemy's Harmonica, the outlined plan of the work would have been
completed. Harmonica III, 4-8, explicates the relation between principles of music and the physiologyand psychologyof man, while Harmonica III, 9-16,thoroughlyexamines musical principlesat workin the
universe.It is interestingto note that, since musica humana and musica
mundana are not treated in the firstfourbooks, this musical division
was not part of the originaltext of Nicomachus. Thus just as Boethius
seems to have coined the concept of quadruviumin the introductionto
De institutionearithmetica
, a concept not foundin Nicomachus' treatise,
so Boethius may have conceived the three-foldconcept ofmusic placed
in certain instruments,music of the human being, and music of the
universe in the proemiumof De institutionemusica. In fact the first
two chapters of the treatise,which expresslyformthe proemium(189,
12), may well representas much Boethius' personal expression concerningmusic just as the rather freeintroductionto the arithmetical
treatise seems to reflectthe Latin author ofthat text as much as it does
its Greek model. On the otherhand, a certainamount ofthe contentsof
the introductions to both the musical and arithmeticaltreatises may
representGreek glosses in the manuscriptsBoethius was using.118The
117Ed. Dring,pp. 70-80.
118See aboven. 7. The proemium
is definitely
not,as Pizzaniclaims,based on a
on Boethius'use of
Latin source.Pizzani,op. cit.,p. 162,buildsthisargument
44
23:17:05 PM
more exact translation in the De institutionearithmeticabegins after

the introduction(i.e. with I, 2), and thus Book I, 3, probably represents
the closer followingof the Nicomachan path in De institutionemusica.
The overall plan for Boethius' musical treatise thus becomes clear:
the work was to consist of seven books, the firstfour based on Nicomachus' Eisagoge and the last three on Ptolemy's Harmonica. The
plan is consistentwith, indeed representativeof, Boethius' mind and
worksas a whole. He was not what we would call an "original thinker"
until he was forced to be in his last years. Boethius was primarilya
moderatorbetween works of Hellenistic civilization and the culture of
his own period, and between conflictingviews expressed by classical
writers.The great philosophical project of his 'life was to translate
Aristotle and Plato into Latin and show the essential agreement between them.119The musical treatise followed the same pattern:
Boethius chose the two most viable theorists known to him, Nicomachus and Ptolemy, served as moderatorbetween them, and translated their basic ideas into Latin. The text of Ptolemy has survived
intact, but we may still read the lost treatiseof Nicomachus in the first
fourbooks of Boethius' De institutionemusica.
Chapel Hill, N.C., U.S.A.
Universityof NorthCarolina
thewordsubphrigius
forhypophrigius.
But subphrigius
is a typographical
error
modisono(185,1) shouldreadsubphrigiimodisono,as
by Friedlein;subphrigii
confirmed
a fewlineslater(185,6) : sonohrigiimodi.
119Perihermeneias
Commenta
ii, 79, 1-80,9, and In IsagogenPorphyria
i, 131,
Plan des Boethius,in:
25-132,2. AlfredKappelmacher,Der schriftstellerische
Zeitschrift
frklassischePhilologie,46 (1929),pp. 215-225,suggestthat this
projectmay not have been an originalconceptionof Boethius,but rathera
featureofBoethius*
Greeksources.
45
23:17:05 PM
Vivarium
, XVI, i (1978)
Consequences et inconsequences de la
supposition vide dans la logique d'Ockham
ELIZABETH KARGER
peut-ilfairefigurede prcurseurde la thoriede la quantification? Si l'on complte le chapitre i, 70 par le chapitre 111-3,
Ockham
32 de SL, on constate que sont quivalentesles propositions:
et
(1)
Tout homme court
(2)
Cet homme-cicourt et cet homme-l court, etc.
ainsi que les propositions:

(3) Quelque homme court
et
(4) Cet homme-cicourt ou cet homme-l court, etc.
D'autre part, il est possible,1 en logique quantifie, de tenir pour
quivalentes les formules:
et
(a)
(x)Fx
(b)
Fa & Fb & . . .
ainsi que les formules:

et
(c)
(Ex)Fx
(d)
Fa v Fb v . . .
Le parallle est certes assez frappant pour suggrer une analogie

entreles termessyncategormatiques"omnis", "aliquis" et leurs variantes (que, suivant en ceci Geach [i], nous appellerons "applicatifs")
et les quantificateurs.2
1 La conditionest que l'on ait autantde constantesque d'objets,ce qui est
impossiblesi le domaineestl'ensembledes rels.
2 Cf.Matthews[2]qui cited autresrfrences.
Il estutilede signaler
que 1intersur les bases qui viennentd'etreindiques,ne se
prtationdes catgoriques,
commeun nonc
aveccellequi littoutecatgorique
confond
pas ncessairement
46
23:17:17 PM
Cependant cette thse a t vivement conteste par Matthews [2].

Nous pensons que ses argumentssont bien fonds,mais qu'il n'en a pas
degag la vritableporte. Nous montreronsque les conclusionsqu'impose un rxamen minutieux des textes sont diffrentes:ce quoi il
faut renoncern'est pas l'analogie entreapplicatifset quantificateurs,
mais trouverchez Ockham une logique une et homogne. Nous verrons que deux logiques sont ncessaires pour justifiertoutes les infrencestenues pour valides par Ockham.
Voyons d'abord les sources. Les commentateurss'appuient uniquement sur le chapitre1, 70 de SL, or est essentielde compltercelui-ci
par chapitre 111-3,32. D'aprs SL 1, 70, il est possible de "descendre"
(conjonctivementou disjonctivement) des propositionssingulires
partir de propositionsgnrales (universelles ou particulires),autrement dit, on peut infrerles premiresdes secondes. On peut galement
infrerune propositionparticulired'une singulire,et, a fortiori,d'une
disjonction de singulires;on peut enfin,d'aprs SL 111-3, 32, infrer
une universelled'une conjonction de singulires.Mais on se tromperait
si on croyait que ces infrencessont "toujours" possibles. Au contraire,
d'aprs SL 111-3,32, l'infrencede singulires,d'une universellecomme
prmisse, ne se fait pas lorsque le sujet de l'universelleest vide, soit,
dans le jargon technique mdival, lorsqu'il ne "suppose pour rien".
Convenons d'appeler ces infrencesde singuliresqui ne sont possibles
qu' condition que le sujet de l'universelle (ou de la particulire) ne
soit pas vide, "descente sur le sujet".
Il n'y a donc pas "descente sur le sujet" partirde toute proposition
gnrale. Ceci vrifi, premirevue, ce qu'avance Matthews[2, p. 22] :
d'une propositionuniverselle(ou particulire)vraie, mais dont le sujet
est vide, on ne peut descendre aucune singulire. Ainsi d'une proposition telle que
(5)
Aucune chimren'est bleue
et qui est vraie d'aprs Ockham du fait justement qu'il n'y a pas de
chimres,on ne pourraitinfrer
(6) cette chimre-cin'est pas bleue
doublementquantifi.Certainscommentateurs,
s'appuyantsur le mmech.
I, 70, ontcrulgitimede considrer
que le prdicatest prcdd'un applicatif
implicite.Par exemple'quelque animal est homme'se liraitcomme'quelque
animalest quelquehomme'.On trouveraune excellenteversionde cetteinterainsique la reprsentation
prtation,
symboliqueappropriedans Price[3]. Or
telleobjectionpeuttrefaite cettedernireinterprtation
(cf.Matthews[2],
le sujet".
p. 18-20)qui ne s'appliquepas cellequi se borne "quantifier
47
23:17:17 PM
et comment le pourrait-on en effet,puisqu'on ne peut en dsigner

aucune ?
Malheureusement,la doctrine Ockham n'est pas simple. S'il est
exact qu'il n'y a pas de descente partird'une proposition sujet vide,
il est faux qu'on ne puisse infrer- par aucune sorted'infrence- ,
une singulire.Justement,on peut de (5) infrer(6).3
En effet,nous lisons, dans SL 111-3, 32, que, dans l'hypothse o il
n'existe pas d'objets blancs actuellement,la proposition
(7)
Aucune chose blanche n'est colore
tant vraie, implique

(8)
Cette chose blanche-ci n'est pas colore
et ceci quelque soit l'objet (prsent) que l'on dsigne.

Il en serait naturellementde mme si Ockham avait pris pour exemple une proposition telle que (5), c'est a dire sujet non pas vide
momentanment,mais toujours (et ncessairement).4
Telle est la doctrinecomplexe qu'il nous faut lucider.5
3 Matthews[2] p. 22 critprudemment
qu'il n'y a pas, dansce cas, de descente
surle sujet,ce qui estexact (cf.note5).
4 II est exclude tenterde comprendre
(8) surle modlede 'Socraten est pas
color'et de penserque (8) est vraiedu faitqu'unechoseblanchepasseserait
dsigne.Un secondexempledonnpar Ockhamdansle memech. 111-3,32 de
commevraiedu faitqu'une
SL montreque la proposition
(8) doittreconsidre
choseactuellement
existante,maisqui n'estpas blancheest designe.
5 Quelquesobservations
de Matthews[2],p. 21-22:il crit
surl'argumentation
gnrales sujet vide. En
qu'il n'y a pas de descente partirde propositions
de (6) partirde (5) n'estpas un cas de descente.
ceci il a raison;l'infrence
sereinequ'il tirede
relativement
c'estla conclusion
ce qui surprend,
Toutefois,
selonlui, qu'il fautrenoncer trouverchez
ce fait.En effet,il s'ensuivrait,
de la quantification
Ockhamunethorieembryonnaire
et,par suite,je suppose,
Notresurles applicatifs
reprsenter
faut-ilrenoncer
par des quantificateurs.
car ellene l'est
soitillgitime,
prisene provientpas du faitque cetteconclusion
soituneconjonction,
pas toujours
gnralesn'impliquent
pas : si les propositions
soit une disjonctionde singulires,
l'quivalenceentreles unes et les autresa
la condition
tantque le sujetdesproposi-.
unevaliditconditionnelle,
galement
tionsgnralesne soitpas vide.Si l'on pensaitavoirdfinipar ces quivalences
la questionse poseraitde savoircommenton doit les
le sens des applicatifs,
sujet vide. Notresurprisevientdu
dans le cas de propositions
comprendre
faitque Matthewsne semblepercevoirici aucun motifd'inquitudepour la
soituneinfrence,
Et pourtant
que signifie
logiqued'Ockhamdanssonensemble.
estde
soitunequivalencequi auraitune "validitconditionnelle"
(l'expression
de la seuleproposition
uneproposition
d'infrer
A, ce
nous)? S'il estlgitime
A et B. Or la forme
soit treen vertude la formeseulementdes propositions
sujet non vide ne se distingueen riende la formed'une
d'une universelle
valided'uneproposition
sujetvide.S'il y a uneinfrence
universelle
singulire
48
23:17:17 PM
Il y a donc deux sortes d'infrencesde propositions singulires

partirde propositionsgnrales,mais Ockham fournit-ilun critrequi
permettede distinguerl'une de l'autre ces deux sortes d'infrences?
Il criten substance que, dans le cas o la prmisseest une proposition
sujet non vide, et seulementdans ce cas, on obtient par infrencedes
singuliresqui sont des singulires "de" l'universelle de dpart. Que
sont des "singuliresd'une universelle"? Elles sont dfiniespar le fait
6
que le sujet de l'universelleest vrifi de l'objet designpar leur sujet.
Un exemple permet de comprendre:la proposition:
(7')
Aucune chose blanche n'est colore
permet d'infrer- par descente sur le sujet - la proposition:

(8')
Cette chose blanche-ci n'est pas colore
seulementsi
(9)
Cette chose blanche-ci est blanche
est vraie, o se trouve dsign le mme objet que dans (8').

Mais avons-nous un vritable critre? Ce que nous savons, c'est que
si (9) est vraie, s'ensuit que le sujet de (7') n'est pas vide. Ou encore,
moins que le sujet de (7') ne soit pas vide, on ne peut infrerla
proposition(8'), entendue comme une "singulirede l'universelle (7')".
(7') serait-ellevide, (8'), commesingulirede l'universelle(7'), n'existerait pas.
Au lieu de fournirun critre,Ockham introduiten faitune catgorie
de propositions(singuliresd'une universelle)qui n'existentque si certains termesne sont pas vides; par suite les infrencesqu'elles dfinissent ne sont-ellespossibles que sous cette mme condition.Mais ce que.
ceci suggre,c'est qu'Ockham a constitu en ralit une logique part
pour les propositions sujet non vide, ou plus precisement,qu" les
infrencespropres ces propositions("descentes" ou "montes" 7) sont
justifies l'intrieurd'une logique dont la dfinitioncontientla stipulation que ls sujets ne sont pas vides. Puisque ces mmes termes
peuvent cependant tre vides, Ockham aurait donc deux logiques dansun cas,la mmeinfrence
devraittrepossibledansl'autrecas. Cependant
Matthewssemblebien supposerque les "descentes"sont, dans la doctrine
valides(p. 17). Dans ces conditions,
comment
d'Ockham,des infrences
peut-il
omettrede s'interroger
sur la natureet la cohrence
de la logiqued'Ockham?
Nousemployons
ce termeau senso Ockhamemploie'verificatur'.
7 Appelons"monte"l'infrence
soit d'une particulire
partird'une disjonctionde singulires,
soitd'uneuniverselle
partird'uneconjonction
desingulires.
49
23:17:17 PM
telles que les "singulires d'universelle" et les infrencesqu'elles dfinissent n'existent que dans l'une. L'analyse de la "deuxime" sorte
d'infrencepermet de confirmercette interprtation.
Dans l'infrencede (8) partirde (7), considronscomme donn que,
les propositionsgnralesngatives tant automatiquementvraies lorsque leur sujet est vide, (7) soit vraie. Mais pourquoi la proposition (8)
est-elle vraie ? Comme nous l'avons vu (cf. note 4), ce ne peut tre du
fait que son sujet soit vide, c'est dire ne dsignerien de prsentement
existant,comme c'est le cas de 'Platon' qui, dans la proposition'Platon
ne court pas', ds lors que celle-ci est nonce aprs la mort de Platon,
ne dsigne aucun individu existant. Il reste deux hypothses:
(i) le sujet (soit 'cette chose blanche-ci') est comprendrecommeun
nom. Mais ceci ne serait dfendable que si l'on supposait qu'en employant cette expressionon peut ne pas avoir gard son sens. Le cas
serait semblable, pour reprendreun exemple familier, celui de l'expression 'toile du matin', employe le plus souvent sans intention
d'indiquer que ce qui est en faitune plante est une toile, ni qu'elle est
visible le matin.8Mais cette hypothse est rejeter: interprteainsi,
l'expression 'cette chose blanche-ci' pourrait bien etre employe pour
dsignerun objet vert, auquel cas (8) serait fausse.
(ii) Il reste la seconde hypothse. (8) est vraie parce qu'elle contient
deux assertions (ici, et dans la suite, au sens de "pense complte" (cf.
Frege [5])) ; elle est comprendrecomme nonant en ralit:
(10)
cette chose-ci n'est pas blanche ou cette chose-ci n'est pas

colore 9
Mais si (10) est la proposition rellementimplique par (7), il devient

clair que l'infrencecorrespondante(de (10) partirde (7)) ne peut en
effettre rgie par la mme regle logique que l'est l'infrencede (8')
partir de (7'). En effet(10) n'a pas la mmeformeque (8'). (8') est une
8 Cecirappellela notionde "dsignateur
rigide"(Kripke[4]).
(9) est comprendre
commenonanten realit:
(9') Cettechose-ciest blancheet cettechose-ciestblanche.
ceci s'expliquedu faitqu'Ockhamdfinitles singu(9) est doncrptitif;
dans la pro un termegnralfigurant
nonpar rfrence
liresd'universelle
Il
auraitt
au
de
l'universelle.
mais
rfrence
par
sujet
positionsingulire,
autrement
obligde procderen deux tapes:
devra tre exprimepar un
(i) stipulerque toute singulired'universelle
dmonstratif
accompagnd'un termegnralpluttque par un nomproprepar
exemple,
de forme'cet F est G' une classe: cellequi
parmiles propositions
(ii) dfinir
que 'cet F est F' soitvrai.
prsupposerait
50
23:17:17 PM
propositionsimple, c'est dire qu'elle ne contient qu'une seule assertion, alors que (10) est molculaire,puisque disjonctive. Dans ces conditions,on ne peut chapper la conclusion que nous avons bien deux
logiques; nous voyons present qu'elles sont dfiniespar des lectures
diffrentesdes propositionscatgoriques (du moins singulires).
La suite de cet expos consistera prciser quelles sont ces deux
lectures des propositionscatgoriques, et comment doivent tre caractrissles deux systmeslogiques correspondants.
La dfinitiond'une "descente" comme infrenceexclusivementd'une
"singulired'une universelle"garantit,nous l'avons vu, que l'infrence
ne pourra se faire si le sujet est vide. Nous suggrionsque la logique de
ces infrencesest dfiniepar la stipulation, ou disons dsormais,prsuppositionque le sujet n'est pas vide; du moins est-il vrai que, "supposant la prsuppositionleve", l'infrencene pourrait se faire. Et on
ne voit pas de quelle autre faon on pourrait rendre intelligiblel'ide
que, sous certaines conditions,une infrenceautrement valide serait
impossible.10
Il faut noter toutefoisque si cette interprtationrend comprhensible la doctrine d'Ockham, elle montre dans le mme temps ce que
celle-ci comporteillgitime: on ne peut "lever une prsupposition".11
Nanmoins Ockham a bien pu commettrecette erreurconceptuelle,vu
qu'il n'tait pas quip pour la dceler.
Nous savons donc de cette "premire" logique, que les sujets (des
propositionsgnrales) y sont prsuppossnon vides. Il est ais de voir
quel systeme de logique symbolique reprsenteracette logique. Puis10II se pourraitqu'aucuncas de descentene soitune infrence
valide: les descentespourraient
n'treque des inferences
"ut nunc",lesquellessontprcismentdes infrences
non valides,puisqu'ellesdpendentde conditions
continvoir SL 111-3,1). Cependant,devrait-ilen tre
gentes(pourleur dfinition,
ainsi,il demeureque d'uneproposition
gnraleil est toujourspossiblede descendre quelquesingulire
dans le cas o le sujet
(cf.SL 1, 70)- sauftoutefois
de la proposition
tantdfinies
gnraleest vide. Ces infrences
par les formes
despropositions
qui y figurent
(parexemple'toutF estG' et 'cetF estG'), dire
que leurpossibilitest empchedans le cas o un termeest vide suggrea
nouveau l'interprtation
propose,notammentque cette conditionde non
vacuitdu sujet est en realitune prsupposition.
dfinit
Cetteprsupposition
unelogiquedanslaquelle,en effet,
certainesinfrences
sontpossibles(quoique,
enl'absenced'information
surles conditions
contingentes
requises,ellesdoivent
resterindtermines).
cette prsupposition
tant "leve", on
Naturellement,
sortde la logiqueainsidfinieet ces infrences
sontds lorsimpossibles.
11On peutnaturellement
avoirunelogiquesansprsupposition
que les domaines
soientnonvides,maison ne peutavoirunelogiqueo cetteprsupposition
soit
faitepourun cas et levepourun autre.
51
23:17:17 PM
qu' toute descente sur le sujet correspondune monte, nous obtenons

les quivalences mentionnesau dbut de cet expos. Ces quivalences
peuvent tre considres comme donnant des conditionsde vritdes
propositionsgnrales et comme justifiantl'analogie entre applicatifs
et quantificateurs.12Il est donc lgitime de choisir pour modle une
logique quantifie. Mais le fait que les sujets soient prsupposs non
vides suggre que leur fonctionest d'indiquer un domaine d'objets, de
sorte que les valeurs des variables lies sont obligatoirementprises
dans le domaine ainsi spcifi. C'est dire que nous devons adopter un
domaines prsupposs non vides.
systme de quantificationrestreinte
Il est inutile de prsenterun systmede faon formelle,notons simplement que les quatre catgoriques sont reprsentercomme suit:
(x)pGx
(x)f-Gx
(Ex)pGx
(Ex)f-Gx
o la lettresouscrite'F' correspondau sujet et la lettre'G' au prdicat.
Reprsenter ainsi les catgoriques gnrales, c'est privilgierune
certaine lecture des ces propositions. Par exemple on interprterala
proposition'tout animal est blanc' comme une propositionpar laquelle
on dit des animaux: 'chacun est blanc', o il devient apparent que le
sujet ne faitpas partie de l'assertion (cf. Geach [2] p. 178). Remarquons
que l'interprtationrespecte la nature des catgoriques, puisqu'elle
s'accompagne d'un principe de lecture capable de justifierque, malgr
la prsence de deux termes, ces propositions ne contiennent qu'une
seule assertion.
Mais comment les "descentes sur le sujet" sont-elles reprsentes
dans cette logique? Puisque chaque proposition gnrale doit tre
quivalente une conjonction (ou disjonction) de propositionssingulires,le procd vident est celui d'associer chaque lettresusceptible
d'indiquer un domaine une liste de constantes individuelles. La rgle
prsidant aux descentes sur le sujet, partird'universelles,correspondra ds lors une rgle d'instantiation individuelle pour logiques
quantificationrestreinte,o provisionest faite pour garantirque l'instantiaton se fait par les constantes appropries.13et14
12L'objectionprsente
[2],p. 22 ne tientplus,puisquele cas des
par Matthews
sujetvide appartient une autrelogique.
propositions
13La descentesurle sujet partirde propositions
correspondrait
particulires
unexemplede raisonnement
existentielle".
Ontrouvera
la rglede "spcification
52
23:17:17 PM
Tel est le systeme de logique qui prside aux descentes sur le sujet,
soit la "premirelogique".
Quelle est maintenant la "deuxime logique" ?.
Nous savons que les singuliressont des propositionsmolculaires.
Se pourrait-ilque, dans cette logique, toutes les propositionscatgoriques doivent etre lues comme des assertions complexes ?
S'fl en tait ainsi, (7) devrait tre lue comme:
(11)
Pour tout objet, s'il est blanc, il n'est pas color
Evidemment (11) est vraie s'il n'y a pas d'objets blancs. Il est apparent
aussi que (11) implique bien (10) quelque soit l'objet dsign.11
Au cas o cette lecture d'une proposition catgorique paratrait au
lecteurpar trop inorthodoxe,signalonsque les noncs des dici de omni
et de nullo en confirmentla lgitimit.Citons le dici de omni: "par
cette proposition 'tout homme est animal' il est signifi(denotatur)
que de tout objet dont 'homme' est dit, du mme objet 'animal' est dit"
(SL 111-1,3.1. 12). Cette paraphrase d'une propositionuniversellemontre bien qu'on peut lire une proposition catgorique comme une pro- de faconerrone
o unerglecomparableestimplicitement
toutefois
employe
- dans SL 111-1,25 1.50SS.
14Cettereprsentation
des "descentessur le sujet" ne semblevalable que s'il
est tabliqu'une descenteest une "consequentiasimplex"et non "ut nunc"
(cf.note10). Nousne pouvons,dansle cadrede ce courtexpos,discuter fond
cettequestion.Notonssimplement
ceci: si les descentesn'taient
jamais quedes
ut nunc,cela indiquerait
consequentiae
que la logiqued'Ockham,o les termes
gnrauxontune extensionvariablede moment moment,ne contientpas de
termessinguliers
de ce pointde vue analogues.Cetteasymetrie
en
expliquerait
effetque les descentesne soientpas des infrences
valides(s'il est vrai qu'elles
ne le sontpas). Soit,pourillustrer
ce point,l'exemplesuivant:supposonsqu'il
soitvraia tAde dire:
toutindividujeune court,
et que, ce moment,
il soitlgitimede'infrer
"ut nunc":
Socratecourt,
t2,l'universelle
la singulire
faussedu
pourraitencoretrevraieet toutefois
faitque Socrateait avancen ge. Il est clairque si, au contraire,
nousavions
des termessinguliers
de dsignertanttun objet,tanttun autre,
susceptibles
il seraitpossibled'avoirdes descentesqui seraientdes infrences
"simplex"et
mmevalides. Il est tentantde penserque les expressions
de la forme'cet
individujeune'peuvent
cettefonction.
prcisment
remplir
15(7) pourraitimpliquer(10) sans
qu'il soitncessairede considrer
(7) comme
quivalente (11): il suffit,tout en conservantle langagede quantification
restreinte
la prsupposition
de nonvacuitdes domaines
propos,de supprimer
(on obtientune logique"libre".Voirpar exempleMeyeret Lambert[6],p. 9).
on nepourraitjustifier
ainsique touteset seulement
lespropositions
Cependant,
ngatives
puissenttrevraies"de faonvide".
53
23:17:17 PM
position o non seulement le prdicat, mais aussi le sujet est "dit de

quelque chose".
Il est vident que le systme quantifi dans lequel il convient de
reprsentercette "deuxime" logique est un systme de quantification
non restreinte,avec prsupposition de non vacuit des domaines.16
Nous savons que chaque propositioncatgorique gnrale est reprsenterpar une propositioncomplexe. Afin de dterminerquelle forme
propositionnelleconviendrait pour chacune, suffitde s'assurer que
les conditions de vrit donnes au dbut de la Partir II de SL, o
provisionest faitepour la suppositionvide, soient respectes.L'universelle ngative, il est vrai, n'est pas mentionne,mais nous savons dj
qu'il conviendra de la reprsenterpar
(x) (Fx->-Gx).
Pour les troisautres catgoriquesgnrales,par transcriptionpratiquement automatique des conditionsde vrit nonces, on obtient:
(x) (Fx -> Gx) & (Ex)Fx
pour l'universelleaffirmative,o '(Ex)Fx' est ncessaire pour garantir
la porte existentielleque doit avoir toute propositionaffirmative(cf.
Ockham [8] fol. 98).
Pour la particulireaffirmative,on a:
(Ex) (Fx & Gx)
et pour la particulirengative:
- (Ex)Fx v (Ex) (Fx & -Gx).
La rgle prsidant l'infrencede (10) partir de (11) correspond
videmment la rgle d'instantiationuniverselle.
Il suffitde doter notre systme de constantesindividuelles de faon
ce que chaque objet (existant prsentement)soit nomm.17
ie Nous mettons'domaines'au pluriel,car ceux-cichangentde moment

dansle sensde Montague[7]seraitvidemmoment(unelogique"pragmatique"
mentapproprie).
17Ce sontles pronomsdmonstratifs
ces constantes.
Si,
auxquelscorrespondent
de penserque les descentes
dansle cas des descentessurle sujet,il taitlgitime
un nompropre,ou un dmonstratif
o figurent
taientpossibles dessingulires
partirde
employseul,ici, par contre,c'est diredans le cas infrences
vraiespar vacuitdu sujet,on ne peutobtenirque des singulires
propositions
dontle sujet est un dmonstratif
accompagnd'un termegnral.Autrement,
d'aucunemanire.
l'infrence
ne se justifierait
commeon peutle vrifier,
individuelles constantes
Remarquonsd'autrepartque la clauserestreignant
les ne nommerqu'un objet existantprsentement
obligebien utiliserun
(cf.note 16) dj
que nous recommandions
systmede logiquepragmatique,
logique.
pourla "premire"
54
23:17:17 PM
Telles sont les deux logiques que nous pensons ncessaires l'intelligence des doctrines se rapportant aux infrences de propositions
singulires.
Nous concluerons par quelques rflexionssur la supposition vide
dans son rapport avec la thorie des modes de la supposition personnelle. Il apparat la lumirede notretude que le concept de supposition vide est tranger la doctrinede la supposition personnelledans
son ensemble. En effetla suppositionpersonnelled'un termegeneralse
subdivise en trois modes principaux,identifisjustement par diverses
formesde "descentes". Mais non seulement ces descentes ne sont dfinies que pour des termes non vides, ce qui est vident la seule
lecturedes textes (SL i, 70), mais encoreles termesvides appartiennent
d'autres formespropositionnelleset une autre logique. Il aurait
donc t impossible Ockham d'assouplir la thorie de ces modes de
faon y inclure la possibilit de la supposition vide et c'est aller
l'encontrede sa logique que de chercher le faire.18On comprendque
dans les chapitres consacrs la supposition personnelle,Ockham ne
mentionnequ'une fois la supposition vide, et ceci seulement dans une
rponse une objection.19
Paris
CNRS
Rfrences
SL Summatotiuslogicae,de Guillaumed'Ockham,Editionsde l'institut
franciscain de l'Universit
de St. Bonaventure,
N.Y. 1974.
andGenerality
Press1962.
, CornellUniversity
[i] Geach,Peter,Reference
in Ockham,
Nos,March
[2] Matthews,Gareth,Suppostiloand quantification
I973[3] Price,Robert,WilliamofOckhamand SuppositioPersonalisin: Franciscan
Studies,30 (1970).
in: SemanticsofNaturalLanguage
,
[4] Kripke,Saul, Namingand necessity,
Davidson & Harman,ed., Reidei,1972.
in: Philosophical
ofGottlobFrege,Geach
[5] Frege,Gottlob,Negation,
Writings
& Black,ed., Blackwell,1966.
[6] Meyer,Robertand Lambert,Karel, Universally
freelogicand standard
in: The Journal
ofSymboliclogic,vol.33,Number1,
quantification
theory,
March1968.
andintensional
Richard,Pragmatics
[7] Montague,
ofNatural
logic,in: Semantics
, 1972.
Language
[8] Ockham,Guillaume,
ExpositioAurea,GreggPress,republi,1964.
s theory
of NotreDame Press
[9] Loux, Michael,Ockham*
ofterms,
University
1974.
18Commele faitLoux [9], p. 34.
19SL i, 72, 1.113SS.
55
23:17:17 PM
Vivarium
, XVI, i (1978)
Mental Verbs in Terminist Logic
( John Buridan, Albert of Saxony, Marsilius oflnghen)
E. P. BOS
Introduction; theproblem
statementscan be describedand analysed scientificallyin
different
ways.1 In the Middle Ages the study of language was
Sentential
conducted by grammarians and logicians. The first were concerned with the description and analysis of sentences: in grammar,
theories were formedabout the correctconstructionof words in sentences. The logicians,on the otherhand, studied the formalstructureof
descriptivesentences or propositionsas well as their constituentparts
to determinethe truthvalue of thesepropositions.2In medieval philosophy semantics,the theoryofmeaning,was closelyconnectedwithlogic.
This essay deals with the field of medieval logic and semantics,
especially that part which is usually designated as terministlogic or
logica modernorum.It was here that, among other things,the theory
of supposition- which originatedin the theoryof fallaciesand twelfthcenturygrammar was developed.3
The theory of supposition is a theory of the use of terms. In fourteenth-centurylogic (the period of our inquiry) supposition of terms
occurs in propositions.4For an example of a definitionof supposition
we may take the one given by the fourteenth-century
logician John
Buridan, who says:5
i
1 Jan Pinborg,Logikund Semantikim Mittelalter

Mit einem
. Ein Ueberblick.
Canstatt1972,p. I02f.
Nachwortvon HelmutKohlenberger.
Stuttgart-Bad
2 Cf.,e.g., W. V. Quine,The WaysofParadoxand OtherEssays, revisedand
ofsuch
For a discussion
enlargededition,Cambridge(USA) 1976;esp. p. 25ff.
and
Sentences
notionsas proposition
, see: R. M. Gale,Propositions,
Judgements,
London-New
in: P. Edwardsed., The Encyclopediaof Philosophy,
Statements,
York 1967.
3 L. M. de Rij, Logica Modernorum.
A Contribution
to theHistoryof Early
Terministic
Logic,2 Vols.,Assen1962-7.
"
4 L. M. de Rijk, The Development
of "Suppositionaturalis in MedievalLogic
(I and II), in: Vivarium,9 (1971)P-91-107;11 (x973)P-43"79>especiallyp.
37ff.ofthesecondpart.
6 JohannesBundanus,Sophismata;I chose:MS Erfurt,
Ampi,r 302,i. i2VD.
56
23:17:29 PM
Est autemsuppositio,
prouthic sumitur,
acceptioterminiin propositione
vel quibusdemonstratis
proaliquo vel pro aliquibusquo demonstrate
per
istapronomina
'hoc' vel 'hic' vel equivalentiaisteterminus
vereaffirmareturde istopronomine
mediantecopula.Verbigratia,in hac propositione
'equuscurriisteterminus
'equus' supponitproomniequo qui est,quia de
verumessetdicere'hoc est equus'.
quocumquedemonstrato
Supposition theory firstand foremostis a theory of the reference
of termsto individual things; the propositionsare analysed primarily
as to whetherthe constituentterms referto individual things in the
external world, in terms of which they are verifiable. Fourteenthcentury logic is directed extensionally.
In this paper I intend to present a survey and analysis of the
behaviour of so-called intensionalverbs and of the termsthat are constructedwith these verbs. I mean verbs like 'inteiligere','cognoscere',
'appetere', 'promittere',i.e. verbs signifyinga mental act. It appears,
in medieval as well as in modernlogic,7that the referenceof the terms
constructedwith intensional verbs, is oftennot very transparent.8
The purpose of this essay is to demonstrate the followingpoints:
1.1 With the fourteenth-century
logicians John Buridan, Albert of
and
of
Marsilius
Saxony
Inghen (whose theories are closely related),
the behaviour of mental verbs in assertorialpropositions(of the form:
subject-predicate)is twofold:the termsthat are constructedwith these
verbs either have a referencethat can be called 'omnitemporaT9 or
have indirect or opaque 10 reference.
1.2 There is a similar opaque referencein propositionsin which the
mental verbs are propositionedoperatorswith respect to the content of
a proposition,the dictumpropositions. I mean propositions like 'I
know that etc/ T promise that etc/. In the case of propositionsthat
can be called 'modal' in the broad sense of the word,11verbs that are
not truth-functionalclash with the theory of supposition which aims
at determiningthe truth value of propositions."Whetherthe mental
See also: JohnBuridan,Sophismson Meaningand Truth,translatedand with
an introduction
by TheodoreKermitScott,New York iq66 p. ioo.
* J. Pinborg,op. cit.(cf.notei), p. 127.
7 See section4, p. 60.
8 The term'transparent*
, of course,is Quine's.See e.g., his Wordand Object,
Cambridge
(USA), i960; forthistermand relatednotionsand fortheproblem
as such,esp. ch. 6.
8 See below,4.2,p. 62.
10Cf.note8.
11G. E. Hughesand M. J. Cresswell,
An Introduction
to ModalLogic,London
ofthe 1968edition),preface,
1972(corrected
reprint
p. IX.
57
23:17:29 PM
verbs occur in assertoral propositions or as prepositional operators,

in both cases conversion to the effectthat the term that was constructedwith the verb is now outsidethe scope of the verb, that is, is
in
preposed to the verb, appears to produce correspondingdifficulties,
medieval as. well as in modern logic.
1.3 The medieval logicians' conception of science and knowledge
plays an importantpart in this matter.
This is not the firsttime the problem has been taken up. It is my
intention to give here a more precise and more complete discussion
than has emerged so far from the work of, for instance, Ernest A.
13
Moody,12Theodore K. Scott and Peter T. Geach.14
2
Scope of thesubject
This inquiryis concernedwith the worksof threefourteenth-century
logicians.
2.1 First we will consider John Buridan. le lived from shortly
before 1300 till about 1360 and was active as Master of Arts in Paris
fromabout 1320 till about 1360.15For the present discussion, two of
his works will be considered:
- The Sumrmde de dialctica, composed 1330-1335, especially the
fourthtract, called De suppositionibus.18 In Maria Elena Reina's
edition of the tract on suppositionswe have a text (largelydifferent
fromthat of Peter of Spain, of whom one is reminded,forinstance
in the firstthree tracts and in the sixth17),togetherwith a running
commentaryby Buridan.
- The tract called Sophismata, which may also be regarded as tract
12ErnestA. Moody,Buridananda DilemmaofNominalism,
in: HarryAustryn
WolfsonJubileeVolume,Jeruzalem1965,p. 577-596.
13Introduction
ofBuridan(cf.note5).
oftheSophismata
to histranslation
14PeterT. Geach,A MedievalDiscussionofIntensionality
of
, in: Proceedings
ofScience,
and Philosophy
the International
CongressforLogic,Methodology,
Oxford1972,p. 129-138.
1964.Alsoin: LogicMatters,
Jeruzalem
15J.Pinborg,Prefaceto: ' TheLogicofJohnBuridan*.Actsofthe3rdEuropean
1975),Copenhagen
Symposionon MedievalLogicand Semantics(Copenhagen,
1976.
1
le GiovanniBuridani,Tractatus
'. Primaedizionea cura di
de Suppositionibus
MariaElena Reina,in: RivistaCriticadi Storiadella Filosofia,12 (i957)>PP175-208and 323-352.It is probablethat a criticaleditionof all the logical
of Buridanwillbe at ourdisposalwithina fewyears (Cf.J.Pinborg,
writings
Preface,(note15)).
17L. M. de Rijk, op. cit.(cf.note4), p. 52, note4. Also J.Pinborg,The SumI: De introductionibus
, in: The Logic .... (cf.note15),p. 72.
mulae,Tractatus
58
23:17:29 PM
IX of the Summule, and was probably written after this work.18

2.2 Secondly, there is Albertof Saxony. He may be regarded as a
fellownominalistof Buridan's. 19In many respectshe can be compared
with him, but there are. also important differences,for instance in
respectofthe suppostilosimplex20His theoriesare not so closelyrelated
to Buridan's as are the works of Marsilius of Inghen. Albert was a
teacher in Paris from 1351 till 1362 (or somewhat later). His main
work on logic is the Perutilis Logica,.21With respect to our inquiry,his
worksSophismata, Insolubilia and Obligationes22have not much to add
to his main work on logic.
Unfortunately,not much has been written about Albert's logic,
despite the fact that Prantl called the Perutilis Logica an independent
compendium of one of the most influentialauthors of the fourteenthcentury,and has a high opinion of him.23
2.3 Thirdly,I will discuss the logical works of Marsilius of Inghen.
Occasionally, but emphatically, Marsilius calls Buridan 'magister
meus' 24 and, in spite of interesting differencesbetween the two
logicians, Marsilius was greatly influenced by Buridan.25 Marsilius
26
composed his Tracts on the Propertiesof Terms between 1362 and
1368. If one considersthe numberof manuscriptsof his works on logic
and of the commentarieson them, one must inferthat Marsilius' work
was highlyinfluentialin the German-speakingcountries.
So much for the works which deal with our problem. They are
18E.A.Moody, Jean Buridan, in: Studiesin MedievalPhilosophy,Science,
and Logic; CollectedPapers 1933-1969,Los Angeles1975,p. 444. Originally
vol. II, 1970.
printedin: Dictionary
ofScientific
Biography,
19GeorgHeidingsfelder,
Albert
vonSachsen.SeinLebensgang
undseinKommentarzur Nicomachischen
Ethikdes Aristoteles,
in: Beitrgezur Geschichteder
Texteund Untersuchungen,
22, 3-4Mnich1927,
Philosophiedes Mittelalters.
p. 91; L. M. de Rijk,op. cit.(cf.note4), p. 63.
20I will discussthisproblemin my
thesis-:Marsiliusof Inghen
,
forthcoming
Tractson theProperties
and
of Terms.A CriticalEditionwitha Commentary
Indexes.
21Venice1522;reprint1974,Meisenheim/Glan.
22Paris1502;reprint1975,Darmstadt.See
esp.bookIII, soph.32-35.
23C. Prantl,Geschichte
derLogikim Abendlande,
Leipzig1870,IV, pp. 60-88.
24G. Ritter,StudienzurSptscholastik
I. MarsiliusvonInghenunddie OkkamistischeSchulein Deutschland,
Heidelberg1921,in: Sitz. Ber. der Heidelberger
Akad derWiss.,Philos-Histor.
Klasse,(4) 1921,p. 11.
25E. P. Bos, JohnBuridanand Marsilius
in: The
of Inghenon Consequences,
Logicof JohnBuridan(cf.note 15),p. 61-70;E. P. Bos, An Unedited
Sophism
byMarsiliusofInghen:"Homo estbos", in: Vivarium,15 (1977),PP-46-56.
ae Cf.note20. Themanuscript
onwhichmyeditionwillbe basedis: MS Mnich,
CLM 4385,f.45rff.
59
23:17:29 PM
general works of logic, and unlike the special treatises which the
27
English and Italian schools devoted to it' under such titles as 'De
sensu composito et diviso', 'De scire et dubitare', there seem to be no
tracts fromthe Parisian school which take up the problemseparately.
Our problem is dealt with by the three logicians under various headings. This is also the case in postmedieval logic.28
I shall deal with the theoryof the behaviour of mental verbs in its
proper sense, not, for instance, with its applications in the tracts on
consequences.29
Approach to theproblem
My approach to the problemfalls withinthe perspectiveof Medieval
logic. It may be called 'immanent',as definedby Pinborgin his thesis.30
I shall not present a systematic evaluation of the medieval theories
from a modern point of view. Though I consider this last way of
dealing with a problem in the historyof logic to be the only fruitful
one, I shall leave this to others,more competentthan' I.
As far as our problem is concerned, Moody makes suggestionsfor
this kind of approach. Geach critically evaluates Buridan's position
concerning intensionality, without, however, making a complete
systematic examination.
3
The problem
In the tracts of the threelogicians a group of verbs characterisedas

verbs signifyingan 'intellectiveact of the soul' 31played a part, that
is: verbs like 'intelligere', 'appetere', 'promittere' etc. In the fourth
chapterofhis SophismataBuridan (whiledealingwithappellation)gives
an enumerationof these verbs, as well as the relation between them:
27 'Tractatusde sensucompositoet diviso': e.g. Williamof Heytesbury,
Paul
of Pergola,Paul of Venice,AlexanderSermoneta,RichardBillinghamet al.
RichardLaverham,
'Tractatusde scireet dubitare': Williamof Heytesbury,
PeterofMantuaetal.
RichardBillingham,
29E. J.Ashworth,I promiseYou a Horse . A SecondProblem
ofMeaningana
and Early Sixteenth
in LateFifteenth
Century
Logic,in: Vivarium,14
Reference
(1976),pp. 62-79,139-155.
29Cf.H. Hubien,Johannis
de consequentiis,
ditioncritique,
BuridaniTractatus
Louvain-Paris1976,(LibriII and IV) ; AlbertofSaxony,Perutilis
Logica,Tract.
Ill; Marsiliusof Inghen,Tractatusduo consequentiarum
(a tracton consequencesin twoparts).
30 J.Pinborg,Die Entwicklung
m: Beitragezur
im Mittelalter,
derSprachtheorie
Vol. XLII, nr.2, 1966,p. 10.
Gesch.derPhilos,u. Theol.des Mittelalters,
81 JohnBuridan,Summule,
ed. Reina,p. 345,. 101-103;Marsiliusof Inghen,
rulei, f.57v.
Appellationes,
60
23:17:29 PM
mediantiad significandum
Etiam notandumest quod nominaimponimus
Ideo etiamistudverbum'significo'
bus rationibus
quibusresintelligimus.
sicud'intelligo'vel 'cognosco'.Ita etiamhoc vertales facitappellationes
bum'appare,et hec verba'scio', 'opinor','puto','credo'etcetera.Postea
sic sequitur
etiam,quia appetitusnostrifiuntnobismediantecognitione,
faciuntnobisista verba 'volo', 'appeto','desiquod similesappellationes
facimusnosdero',etcetera.Adhucetiam,quia sub aliquibusconceptibus
traspromissiones
et obligationes,
et consequitur
quod ista verba'debeo',
faciunthas (hiis
'obligo'etceteraet alii terminiex terminisdescendentes
MS) appellationes.32
Buridan reduced all these verbs to verbs signifyingan act of the
intellect. Nevertheless, there are differencesin the various kinds of
mental verbs, and it is important to note what specific division of
verbs a logician has in mind when he seems to discuss the group as a
whole. This will play an importantpart in my essay.
The three Parisian logicians discussed the mental verbs in their
tracts on:
1)
2)
3)
4)
suppositionaturalis (only in Buridan's work);

suppositioconfusatantum;
appeatio;
ampliatio.
As has been observed before,these verbs occur in two kindsofpropositions, viz. assertorial propositions and modal ones (in the broad
sense of the word). I present examples of both kinds:
'
'
ia) cognoscovenientem;
'
.33
2a) astrologusseit aliqua astra esse super nostrumhemispherium
In ia) and 2a) the mental verbs affectthe subsequent parts of the
proposition,at least the parts that are constructedwith them. In both
kinds of propositions these same object-termscan stand outside the
scope of the verb, viz., precedingthe verb. I mean:
'
'
ib) venientemcognosco;
'
'
2b) aliqua astrologusseit esse super nostrumhemispherium
In 2b) we recognisea propositionin sensus divisus in contradistinction to the one in sensus compositusof 2a).
There is a relation between the various paragraphs in the logical
works that deal with our problem. Two modes can be distinguished
32Erfurt,
MS F 302,f. i68rb(cf.Scott,Ch. 4, remark9).
33Ib., f. i69vb.
61
23:17:29 PM
in the discussion of the behaviour of the mental verbs by the three

logicians: these interpretationsare to be found in the theoriesabout :
1) suppositio naturalis and ampliatio;
2) suppositio confusatantumand appettaiio.
'
4.2 Mental verbsin thetheoryof'suppositionaturalis' and of ampliatio'
4.2.1 First, the theory of natural supposition. It seems best to
quote from Buridan's Summule, first the text and, after that, his
commentary on it.
Communissuppositiosolet dividiin suppositionem
naturalemet in suppositionemaccidentalem.Suppositionaturalisvocatursecundumquam
indifferenter
terminus
supponitpro omnibusproquibuspotestsupponere,
utimur
tampraesentibus
et hacsuppositione
quampraeteritis
quamfuturis,
in scientiisdemonstrativis.
Suppositioaccidentalisvocaturilia secundum
et
vel pro praesentibus
supponitsolumpro praesentibus,
quam terminus
vel propraesentibus
et futuris
secundumexigentiam
verborum
praeteritis,
ut postdicetur.Et hac suppositione
utimurin sermoniet praedicatorum,
bus historialibus,
qua etiammaximeutuntursophistae.34
De Rij gives a survey of the developmentof natural supposition35
and points to the importance of this kind of supposition as a touchstone fora philosopher'spoint of view.36
Probably, John Buridan was the reintroducerof natural supposition
in the fourteenthcentury. He conceives it as omnitemporal(i.e. supposition for all times), as is also clear from the text cited. Buridan
comments that this supposition has not been accepted by many 'moderni'. The text concerns, Buridan continues,verbs like 'inteiligere',
'scire', 'cognoscere', 'opinari', 'significare'and many others. He gives
as example: 'homo qui fuit,intelligitur,ergo homo intellgitu.
One should note that Buridan mentions cognitive verbs that are
constructed with an 'absolute' or 'abstract' term (that is: a term of
the firstcategory of Aristotle). They are not constructedwith a 'connotative' term, i.e. a term that signifiessomething'in addition'. The
differencebetween these sorts of terms is clearly pointed out by
Marsilius of Inghen:37
est sciendumquod terminorum
Pro qua divisionepleniusintelligenda
quiet de rectalinea
substantie
de predicamento
dam suntabsolti,ut termini
ut isti termini'Gherardus','Johannes','homo','animai'
predicamentali;
34Ed. Reina,p. 206,11.231-9.
36Cf.note4.
3e Middeleeuwse
Traditieen vernieuwing.
Assen1977.
wijsbegeerte.
37Appellationes,
f.56'.
62
23:17:29 PM
Alii sunt terminiconnotativi,

etcetera.
ut isti termini'sedens',
et sicde aliis.Quodautemistitermini
suntconnotativi,
'album','coloratum'
me,ut notumest,nontarnen
patet,namsi egosedeam,li 'sedeam'significai
absolute,sed connotatme actumsedendihabere.
In demonstrativescience, Buridan says, use is made of this kind of
supposition: knowledge,as he conceivesit, was about individual things.
The intellect can conceive 'homo' and 'animal' as 'conceptus comthat is, without
plexivus', withouta particulartempus,indifferently,38
the
existence
a
of thing designated by that term.
implying
Buridan is rightin remarkingthat this kind of supposition was not
accepted by many moderni. In fact, Marsilius also omits this supposition in his works on Logic.39
Albert of Saxony does not deal with natural supposition either,but
Scott 40 correctlyremarksthat Albert does mention 'materia naturalis' : a proposition in 'materia naturalis' is an analytic proposition.41
4.2.2 In a correspondingmanner the mental verbs influencethe
termsthat are constructedwith themin the theoryof ampliatio. In the
words of Marsilius,who is fondof discussionsabout definitionand who
complains about the lack of definitionof ampliation:
Ampliatioest suppositioterminipro suis significatis
respectudiversorum
indifferenter.
temporum
Marsilius proves, correctly,that also when defined as a kind of
acceptio, as it is with Buridan, in fact ampliation is a case of supposition.42
In the Summulae, we find as text:
futuraet possibiliasi construatur
Quinto,terminus
ampliaturad praeterita,
cumverbosignificante
actumanimaeintellectivae,
sivea parteantesivea
partepost.48
The examples are : 'hominemintelligo',or 'intelligorosam', a construction of an absolute term and a cognitiveverb. Even if thereis no rose
58De Rijk doesnotdeal withthistractin his
'Development
(note4).
He mentions
naturalsupposition
in hisCommentary
on The Sentences,
book
III, 9.13,f.447ra,buthe doesnotacceptit.
40T. K. Scott, JohnBuridan on the
Science,in:
Objectionof Demonstrative
Speculum,40 (1965),654-673;esp. p. 670.
41F. 23ra(Book III, Ch. 10). Cf.Petrus
ed. De Rijk,p. 7,
Hispanus,Tractatus,
.4-s.
42Marsiliusof Inghen,Ampliationes,
f. 5ir.
43Ed. Reina,p. 349,11.33-5.
63
23:17:29 PM
at all, this is trae: 'intelligorosam'. The conditionis that in 'rosa est'

supposition is possible, and this in contradistinctionto 'chymera est',
says Marsilus. Marsilius goes furtherinto this particular problem,
Buridan does not. In this way, Marsilius says, there is ampliation in
'vacuum intelligitur',because, though 'vacuum' does not exist, it could
exist. The solution of this problem is, of course, only possible if one
accepts 'possible' as a 'tempus' in addition to the traditional'tempora'
'past', 'present' and 'future'. Marsilius will only accept this, he says,
if the word 'tempus' is not used in the strictsense. In the doctrine of
natural supposition, this 'tempus' is not taken into considerationby
Buridan, an inconsistency,I feel.
Albert of Saxony gives a correspondingdefinitionand corresponding
examples in his rule VIII. He says:
sinedifferentia
aliquatemporis,
(...) et ratiohuiusquia respotestintelligi
videlicetabstracteab omniloco et tempore.Et tunc,quandoressic intellique fuitvel eritvel potestessesicutresque
gitur,ita beneresintelligitur
in mentemea a quo sumitur
est. Unde si habeo conceptumcommunem
hoc nomen'homo',indifferenter
preintelligoomneshomines,presentes,
tritoset futuros.44
Buridan is now entitled to conclude that this is true: 'hominem
intelligo,si Aristotelemvel Antichristumintelligo'. A comparison can
be made with the theoryof natural supposition,whereit is said: 'homo
qui fuit,intelligitur,ergo homo intelligitur'.
The relation between this kind of ampliation and the concept of
knowledge is clear. There is, of course, in this particular case, no differencewhetherthe same substantive noun is in subject or predicate
position, though in other parts of ampliation theory this is a fundamental issue. Knowledge is not made dependent on existence according to this theory.
4.3 Mental verbs in the theoryof 'suppositio confusa tantum' and
'appellaiio'
The foregoingtopic has not been the subject ofintensestudy,mainly
because the text of the Summule of Buridan, as edited by Miss Reina,
was not very widely used. The problems in the ensuing paragraphs
have more oftenbeen the subject of investigation.
The theoryof suppositionis a theoryof reference.However, mental
verbs cause the terms constructedwith them to behave differently;
this is the case in this section of terministlogic, viz. in the theoryof
merely confused supposition and of appellation. In the works of our
44Perutilislogica,Tr. II, Ch. X, r. 8 (f.I5vb).
64
23:17:29 PM
threelogicians appellation is also a case of supposition (except forthe

case of 'chymera' etc., i.e. names of fictionand other terms45).
As has been said before,one should carefullynote the examples that
were given as illustrationof the rules.
4.3.1
tion:
First the definitionby Buridan of merely confused supposi-
Sed suppositioconfusatantumest secundumquam non sequituraliquod

seorsumretentisaliis in propositione
singularium
positis,nec sequuntur
licet forte
disiunctivam,
singulariadisiunctivesecundumpropositionem
de disiunctoextremo.46
sequantursecundumpropositionem
'Suppositio confusa tantum' is differentfrom*'suppositio determinata', wherethe termsstands fora definite'suppositum'.47The relation
can be expressed as follows: fromdeterminatesuppositioninferenceto
merelyconfusedsupposition is valid, not vice versa.
There is a clash when mental verbs are introduced:
et multaalia,de quibus
Quandomodoistaverba'scio','cognosco','intelligo'
sine
et eorumparticipiaconfundunt
magisdictumest in Sophismatibus,
distributione
terminossequentesterminantes
transitus(trnsitos,
Reina)
non sequitur'ergo cognosco
ipsorum,ut si dico 'cognoscotriangulum',
isoscelem'vel 'cognoscoisopleurum'
et sic de aliis.48
Buridan comments that there was a considerable differencefrom
other cases of merelyconfusedsupposition:
First, no descent is possible to individual things within the reach
of the universal term because the mental verbs cause the universal
term that followsthe verb 'appellare suas rationes, scilicet secundum
quas impositi sunt ad significandum'.49The reason is, says Buridan,
that these nouns in the accusative case participate in 'suppositionem
materialem', which covers, according to him and Marsilius, also the
'suppositio simplex'.50
51
Secondly, the conversion froma merelyconfusedtermto a determinate one is valid: 'cognosco triangulum,ergo triangulumcognosco'.
46JohnBuridan,Summule,
p. 184-5(ed. Reina); Marsiliusof Inghen,Appellationes
, MS CLM 4385,f.57r.
p. 324,11.373-6(ed. Reina).
47p. 323,11.335-40(ed. Reina).
48P-383.783-9(ed. Reina).
p. 334,. 002-3(ea. ).
60p. 201,1.34f.(ed. Reina). Albertof Saxonydoes acceptsimplesupposition.
For thedevelopment
ofsuppositio
thesis.
see myforthcoming
simplex,
51Reina: 'consequentia'.
65
23:17:29 PM
This was not permittedin other cases of merelyconfusedsupposition.

A third differenceis the validity of the conversion from a merely
confusedtermto a distributedone. For instance: 'cognosco triangulum,
ergo omnem triangulum cognosco'. Vice versa, the inferenceis not
valid : one can have knowledge of every triangle,but not by virtue of
the notion of the species of triangle.Neitheris valid: 'cognosco omnem
triangulum,ergo cognosco isoscelem'. One does not necessarily have
knowledge of a specifickind of triangle.
It is quite remarkable that Albert of Saxony, under the heading of
merely confused supposition, gives the followingexamples in connection with mental verbs: 'appetere', 'desiderare', 'promittere','teneor',
'debeo'. He denies this inference: 'promittotibi denarum,ergo denarum tibi promitto',52which was valid according to Buridan's rule.
Buridan, however,gives examples of purely cognitiveverbs.
As
Geach rightlyjudges this conversion to be full of difficulties.53
counter-exampleshe gives the inferencesthat Albert gives, and which
in fact Albert himselfconsiders to be invalid. Conversion is possible
in propositionswith purely cognitive verbs like 'intelligere',especially
if it is conceded, as is Buridan's opinion, that knowledgeis not about
the ratio as such, but that the ratiois linked up with particular things.
This is Buridan's anti-Platonic attitude in this matter.
Afterconversion,the term precedingthe verb has determinatesupposition. Albert of Saxony remarks that equivalent terms have a
correspondinginfluence,e.g. 'necesse', another modal operator. With
this operatorthe propositionsare constructedeitherin sensu composito
or in sensu diviso. Albert sketches the relationbetween the two kinds
of propositions:
terminos
hishabentvimconfundendi
Sic enimtermini
sequenequivalentes
tes se confusetantum,sicud est ly 'necesse',ut dicendo'necesseest me
habereoculum',si debeovidere:ly 'oculum'supponitconfusetantum.Et
: necesseestmehabereoculum,si debeo
propterhocnonvaletconsequentia
enima
videre,ergooculumnecesseestmehabere,si debeovidere.Arguitur
stanteconfusetantumad eundemstantemdeterminate.64
termino
4.3.2 We meet with similar problemsin the theoryof 'appellatio'.
Right in the beginning of his tract, Marsilius speaks of two kinds of
appellation. This division takes into account the specificbehaviour of
mental verbs:
" PerutilisLoeica.TractII. Ch. VII. rule..f. i3vb.
13Cf.note14,p. I33ff.
54Perut.Log.,Tr. II, VII, f. 14.
66
23:17:29 PM
formali(
Est enimaliquandoacceptioprosuo significato
), aliquando
estacceptiotermini
significat.65
prorationesecundumquam talisterminus
In the second case we find appellation of a concept, that is not an
individual thing,its primarysignificate.
Buridan does not define appellation so elaborately as Marsilius.
However, he does underlinethat there is a strikingdifferencebetween
terms constructed with mental verbs and other kinds of verbs. He
gives an exposition of the behaviour of mental verbs in his discussion
of the possibility of substitutionof universal terms constructedwith
them: if the universal termfollowsthe verb, thereis only substitution
of synonymousterms; if the termis preposed, i.e. is outside the scope
of the verb,substitutionof termsis also possible if only the supposition
is the same.55In this last case, the ratio does not matter.57
Buridan, as is evident, acknowledges the validity of this inference:
'cognosco venientem,ergo venientem cognosco' : knowledge is linked
up with particularsin Buridan's opinion.58
Mental verbs, Buridan continues, occur with concepts in various
ways: 'intelligere',forinstance, is about thingsby way ofnon-complex
concepts; 'scire' and other verbs, on the other hand, are about reality
by way of complex concepts as well. In this last case, the proposition
is eitherin sensu composito
, or in sensu diviso. This would imply that
the inferencefroma propositionin the composed sense to one in the
divided sense is not valid. However, Buridan says, Aristotlewas wrong
in concludingthis for all cases, because there is, in fact, a valid inference in the case of cognitiveverbs. Buridan gives several examples in
his Sophismata.59All propositions with 'scire', 'opinari' etcetera, are
true in the divided sense, where the term before the verb has determinate supposition, for instance 'quibet sciens hanc conclusionem,
videlicet: omnis triangulushabet tres equivalentes duobus rectis, ysoschelemseit habere tres ngulos equivalentes duobus rectis'.
Knowledge is about the particular.
Marsilius and Albert, on the other hand, findit hard to accept this
theory. Marsilius denies the validity of 'cognosco venientem, ergo
venientem cognosco' : there is no appellation of the ratio in the consequent, he says.60 If one accepts, as Geach does, that knowledge
65Appellationes,
MS CLM 4385,f.5.
66Summule,
p. 345-6(ed. Reina).Also: Moody,op. cit. (cf.note12),p. 584ff.
57Marsilius
ofInghen,Appellationes,
f.6or;JohnBuridan,Sophismata
, f. i68va.
58Also: Moody,op. cit.,p. 585.
59Sophismata
, f. i68ra-i7ivb.
60Appellationes
, Rule VII, f.6.
67
23:17:29 PM
is about particulars, Buridan seems to have the right position.

Albert of Saxony is very reluctantin the presentationof the theory:
he presentsa theoryof otherlogicians, he says. Who they are I cannot
determine. Afterthe cognitive verb, there follows a complex of substantive and nominalized adjective. If it is, forinstance, the case that
you do know Socrates, while you do not know that he is approaching,
this is true: 'Sortem venientem tu cognoscis', but not: 'tu cognoscis
Sortemvenientem'. As to the conversionof thislast proposition,Albert
says nothing.1
To conclude his sophisms on appellation, John Buridan gives the
sophism: 'debeo tibi equum'. There is a close connection with the
earlierdiscussion: the consequence to 'equum tibi debeo' is true,where
'equum' has determinatesupposition.62I do owe you particularhorse,
and not a Platonic ratio. Ethics is saved.
The inferenceby conversion,as mentioned above, is, without any
doubt, not without difficulties.If one accepts that knowledgeis about
particular things,the inferenceis valid in the case of purely cognitive
verbs, as Geach rightlyacknowledges.The analysis of propositionsalso
brings the problem of universais into the discussion.63As counterexamples to Buridan's rules, Geach gives, among other propositions,
some with the verb 'to look for' etc. It must be said that in this case,
the inferenceis not valid, but it should also be added that Buridancircumspectly?- does not give these examples and that Albert denies
the validity of this inferenceexplicitly.
are far fromagreeingin all respectson
The three fellow-nominalists
this problem.
5
-
Conclusion
I hope to have shown that mental verbs, especially 'intelligere'
etc., occur in supposition theoryin a twofoldway: with omnitemporal reference,and, on the other hand, without descent.
Our authors' conception of science plays an importantpart in this.
It is clear that the three fourteenth-century
logicians of our in-
" Perut.Log.,TractII, Ch. XI, f. i6vb.

* F. 872vb. Cf. also: JohnA. Trentman,VincentFerrerand his Fourteenthin: Artslibrauxet philoon a ProblemofIntensionality,
Predecessors
Century
sophieau moyenge, Montraland Paris, 1969,pp. 949-958.I disagreewith
Trentmanin this respectthat he does not treatthe problemof 'debeo tibi
withthe behaviourof mentalverbs
equum' in Buridan'scase in connection
as such.
3 Geach,op. cit.,p. 134.Cf.Moody,op. cit.,p. 579.
68
23:17:29 PM
quiry, who are so closely related in other respects of theirnominalist attitude, did not agree on every point of this problem.
It is also clear that theories about sense and referencedo not
originate with Frege.
Leiden
FilosofiseliInstituid
Witte Singel 71
69
23:17:29 PM
XVI, i (1978)
Vivarium,
Contribution l'histoire des termes 'natura naturans'
et 'natura naturata' jusqu' Spinoza
OLGA WEI JERS
n'ayant pu dire quelle source Spinoza a puis l'expression natura naturans, devenue un concept clef de sa philosophie,
Personne
de mme que son pendant natura naturata, l'histoire de ces
termes mrite d'tre tudie mieux qu'on ne l'a fait jusqu' prsent.
On trouvera ci-dessous quelques prcisions sur l'origine des termes
tudis, suivies d'un tableau de leur utilisation travers les sicles,
et finalementune note sur l'emploi qu'en a fait Spinoza par rapport
ses prdcesseurs.
I. L'origine
Bien que le concept exprim se trouve dj dans la philosophie de
Jean Scot Erigne sous la formenatura crcans et non creata} les expressions natura naturans et natura naturata ne se sont formesque
dans la premire moiti du XIIIe sicle.2 Leur naissance peut tre
localise dans les traductions latines des commentateurs arabes
d'Aristote, en particulier d'Averros. Le verbe naturare apparat en
premierlieu. Si l'on compare les traductionsd'Abumasar par Hermann
de Carinthie et Jean de Sville, on constate qu' la place des mots
elementanset elementatumutiliss par Jean de Sville, Hermann a
? Dans les traduccrit natura (pas encore naturans) et naturatum
tions des commentaires d'Averros sur Aristote Phys. et De clo>
comme l'a fait remarquer H. Siebeck,4 on trouve le verbe naturare
1 Cf. dj August.,De Trin.XV, 1, 1 o Dieu est natura. . . non creatased
creatrix.
* Une priodedanslaquelle,par le truchement
d'Amauryde Bne et David de
Dinant entreautres,Jean Scot connutun intrtrenouvel Paris. Il faut
noterque lorsqu'audbutdu XIIIe s. Ps-Denysl'Aropagitefutde nouveau
ne semblentpas eux non plus avoiremploy
traduiten latin,les traducteurs
les termes.
3 Cf. R. Lemay,Abu Ma'shar and Latin Aristotelianism
in thetwelfth
century,
Beirut1962.
4 berdie Entstehung
der Termininaturanaturansundnaturanaturatadans:
derPhilosophie,
III (1890).
ArchivfrGeschichte
70
23:17:36 PM
et mme le terme natura naturata. Le dernier figure aussi dans la

traduction du commentaire d'Averros attribue Michael Scot.5 Il
est possible que ce fut Michael Scot qui cra la formeactive natura
naturans. C'est du moins dans son Liber introductorius
qu'on la trouve
la
fois
un
relatif
contexte
premire
l'espoir d'engendrer
pour
(dans
et au pouvoir de Dieu de modifierl'infertilitnaturelle): cum Deus
sit natura naturans et ideo superet naturam naturatam (Mnchen
C.L.M. 10268 fi4ira).e
Pour une meilleure comprhension du processus selon lequel ces
termesont pris naissance et de quels mots grecs ils sont la traduction,
je renvoie l'article de Siebeck.7 Il est certain que leur signification
dans ces premiressources - et qui ne changera d'ailleurs pas beaucoup pendant des sicles - est "la nature qui cre" l'opposition
de "la nature qui a t cre". Dj Michael Scot fait le rapprochement Deus - natura naturans, qui va tre rpt par tous ceux qui
conoivent Dieu comme le principe crateur de la nature. Parfois,
cependant, on notera une confusionentrecette significationet l'identification de natura naturans natura universalis,8 qui n'implique
d'ailleurs nullementune pense panthiste.
2. U utilisationavant Spinoza
Le but de rmunrationdes sources qui suit, est de dmontrerque
l'accueil qu'ont trouv les termesconcerns,a t plus ample qu'on ne
l'a cruet qu'ils onten ralitt le sujet d'une traditionconstante.Le tableau qu'en fait.Lucks9 est trs insuffisant,d'abord en ce qui concerne
les XIVe et XVe sicles qui, chez lui, manquent presque compltement,
ensuite parce qu'il omet de noter qu'outre la tradition purementphilosophique les termesont t utiliss par des thologueset des juristes.
a) Le XIII sicle
Un des premierstextes dans lesquels natura naturans est nonce
depuis l'origine dcrite ci-dessus, est le curieux trait De disciplina
5 Cf. l'ditionde F. StuartCrawford
(The MediaevalAcademyof America,
Mass.,1953)p. 187Deindedeclaravit
Cambridge,
quodistudpropter
quodNatura
esseanimain animalibus,
et nonsolumin animalibussed
, videtur
agitnaturata
rebusnaturalibus.
in omnibus
* Cf. L. Thorndike,
MichaelScot,
1965,p. 105.
7 Cf.ci-dessusnote4. Cf.aussi H.London/Edinb.
- naturanaturata
A. Lucks,Naturanaturans
dans: The New Scholasticism,
IX (1935) P- 1-24; pour l'originedes termes
p. 12 sqq.
8 Cf.les citationsde saintThomasp. 72 et cf.p. 78.
9 Cf.ci-dessus
note7.
71
23:17:36 PM
scolarium du Ps-Boce, crit probablement Paris entre 1230 et

1240. Dans un passage contenant les "Questions de Craton" et,
plus particulirement,dans une question sur les comtes, on trouve le
terme dans le sens de Dieu crateur ou le principe crateur: Tercio
an planete prodigialiterrubentescircularitercum ceterispercurrantan
createin suum chaosrevertantur
naturepermissionenaturantis
destructive
10
6
106,
15).
(, p.
L'utilisation par les juristes, dcrite par Brian Tierney,11commence dj avec Accursius, ad Dig . 12, 1, 38 et ad Inst. 3, 16, 6,
rpt par Odofredussuper Digesto Veteriad i, i, 1, 3 et par Hostiensis, Summa Aurea, De consuetudinecol. 81.12
Le plus proche Accursiusest le passage dans la grande encyclopdie
de Vincent de Beauvais: Speculum doctrinale15, 4 (d. Venise 1591
fo 257ra), o un auteur non identifiest cit (Auctor): In summa vero
notaquod naturaprimodiciturdupliciter.Uno modonaturanaturans,idest
ipsa summa lex naturae quae Deus est ... Aliter vero dicitur natura
naturata, et haec multipliciter.
Parmi les philosophes du XIIIe sicle figurentles deux grands,
saint Bonaventure et saint Thomas:
Saint Bonaventure, in Sent. Petri Lombardi III dist. 8 dub. 2 (Opera
omnia III, Quaracchi 1887, p. 197) non vultdicerequod generatioFilii
sit supra nturmaeternamquae estnatura naturans, sed super naturam
creatamquae consuevitdici natura naturata.
Saint Thomas, Summa theol. I, II, qu. 85, art. 6 (Turin-Rome 1948,
p. 397) Natura vero universalis est virtus activa in aliquo universali
principio naturae, puta in aliquo caelestium corporum; vel alicuius
superioris substantiaesecundumquod etiam Deus a quibusdam dicitur
natura naturans.
id., De div. nom. IV, 21 (Turin-Rome 1950, p. 206) Est autemDeus
universaliscausa omniumquae naturaliterfiunt; unde et quidam Ipsum
nominantnturmnaturantem.
Notons cependant que ce n'est pas saint Thomas lui-mme qui appelle Dieu natura naturans, mais qu'il note seulement cet usage chez
"quelqu'uns" (quidam).
10Cf.Ps-Boce,De disc,scoi.,d. Olga Weijers,Leiden/Kln

1976.
11BrianTierney,Naturaid estDeus: a caseofjuristicpantheism
? dans: Journal
oftheHistoryof Ideas, XXIV, 3 (juillet-sept.
1963)p. 307-22.
12Cf.Tierneyp. 318.
72
23:17:36 PM
Dans la premiremoiti du XIIIe sicle se situe un manuscritcontenant une "Wissenschaftslehre":13

Mnchen C.L.M. 14460 f 31 Ad secundum dicendum quoniam
iusticia naturalis dupliciterpotestconsiderati. Uno enim modo potest
consideranper comparationemad res de quibus est . . . Alio autem modo
potestconsideratiper comparacionemad naturamnaturantema qua est
tamquamaprimofundatore'etsic intenditPlato in ipsa naturali iusticia.
Dans la deuxime moiti du XIIIe sicle se situent Barthlmy de
Bologne et Pierre de Falco:
Barthlmy de Bologne, Quaest. disp. de fide, qu. 2 Et ideo nota
quod sicut natura naturata et creata descendita natura naturante,id est
a divina natura, tarnenreservavitsibi natura divina quaedam opera, ad
quas facienda non potestattingereista natura creata, ut sunt miracula,
sic etiam lex aeterna reservavitsibi quaedam dictamina et scibilia ad
quae docenda non attingitilla lex naturalis nobis impressa.1*
Pierre de Falco, Quest, disp., qu. IX {ad argumentain contrarium)
ad p. 358, 12-3 Vel dicendumest quod etsi Deus hocposset, tarnennon
decet,quia Deus dedit intellectuialiquid de possibili ita et voluntati,
unde utraque potentia nata est reciperehabitm informantemet perficientem
, natura veronaturans non deficitin necessariis sicut natura
naturata.15(Cf. Arist. de anima III, 9, 432021 sq. r' <p<n(xyjtsto tei
ti tcvvayxaov;saint Thomas d'Aquin,
(agcttjv
[X7)0v
(jltjts7uoXe7cei
.
de
Quaest disp.
potentia 3, 7 ob. 1 natura neque deficitin necessariis
in
abundai
neque
superfluis; id., Summa theol.I, II, 91, 2, 1).
id., qu. XXIII Ad argumenta10 p. 813 Ad ittud: desideriumnaturale
non potestfrustrati, respondeo: verumest quantum est ex parte naturae
naturantis,scilicetDei naturameteantis,quia Deus est causa omniumentiumet bonorumet malorumdeficientium. . . Item,hocverumestexparte
naturae naturatae,scilicetcreatae, quia Deus talem virtutemnaturae indidit, quod agens virtutesua adiva inducit in materia illud ad quod est
in potentia.
b) Le XIV* sicle
Au dbut du XIVe sicle matre Eckhardt, mystique Dominicain,
parle de "ungenaturten nature und genaturten nature Gottes" dans
13Cf.M. Grabmann,
Mittelalterliches
Geistesleben
II p. 197.Bienque Grabmann
ait constatune granderessemblance
entrece manuscrit
et le ms. Ripoll109,
je n'ai pas trouvle passageconcerndans ce derniermanuscrit.
14d. P. MeinolfMckshof
f dans: Beitrge zur Geschichteder Philosophie
undTheologiedes MA, 24, 4 (1940)p. 25, 7-8.
15d. A.-J.Gondras,Louvain/Paris
1968.
73
23:17:36 PM
une discussion de la Trinit et sans rapport avec les expressions

latines.16
Commejuristes du XIVe sicle qui ont utilisles expressionstudies,
il faut mentionnerCynus de Pistoia, au dbut du sicle, Albericus de
Rosate un peu plus tard, et vers 1400 Aegidius de Bellamera.17
Tout au dbut du XIVe s. Raymond Lulle est actif Paris. A de
nombreux endroits dans son uvre il utilise nos deux expressions,et
en plus les participes substantivs naturans, naturatumet le verbe
naturare. On trouve natura naturanset natura naturatadans :
- De efficienteet effectu10 (d. Harada 175, 155-160 p. 278)18
Ad consequentiamistius sequiturquod Deus sit natura naturansfactive
etactive, etnatura naturatasitfacta etpassa, ethocper totamsubstantiam
universi et accidentia eius. Aliter, si natura naturata esset aeterna,
distaret a natura naturante; et destrueretur
finis ultimus naturae in
subiectonaturato; quod est impossibile. Pat et ergoquod natura naturata
est sub natura naturantefacta et passa .
- Liber contradictionis15 (d. Harada 171, 585 p. 153); id. 17 (d.
Harada 171, 640 p. 154); id . 19 (d. Harada 171, 741 p. 156).
En combinaison avec naturans, -atum et naturaredans:
- De consolationeeremitae9 (d. Stegmller I p. 99)19 . . . cum
omnis natura requirat distinctosrelativosreales, scilicet naturans, naturatumet naturare, sine quibus esset vacua et otiosa. Eremita, dixit
Raimundus, differentiaest internoturamet nturm, scilicetquod una
natura est naturans, quae est superius, spiritualis, infinitaet aeterna; et
alia est inferius, quae est naturata, nova et finita, quae natura non
facit se ipsam. Et in natura superioriest naturans, qui de se ipso pro
ducit naturatum, et ex ambobus se ipsis spirant naturare; et nullum
istorumtriumfacit se ipsum.
Les termes naturans, -atum et naturare se trouvent indpendammentdans:
- De scientiaperfectaIII De natura (d. StegmllerI p. 228).
16Cf. H. Denifle,MeisterEckehardsLateinische
und die GrundanSchriften
schauungseinerLehre dans: Archivfr Literaturund Kirchengeschichte,
II (Berlin1886)p. 456; Lucks (voirnote7) p. 7-8.
17Cf.Tierney(voirnote11) p. 319-20.
18 Opera latina 168-177Parisiis anno MCCCXI composita,
d. H. Harada,
Medievalis32); cf.P. O. Keicher,
Turnhout1975 (CorpusChrist
., Continuatio
R. L. und seineStellungzur Arabischen
Philosophiedans: Beitrgezur GeschichtederPhilosophiedes MA,Band VII Heft4-5,Mnster1909p. 55.
19 Operalatina,d. T. Stegmller,
tomeI, Palmae Maiorcarum
1959.
74
23:17:36 PM
- PropterbeneAnteiligere
, diligereetp ossificare13 De divina natura
I
(d. Stegmller p. 194).
- De divina natura 1 De unitate (d. StegmllerI p. 352).
- De medionaturali 12 De contrarietate
(d. StegmllerI p. 215).
- Ars brevisXXIII, 5 Natura estformacui propriecompetit
naturare.
Vers 1315 Pietro d'Abano, professeur de mdecine et sciences
inter
physiques Padoue, crivit dans son Conciliatordifferentiarum
et
medicos
f.
G
Venise
:
natura
,
philosophos
diff.9, 14, 3
(d.
1476)
duplex existit: natura quidem naturans ut primum, sicut quidam theo, et natura naturata quae est principium motus et
logizantesdixerunt
in
eius
est
quo
primoper se et non secundumaccidens,
quietis
La note dans la revue "Damaris" 1865 p. 85 dont parle Siebeck20
et dans laquelle E. Bhmer raconte que "ein grndlicherKenner der
Scholastik" lui a dit qu'Ockham est le premier utiliserles termes
natura naturans et natura naturata, a dj t dnonce par Denifle,21
en mme temps que quelques autres auteurs qui avaient rpt cette
observation. Ockham, si les termes figurentchez lui, est videmment
loin d'tre le premier. On aimerait bien d'ailleurs avoir quelques
prcisions sur l'endroit o ces termes doivent tre cherchs dans
l'oeuvre d'Ockham.
Le mot natura dans une phrase d'Aristote est expliqu ainsi par
l'auteur des Sententiae ex Aristotelecollectae (P.L. 90, col. 1017C) :
"natura dirigiturab agente infallibili" (1Comment
. XII Metaph.) inde
natura
naturante
scilicet
Deo
vel
etiam
de natura quidem
,
,
telligitur
naturata, sed ut plurimum duntaxat simpliciterintelligiturde natura
naturante
, scilicetDeo.
Vers la fin du sicle, un philosophe nerlandais, Marsiliusd'Inghen,
utilise le terme natura naturans (en y ajoutant la prcision omnia)
dans ses Abbreviationessuper octolibrosphysicorumAristotelis(Venise
1521) f 3 RA 13 Potestautemnatura capi tripiiciter. . . uno modopro
natura omnianaturanteetestDeus ; id. f8 RB 11 prima natura naturans
omnia.
c) Le XVe sicle
Parmi les juristes, dont Antonius de Rosellis, l'usage de nos termes
continue.22
En 1405 Jean Dominici (Giovanni Dominici), dans Lucula noctis,
20Cf.op.cit.(note4)
p. 370.
21Dans: Archiv
frLit. und Kirchengeschichte.
II v. 46.
22Cf.Tierney
(voirnote11) p. 320.
75
23:17:36 PM
tmoigne de la tradition ininterrompue(ch. 34, d. E. Hunt, Notre

Dame, Indiana 1940, p. 275 :) quia philosophiillas propositiones
fatentur
veras secundumcursumnaturenaturateet non iuxta posse ipsius summe
naturenaturantis, cui communiconsensuinfinitampotestatemascribunt.
L'universalit de la distinction, galement au XVe sicle compltementoubli par Lucks,23peut tre dmontrepar les tmoignages
de deux philosophes,italienet nerlandais,tous les deux mortsen 1499.
Franciscus Sanson, Quaestionessuper totumopus de physico auditu
Aristotilis secundum Aristotilis,Averrois et Scoti doctrinam(Venise
1496) II qu. 5 et VIII qu. 2.24
Lambertus de Monte, Circa octo libros physicorumAristotelis(Cologne 1498) 2 p. 55 A 43 quamvis naturalia aut natura naturata non
cognoscatfinem.
En Pologne aussi, les termes n'taient pas inconnus:
Ms. Cracovie, Bibl. Jag. 513 f 44v Primo namque dixi quod in verbis
premissis tangituromniumrerumconditoret auctorin hoc quod dicitur
" natura
", natura enim communi dividitur divisione in naturam
naturantemet noturam naturatam. Primam quidem naturam omnium
esse conditoremet auctorem,si secundumfidem loquimur, non constat
cum hec omnia fidei rudimentoenotescere
esse immanifestum,
faciunt et
ostendunt.2*
De l'autre ct de l'Europe un ms. anglais, Cambridge, Gonville
and Caius 182/215,du XVe s., explique la distinction: (p. 157) Natura
est duplex, scilicetnatura naturanset natura naturata. Natura naturans
est prima causa , ut Deus ; natura naturata est duplex, scilicetmateria,
forma, et componiturex hiis.2
Par suite du fait que les termes figurentchez trois thologuesnerlandais du XVe s.,27on peut tranquillementsupposer qu'ils taient un
bien commun l'poque. Cette supposition est confirmeen l'occurrence dans un glossaire, le Vocabularius ex quo (Zwolle 1479) : motora
naturans est deus qui creavit naturam et natura naturata est
quelibetres a deo creata de similibus similia procreans.
Cf.. p. 71.
24J'ai trouvles citationsdansl'articlede Siebeck(voirnote4) p. 371 note7,
auquelje renvoie.
25Cf. P. Czartoryski,Wczesna Recepcja "Politykx Arystotelesa
na UmKrakw1963) AneksIV, Teksty
mer
sytede Krakowskim
(Wroeiaw-Warszawai, p. 184.
26Ce manuscrit
m'a t indiqupar L. M. de Rijk.
27Denys le Chartreux,
De dignitate
et laudibusB. V. Mariae 2, 2; Johannes
Liidwine,d. A. de Meijer(Groningue
1963)3, 10
Brugman,Vitaalmevirginis
De bonitate
divina(d. Cologne1534)3, 3, 1 p. 245 28.
p. 152,6; Blomevenna,
76
23:17:36 PM
Tirons en la conclusion que pendant tout le moyen ge il y a eu une

tradition constante de l'utilisation des expressions concernes.
d) Les XVIe et XVIIe sicles
Outre les juristes qui continuent la tradition,28il semble que les
termes soient tombs en dsutude, si Ton veut en croire Lucks,29
depuis la fin du moyen ge. Il est vrai, au moins, que les humanistes
commencent critiquer les expressions pour leur forme: elles sont
considres comme des barbarismes, comme tant d'autres termes
mdivaux. Ainsi fulminedj le Nerlandais Alexander Hegius (env.
3:433-98)dans son Invectivacontramodossignificandi31-2 : pretermisso
itaque barbarissimoilio operequod ScholariumDisciplina <inscribitur>
. . . quoi Boetium scripsisse mentiuntur.Qui libellus tam ineptis tropis
indignissimus sit quo pueri instituantur.Huic
refertusest <ut>
debemusquod dicimus naturamnaturantemet naturatami
Et aussi Jean de St-Thomas, Phil. Nat . I, P.O. IX, De natura, arte
et violentia,Art. 1, p. 148-9 natura . . . aliquando sumiturpro ipso auctore naturae qui non tam est natura quam origo ipsa principiumque
totiusnaturae et ab aliquibus appellari solet asperiore vocbulo natura
naturans, id est ordinansnaturam(d. Vives, Paris 1883).
Or, non seulement ce philosophe thomiste, mais d'autres aussi
utilisent toujours les termes malgr leur ton barbare. Par exemple
Bernardus Sannig dans son Cursus Philosophicus bas sur Duns
Scotus31et au XVIIe s. D. Attilio Brunaccio dans les Commentairesde
Coimbresur les Physiques d'Aristote.32
A la fin du XVIe s. le philosophe anglais Francis Bacon s'carte de
l'interprtationtraditionnelledes termes: Nov. Org. II, I (d. Fowler,
Oxford 1889) Super datum corpus novam naturam sive novas naturas
generareet superinducereopus et intentioest humanae potentiae. Dat
autemnaturaeformamsive differentiam
veramsive naturamnaturantem
sive fontem emanationis . . . invenire opus et intentio est humanae
scientiae.Ici natura naturans ne se rfrepoint Dieu mais dsigne la
"forme", la cause immanente de la proprit des choses. Ce n'est
28Cf.Tierney(voirnote11) p. 321.
29Cf.Lucks (voirnote7) p. 7.
30 Ed. J. Ysewijn dans: Forum for modem language studies,VII = oct.
I97131Cf.Lucks (voirnote7) p. 6.
32Cf. Lucks (voir note 7) p. 7; M. Guroult,Spinoza I (Hildesheim1968)
une confusion
de
entredeux significations
Appendice13 p. 565, qui dmontre
naturanaturans
; cf.p. 9.
77
23:17:36 PM
pas la thorie orthodoxe, mais ce n'est pas non plus l'interprtation

panthiste de Spinoza, bien qu'on puisse peut-tredire que Bacon la
prpare en utilisant les termes pour une opposition immanente la
nature au lieu de l'opposition entre un principe crateur et la nature
cre. Il n'y a qu'un pas faire: appeler cette cause immanente,
responsable des individus, "Dieu".
C'est en fait l'implication de la philosophie de Giordano Bruno qui
utilise natura naturalis pour Dieu {Della causa, dial. 4, d. Gentile e
Croce, Bari 1907-23, p. 219; d. A. Guzzo - R. Amerio, Milano 1956,
p. 378) et qui, dans son oeuvre, ne fait pas de distinctionentre Dieu
et la cration. Mais il faut avouer, avec Lucks, que ce n'est qu'une
seule rfrenceindirecte.
Deux contemporains,un peu plus gs, de Spinoza utilisentles expressions dans leur sens traditionnel:
Heereboord, philosophe de Leyde (1614-61) dans ses Meletemata(d.
33
Nimgue 1665) CollegiumPhysicum,Disp. 2 Thesis 1 4 p. 66.
Clauberg (1622-65) Opera (Amsterdam 1691) p. 62934 et Paraphrases in R. Descartes Meditationes (1658) p. 327.35 Le dernier passage, comme le dmontreGuroult,rvle une confusionentre natura
naturans = Dieu crateur et natura noturans dans le sens de Nature
universelle.3Cette confusion, qui semble dj indique chez saint
Thomas (voir ci-dessus p. 71 et 72), est due une conceptionqui tend
marquer le caractre analogique de la communaut ontique entre
Dieu et la Nature, mais qui n'implique pas une pense panthiste.
En conclusion on peut dire qu' l'exception de Francis Bacon et
Giordano Bruno, tous les prdcesseurs de Spinoza, en utilisant les
termes tudis, n'ortvoulu qu'indiquer le principe crateur distinct
de la cration, mais en mme temps troitementli elle, et que
parfois on a indiqu avec natura naturans la Nature universellesans
cependant identifiercelle-ci Dieu.
3. Spinoza
Il a t depuis longtemps dmontr que Spinoza utilise les expressions tudies dans un sens qu'ils n'ont jamais eu avant lui, en faisant
d'eux des termes techniques de sa philosophie panthiste. Natura
33Cf.Guroult,op. cit.p. 564.
34Cf.id. ibid.
Cf.id. p. 565.
33Ainsique les Commentaires
de Coimbre
(voirp. 77).
78
23:17:36 PM
naturans est la cause divine (= Dieu) immanente son effet (la

nature cre = natura naturata)*1
Si l'on tudie comment Spinoza a pu utiliser ces termes traditionnels de la scolastique pour sa philosophie panthiste,il faut se rendre
compte qu'ils taient en fin de compte trs aptes cet usage. Exprim de faon simple ce phnomne s'exprime ainsi: natura naturans
et natura naturata sont, en un sens, opposes, parce que la premire
est la cause du procs exprim par naturare, la dernireen est l'effet.
En un autre sens "elles ne font qu'une seule et mme nature, qui
unit en elle la prima causa et le primm causatum, tant ainsi, d'un
ct, naturans,et, de l'autre, naturata. Le concept de Dieu peut apparatrealors comme tant un seul et mme contenu sous deux aspects
diffrentslis entre eux par un procs dynamique" (Guroult,
Spinoza I App. 13 p. 566). "Puisque les termes de natura naturans et
natura naturata introduisent la notion de l'immanence de la cause
divine son effet,on comprend que Spinoza ait pu les utiliser pour
exprimerson concept d'immanence absolue que ces termes n'avaient
jamais signifi"(1id. ibid. p. 567).
L'emploi de naturareet naturari pour dcrirel'unit dynamique du
Crateur et du cr a probablement sa source dans le Commentairede
la Physique d'Aristotepar Averros.38Il n'y a cependant pas de raison
de supposer que Spinoza ait connu ce commentaireet y a trouv cet
emploi. Ni qu'il a d lire saint Thomas, comme le pensent certains.39
Il n'y a qu'un seul endroit o Spinoza parle de l'utilisation des expressionspar d'autres, au ch. VIII de la Partie I du Court Trait (d.
Van Vloten-Land, La Haye 19143) "Gelijk 00k de Thomisten bij
37Pour la signification
des termesdans la philosophiede Spinozaje renvoie
aux tudessuivantes:J.Martineau,
A StudyofSpinoza(Londres1882)p. 224-6.
- H. A. Wolfson, The Philosophyof Spinoza (Cambridge,
Mass., 1934) P- *5
sqq.- F. A. Walsh,compterendudu livrede Wolfsondans: The New Scholasticism,VIII (oct. 1934) P- 352'5-- id., The GodofSpinoza (compterendude
R. McKeon, The Philosophyof Spinoza, New York 1928) dans: The New
III (juillet1929).- H. A. Lucks,op. cit.(voirnote7) p. 10-12.
Scholasticism,
- Il fautnoter
M. Guroult,
op. cit.(voirnote32) p. 345$^. et App. 13 p. 567-8.
du fond
que dans ce que Tierney(p. 319 note35) appelleune tudeperceptive
mdivalde la doctrinede Spinoza,A. O. Lovejoy,TheDialecticofBrunoand
: Philosophy,I, Berkeley
Spinoza dans: Univ. of CaliforniaPublications
1904p. 141-174,on ne trouvesur l'histoireconcrtedes termesqu'un renvoi
Siebeck(voirnote 4). - Pour les endroitso l'on trouveles termeschez
LexiconSpinozanum
Spinoza,cf.E. GiancottoBoscherini,
p. 735-6.
38Cf.Siebeck(voirnote4) p. 372.
39Par exempleH. A. Wolfson,The
Philosophyof Spinoza (voir note 37) 1
p. issqq.
79
23:17:36 PM
hetzelve God verstaan hebben, dogh haare Natura naturans was een
wezen (zy zo noemende) buyten alle zelfstandigheden"("Comme les
Thomistes ont entendu par cela Dieu, mais leur Natura naturalis
tait un tre (ils l'appelaient ainsi) extrieur toutes substances").
Il est assez clair que Spinoza ne vise pas saint Thomas lui-mme,
puisque ce dernier,comme on l'a dj remarqu,40n'a pas appliqu
le terme natura naturans Dieu, mais rapporte seulement que "certains" ont fait ainsi. Lorsque Spinoza parle des Thomistes, il est
vraisembable qu'il pense des adeptes de la doctrinede saint Thomas,
comme par exemple Jean de saint Thomas, qui ont, eux mmes,
identifinatura naturans Dieu.
Cela n'implique pas une dtermination de la source laquelle
Spinoza aurait puis les termes. Dans la phrase cite, Spinoza ne se
rfre qu' l'emploi par les Thomistes, sans parler d'autres, avant
eux, contemporainsou plus tard, qui s'en sont servis. Le but de cet
article est justement de montrerque les termes concerns font l'objet de toute une traditionphilosophique,thologiqueet juridique, qu'ils
taient, depuis le XIIIe s., des expressions courantes dans ces disciplines,et qu'il est absolument inutile de se demander ce que Spinoza
doit avoir lu pour les connatre.
Voorburg (N.L.)
Nicolaos Beetslaan 91
" Cf.p. 72.

80
23:17:36 PM
VivariumXVI, 2 (1978)
On Ancient and Mediaeval Semantics and Metaphysics (2) *
L. M. DE RIJ
The MultiplicationofBeing in Aristotle'sCategories
3.1 Introduction
of the results of the preceding section may be that Lloyd
One (1956, 59) seems to be wrong in asserting that in Plato's view
the rle of the universal is played by the Idea exclusively, and
that only by the time of the Middle Academy, that is, forthe Platonists
of the firsttwo centuriesA.D., the performersof this rle have been
multiplied.As a matter of fact the distinctionbetween Plato and his
followersof the Middle Academy on this score would seem to be a
differentone. The ontological problems of participation were felt as
early as in the Platonic dialogues (see our section 2), as well as the
logical ones concerningpredication (which will be discussed in a later
section). Well, the Platonists of the firsttwo centuriesA.D., introduced
1
explicitlya threefolddistinction of the Platonic Form or rather of
its statuswhich was (only) implied with Plato. I think,Lloyd is hardly
more fortunatein ascribing {ibid.) this introductionchieflyto the influence of Aristotelianlogic on Platonic interpretation.It is true, in
'
stating the basic distinctionbetween en hypokeimeniand kath hypokeimenouAristotletried to face the same clusterof fundamentalproblems which induced later Platonists to the distinction of the Forms
as taken beforeor after the methexis(cf. Simplicius, In Arist. Categ.y
7912ff.).However, Plato's disciple, Aristotle (the most unfaithfulone,
in a sense, as must be acknowledged) was as deeply engaged on the
same problems as were his condisciples and the Master himselfin his
most mature period. It is certainlynot Aristotlewho played the rle
of a catalyst and was the firstto provoke the multiplication of the
* PartI ofthiscontribution
is foundin thisJournal15 (1977),81-110.
1 I cannotsee whyLloydthinkstheMiddle
Platonists,insteadofmakingone
three-fold
to havedrawntwoindependent
distinctions.
For thatmatter,
division,
whensumming
doesspeakofa threeup (61) hisownexplanations
Lloydhimself
folddivision.
81
23:15:40 PM
Platonic Form in order to solve problems which were not recognized

beforein the Platonic circle.On the contrary,Plato himselfhad saddled
his pupils with a basic and most intricateproblem,that of the nature
of participation and logical predication. It was certainlynot left quite
unsolved in the later dialogues, but did still not have a perspicuous
solution which could be accepted in the School as a scholastic one.
So any of his serious followers,(who were teachers in the School, at
the same time) was bound to contrive,at least, a scholastic device to
answer the intricatequestion. To my view, Aristotle'ssolution should
be discussed in this framework.For that matter, Aristotle stands
wholly on ground prepared by his master to the extent that his works
on physic and cosmology,too, are essentiallydiscussions held within
the Academy (Cp. Werner Jaeger, Aristotle.Fundamentals of the
historyof his development,Oxford 1949, 308).
3.2 Aristotle'sclassificationof beingas givenin theCategories
From the logical (semantic) point of view the problem of participation concerns, among other things, the differentmeanings of terms
signifyingboth essential and accidental qualities (including the socalled "true nature" of things). So it seems to be quite natural to
draw the firstbook of the AristotelianOrganon,the Categoriesinto our
discussion. However, a troublesome stumbling-blockis found on the
interpreters'way, viz. the problem of the proper meaning of the
categories.
3.2.1 The commonview: categories= predicates
In his GreekFoundationsofTraditionalLogic Ernst Kapp maintained
that much trouble in the study of philosophy and its development
had been caused by the fact that students seemed to have to begin
their study of Aristotlewith a little book "whichtreatspredicatesas if
they were not predicates" (p. 21; italics mine). On his view this little
book, Aristotle's Categoriesmarked an incisive turn in the historyof
thought,which was practically unavoidable at the same time: 'predicate' (Greek: katgoria,transliteratedcategory)was going to be used
in the sense of 'class of predicates' and, accordingly,as 'class of things'
signified by such predicates. The (unfortunate) next step (still on
Kapp's interpretation)simply was to isolate the 'meanings' involved
and to discuss them as meanings of words not combined into statements, i.e. taking the terms on their own, quite apart from any
predicative relationship(see Kapp, 25-27).
82
23:15:40 PM
Kapp thinksto findsome support to his view in the pivotal position

of argumentationin Aristotle's logic. As is well-known,the arrangementof the subject-matterof the Aristoteliansystemoflogic: (i) term,
(2) statement,(3) syllogismas suggested by the orderof the firstthree
books of the Organon (Categoriae, De interpretatione
and Prior Analytics) is not the one practiced by Aristotlehimself: the two Analytics,
Topics and SophisticiElenchi concern all of them just argumentation
to the effectthat they do not presuppose any separate treatmentof
the ingredientsof argumentation: statement and term. It is highly
remarkable,indeed, that the scholastic division into the logic of term,
of proposition,and of argumentation,which as early as the Hellenistic
periodwas reduced to the "Aristotelian" distinctionofthe threemental
operations of conceiving, stating and reasoning does not find much
support in Aristotle'sown practice as a logician.
The openingpart of the Prior Analyticsseems to be most instructive
on that score. It contains the well-knowndefinitionof term (horos) :
"I call that a term into which the premise is resolved, i.e. both the
predicate and that of which it is predicated" (24 b 16-17). The horos
(Latin terminus)has nothingto do really with 'concept' or 'meaning'
(in the semantic sense) but only concerns the "empty place" in the
implicationformula (later on, the inferencescheme); it is filled in by
Aristotlehimselfwith symbols (letters),to the effect,indeed, that the
letters may stand for anythingthat can occur either as subject or as
predicate of a statement,rather than stand for a meaningfulterm as
such. As a matterof fact,Mediaeval logicians were in a constant habit
of stressingthe most specificsense of 'term'as occurringin the opening
part of the Prior Analytics; unlike modernpeople Ancientand Mediaeval logicians were fully aware of the etymology of the word 'horos'
(terminus').Kapp could rightlypoint (29) to the correctmeaning of
this 'horos' and its disastruous confusionwith the metaphysical sense
and usage, which lies at the basis of Prantl's obstinate rejection of
Ancient and Mediaeval formal systems of logic. So Kapp is of the
opinion (see op. cit., p. 39) that, in fact, the categories,far fromtaken
(as they should be, on his view) as classes of predicates, are usually
knownin a functionthat is almost contradictoryto theiroriginalconstitution,i.e. not as a due and timely attempt to differentiate,for
certain dialectical and philosophical purposes, the sometimes misleading uniformityof grammatical predication, but as an arbitrary
and superficialsurvey of classes or general headings to whichAristotle
tried to reduce all the objects of our thought.
83
23:15:40 PM
Several points can be made against Kapp's view. First. The (unfortunate)arrangementas contrastedwith the Prior Analyticspassage
proves absolutely nothing in favour of his view of the categories as
'predicates', i.e. as basically elements of sentence and argumentation.
The only point he made is in evidencing the rather loose connection
(if any at all) of the firsttwo works of the Organon with the other
books.
Anotherobjection can be put forwardagainst Kapp's sharp distinction between the logical and the ontological(metaphysical)areas. Even
if thereshould be assumed in the historyof logic (includingAristotle's
own development) a transitionfroma logico-grammaticalto an ontological use of the categories,it would be more conspicuous and illegitimate to modern than to Ancient eyes. As a matter of fact, the sharp
distinctionbetween the ontological and the logical aspects is due to a
modern way of thinking.2Thereforethere is a great deal of confusion
in Kapp's speaking (39) of Aristotle's "abandonment of his original
point of view" which he thinks to perceive in the (otherwise undeniable) fact that in various writingsAristotle uses a more or less
complete enumerationof the ten 'categories'as a convenientlycompendious inventoryof the main aspects of reality.
However, some far more fundamentalobjections must be adduced
against Kapp's most unfortunateview of the Aristoteliancategories,
their original functionand true nature, which he supposed to have
been so drastically misunderstood in the history of logic. For that
matter, Kapp presented the common view of the proper meaning of
the categories in its most extreme fashion, that is, with an absolute
neglect of the semantic purport of the doctrineof the categories.
The problem at issue deserves a thoroughinvestigation.I start with
a closer inspection (our sections 3.2.2.-3.2.7) of what Aristotle himself
has to say about his classification of 'being' ("things there are", he
status
says, i a 20) ; some special attentionwill be paid to the different
the
the
nature
of
involved
of the 'things'
categorieswill
(3.2.6). Next,
be clarifiedto some extent by contrastingthemto the predicables (3.3).
A further discussion will be found in our section 4, where I shall
make an attempt at corroboratingtheircharacterizationas classes of
names, ratherthan predicates, sketched roughlyin section 3.3 (below,
p. ioif.).
2 Cf. L. M. de Rij, The Categoriesof Being in Aristotle's
Philosophy(diss.
Utrecht)Assen1952,p. 4. Cp. Lloyd,66. See also below,p. 88.
84
23:15:40 PM
3-2.2 The thingssaid 'aneu symploks'

, ch. 1-9 (1 a 1-11 b 8, which seems
Considering,now, the Categories
to be the only part of the tract that can claim to be authentic),3we
have to notice that it contains an extensive and fairlyclear description
ofthe firstfourcategories(Substance, Quantity,Relation, and Quality)
froman undeniably ontologicalpoint of view.
The author startsfromthe assumptionthat thereare "things 4 which
are said" and divides them into two groups: some are said in the
context of a statement,others are said apart from any such composition {aneu symploks,ch. 2, i a 16-17). The latter group is the one
discussed in thistreatise.Its members,designated as 'things'or 'beings'
(ton ontn,i a 20) are subdivided. Some are said of a subject (katK
hypokeimenoutinos), but are not presentin any subject (en hypokeimeni de ouden), such as 'man is said of the individual man as its
subject,5 but is never present in a subject. (The author adds that by
'being present in a subject' he does not mean "present as parts are
in a whole" but "being incapable of existence apart fromthe subject
involved"; 1 a 24-25). Others, again, are presentin a subject but are
notsaid ofany subject. For instance,the individual grammaticalknowledge (i.e. some actual knowledgeofgrammarpossessed by some specific individual) is presentin the mind as its subject (substrate) 6 but is
not said of any subject, such as the individual whitenessis present in
the body as its subject, yet is not said of any subject. Other things,
again, are both said of a subject and present in a subject, such as
knowledge is present in the mind as its subject and is also said of
grammaticalknowledgeas its subject. Finally, thereis a class of things
which are neitherpresentin a subject nor said of a. subject, such as the
individual man or the individual horse. For that matter,to speak more
generally: that which is individual and numericallyone (ta atoma kai
hen arithmoi)is never said of a subject but there is nothingto prevent
some of them frombeing present in a subject, such as an individual
3 Unlikesome30 yearsago
(seemypaperspublishedin Mnemosyne
1951), the
of the first
presentauthorhas his seriousdoubts,now,on the authenticity
treatiseof the Organon
; in viewof the scopeof thisstudy,however,we need
notworry
aboutthequestionofauthenticity,
all theless,indeed,onmyassumptionthattheyounger
whois commonly
considered
thepossibleauthor
Aristotle,
ofthetreatise,
washimself
a discipleofPlato'sand stillbelongedto thePlatonic
circle.
4 The modernterm
foritemsin any
'things'is used hereas a blanket-term
category.
6 manbeingthesubject.
e mindbeingthesubstrate.
85
23:15:40 PM
piece of grammaticalknowledgebelongs to the class of thingsthat are

present in a subject (i a 23-b 9).
From the ontological point of view and, accordingly,fromthat of
participation, the contradistinctionof beingpresentin thesubjectand
not beingpresentin the subject seems to be the most important one.
As a matterof fact it may be (and always was) interpretedas a distinction between non-substance {puh ousia, or accidental being) and substance (ousia), respectively.Lloyd (65) seems to be rightin saying that
the conception of en hypokeimenion was readily assimilated in the
old Platonic circlebecause it was identified(probablyin a correctway)
with that of en alli on, which Plato had regardedin the Timaeus as a
propertyof thingscomingto be (gignomena)as opposed to thingsbeing
{onta).
So the proper sense of what fromthe philosophical point of view is
at stake here can only be gathered after a discussion of the author's
doctrine of substance, given in ch. 5.
3.2.3
The doctrineof substancegiven in theCategories
Substance, in the truest and primaryand most definitesense of the

word, the author says (2 a uff.), is that which is neithersaid of a
subject norpresentin a subject, e.g. the individual man or horse; but
in a secondary sense those things are called substances (which are
called 'secondary substances') within which, as species, the primary
substances (i.e. the substances of the formertype) are included, as also
are those which, as genera, include these species. For instance, the
individuai man is included in the species man, and the genus to which
the species belongs is animal ; that is why they are termed 'secondary
substances'.
Next, the authorcomes to speak about the applicabilityof thisgroup
of the unconnectedterms:
2 a 19-b6 (AckrilVs
translation
7): It is clearfromwhathas beensaid that
is said ofa subjectbothits nameand its definition
ifsomething
are necessarilypredicatedofthesubject.For example,manis said ofa subject,the
individualman,and the name is of coursepredicated(sinceyou willbe
manoftheindividualman),and also thedefinition
ofmanwill
predicating
be predicated
oftheindividualman(sincetheindividual
manis alsoa man).
Thus boththe nameand the definition
willbe predicatedof the subject.
But as forthingswhicharein a subject,in mostcasesneither
thenamenor
7 Aristotle's
andDe interpretations
withnotesand glossary
, translated
Categories
by J.L. Ackrill,Oxford*1974; this workis mostuseful,as still is Tricos
Avistte
. Organon,I Categories,
Nouvelletraduction
et notes,Paris 1i94.I am
muchindebtedto bothofthem.
86
23:15:40 PM
is predicatedof the subject.In somecases thereis nothing

the definition
to preventthe name frombeingpredicatedof the subject,but it is imto be predicated.For example,white,whichis
possibleforthe definition
in a subject(thebody),is predicatedof the subject;fora bodyis called
white.But thedefinition
ofwhitewillneverbe predicatedofthebody.
All theotherthingsare eithersaid to theprimary
substancesas subjects
or in themas subjects.This is clear froman examinationof cases. For
also oftheindividual
example,animalis predicatedof man and therefore
man; forwereit predicatedofnoneoftheindividualmenit wouldnotbe
also in an
predicatedofmanat all. Again,colouris in bodyand therefore
individualbody; forwereit not in someindividualbodyit wouldnotbe
in a bodyat all. Thus all the otherthingsare eithersaid of the primary
substancesas subjectsor in themas subjects.So iftheprimary
substances
did notexistit wouldbe impossible
foranyoftheotherthingsto exist.
Then our author goes on to speak about the ontological differences
between the primary and secondary substances (2 b 7-38). Of secondary substances, he says, the species is more trulysubstance than the
genus since the formeris more closely related to primarysubstance.
As to the primarysubstances, they are most properlycalled substances
in virtue of the fact that they are the entities which underly every, 2 b 15-16), and that
thing else (dia to tois allois hapasin hypokeisthai
else
is
either
said
of
them
ta
alla kaia tontnkateverything
(panta
or
in
them.
the
same
relation
which exists
Now,
goreisthai) present
between primarysubstance and everythingelse is also found between
the species and the genus; thus there is another reason for asserting
that the species is more truly substance than the genus. Again, it is
for good reasons that of all that remains, when we exclude primary
'
substances,we concede to species and genera alone the name secondary
substance', forthese alone of all what is predicated (tonkatgoroumenn,
2 b 31) convey a knowledge of primarysubstance. Of all other predicates that we state, such as ' (is) white, 'runs', and so on, this cannot
be conceded. Further,primarysubstances are most properlyso called
because they are the subjects (substances) of everythingelse (2 b 38;
cp. 2 b 15-16).
Finally, the distinctivepropertiesof substance are discussed (3 a 7b
4 19). First a characteristicthat is commonto primaryand secondary
substance is that it is never present in a subject (the characteristic
mentioned above, p. 86: to m en hypokeimenieinai, 3 a 7-8). As a
matter of fact, secondary substance, too, such as the species man or
the genus animal, is said of a subject, yet not presentin any subject,
forneitherman (nor animal) is present in an individual man, while it
is said of him as a subject. Again, all substance (both primary and
secondary) seems to signifya certain this (tode ti smainein, 3 b 10).
87
23:15:40 PM
This is indisputably true as forprimarysubstance. In respectof secondary substance, however,thisis not reallytrue,our author adds; rather,
it signifies a certain 'being qualified' (mallon poion ti smainei, 3 b
15-16), because the subject is not, as in the case of primarysubstance,
numericallyone but, for instance, man and animal are said of many
things.On the otherhand, secondarysubstance does not simplysignify
a certain 'being qualified', as 'white' does, which signifiesnothingelse
than the qualification involved; genus and species, on the contrary,
mark off the qualified being of a substance: to the effect that they
signifysubstance as differentiatedby some qualification.8The other
characteristics(3 b 24-4 b 19) can be dismissed at present.
3.2.4 The ontologicalcharacterof theclassification
As has been remarkedbefore (p. 86), the conception of en hypokeimeni on found in our tract may be set by the side of the Platonian
1
one meant by the phrase en alli on*Plato made use of in the Timaeus
to signifya propertyof things comingto be {gignomena)as opposed to
things being {onta). It should be noticed, now, that for Plato the
gignomenaare found in preciselythat area where Aristotle's onta have
their locus naturalis.
'
But, unlike the Platonic term 'allori, the word hypokeimenoriis,
fromthe modern point of view, at least, rather equivocal in meaning
both ontological substrate-subjectand logical subject-substrate
; the
formermore explicitly in the phrase 'en hypokeimeni
the
latter
in
,
'
. Lloyd (65) seems to be wrong in assertingthat
kath*hypokeimenori
the formula'en hypokeimeniori reflectsas much a logical as a metaphysical notion of substrate (which seems to be correct), whereas the
'
'
formula kath hypokeimenorishould be a "purely logical phrase".
'
His remark is rather confusing,it seems. How could the phrase en
'
'
'
hypokeimeni reflect a double notion if its counterpart kath hypokeimenoriis "purely logical" ? Elsewhere (p. 154) he seems to be no
more fortunate(and inconsistent,at least) in assertingthat the former
phrase should be taken as metaphysical and concerned with properties, the latter as logical and concernedwith predicates. The decisive
point I would think, is that, by the lack of a sharp distinction in
Ancient philosophybetween the logical and the ontologicalareas, both
formulasare (strictlyspeaking) not "purely" logical, c.q. metaphysical,
8 It shouldbe noticedthatin thiscontextthequalification
is alwaysan essential one.
88
23:15:40 PM
and only differin theirprimarysenses,which,if taken apart fromtheir

secondary ones, are anachronistic (modern) ones, indeed. See also
above, p. 84.
'
ProfessorAckrill is rightin remarking(74-75) that the phrase en
'
hypokeimeni as said of non-substance categories concerns the requirement that every instance of a property being in some kind of
substance should belong to some individual substance of that kind.
For example, it is not white(ness)taken as a species that is present
in some individual man as its subject, say Socrates, but thisinstanceof
whiteis presentin thisindividual man Socrates. That is why Aristotle,
in defining(1 a 24-25) the phrase 'being presentin <say A as > a subject' does not use a formulasuch as 'present in A, and incapable of
existing separately from A' but has 'present in A and incapable of
existingseparatelyfromwhatit is in*, i.e. fromwhatever- be it A or
anothersubject whatsoever- which it actually is in; thereforehe adds
(1 a 28) : "for all colour is in a body".
Ackriirs remark seems to be right to the extent that what is requisite is that every instance of a property should belong to some
individual substance of that kind. However, he seems to be definitely
wrong in concluding: "thus the inherence of a propertyin a kind of
substance is to be analysed in terms of the inherence of individual
instances of the propertyin individual substances of that kind". As
will be argued below (p. 95f.), the inherence of a universal in another
universal seems not to be excluded by our author.
In fact the passus 1 a 20-b 9 offersa fourfolddivision of 'things
thereare' {tnontn), as has been oftennoticed. But the ratiodividendi
is based upon a remarkable(and ratherodd) combinationof the criteria
en hypokeimeniand katK hypokeimenou.I come back to this later on
(p. 90).
The formerphrase, as is well-known,serves to distinguishquantities,
qualities and itemsin otherdependentcategories (in a word: accidents)
fromsubstances, which, unlike accidents, exist independentlyand in
theirown right; the latter,on the contrary,aims at markingoffspecies
and genera (i.e. universais)fromindividuals. This produces fourclasses
of 'things there are', which, fromthe ontological point of view, are
interrelatedin a remarkableway:
(1) species and genera in the category of substance, viz. the so-called
secondary substances, such as man, tree
individuals
in non-substancecategories,viz. the so-called concrete,
(2)
89
23:15:40 PM
or particular,accidents, e.g. thisparticularwhite of thisparticular

wall
(3) species and genera in non-substance categories, viz. the so-called
universal accidents, such as 'whiteness' or 'the white'
in the categoryof substance, viz. the so-called primary
individuals
(4)
substances, e.g. this man, this tree.
Let us have a closer look at all this. First our assertion that the
ratio dividendi should be considered rather remarkable, needs some
explanation as it might raise some confusion.Ackrfflis quite rightin
rejecting(p. 75) the view that 'said of' and 'presentin' should introduce
notions of radically differenttypes, the formet being linguistic or
grammatical,the latter metaphysical or ontologica!; and that, accordingly,the word translated 'subject' (literally,'what underlies') should
mean grammatical (or logical) subject in the phrase 'said of a subje
and 'substrate' in 'presentin a subject'.For that matter,I have already
pointed to the absence of such radical distinctionsand differencesin
Ancient philosophy (above, p. 84; 88f.). However, when defendingour
author ih stressing,with Ackrill,that his fourfolddivision is a classification of things('things there are', 1 a 20) ratherthan names, and that
what is said ofsomethingas subject is itselfa thing(a species or genus),
not a name, one seems to be just shiftingoffthe difficulty:our author's
basing the classificationupon things,not names, should be precisely
consideredthe reason why it lacks clarity.To say it more specifically.
When both in 'whiteis said of and in 'whiteis present in' the subject
term stands for a thing, not a name (as ProfessorAckrill rightlyremarks), therestill is no single warrantthat it stands forthe same thing
each time. As a matter of fact it does not. And that is what seems to
trouble, at firstglance, at least, the ratio dividendi and, accordingly,
the classificationitself.
3.2.5 Some obscuritiesof theclassification
This lack of clarityis on a line with the obscurityof the relationship
between primaryand secondary substances, and that between particular and universal accidents. When speaking (ch. 5, 3 b 10-23) about
substance's characteristicof signifyinga certain 'this', which may not
thought, however, to be indisputably true of secondary substance,
Aristotle still insists that secondary substances are kinds of primary
substance (see above, p. 87i.). Such an assertionseems to encroachupon
the fourfoldcharacter of the classification.The same objection may
90
23:15:40 PM
be raised {mutatismutandis) concerningthe relation of particular and

universal accidents.
For the rest,it cannot be denied (as Ackrill agrees) that sometimes,
(e.g. 2 a iff.) Aristotle does speak of 'saying' or 'predicating' a' name
of a subject, be it true that it still is not linguisticitems but the things'
they signifywhich are 'said' or 'predicated'. However, it is just the
fact that Aristotlesometimes (e.g. at 2 a 31-34) is 'careless' (to speak
withAckrill)in that he says that whiteis in a subject and is predicated
of that subject which should draw our special attention.I would think,
ProfessorAckrill is somewhat inconsistentin more than one respect.
First. Aristotle'ssupposed carelessnessshould fitin very well with the
absence of 'notions of radically differenttypes' Ackrill is quite right
to deny in Aristotle's views of the matter involved. Secondly. When
Ackrill adds that Aristotle should have said (at 2 a 31-34) that white
is in a subject and that its name is predicated of that subject, he is
'
bound to recognize that whiteis in a subject and white'is said of it.
'
'
Well, on this assumption, he should agree, white is a name, not a
'
too, would be a name,
thing,and, accordingly,the kath hypokeimenou,
not a thing and the en hypokeimenikai kath' hypokeimenoua rather
remarkablejuxtaposition.
The best thing we can do, it would seem, is to reject the label
'linguistic'or 'grammatical' in a more radical way than does Professor
Ackrill, and assume, indeed, that Aristotle is always speaking of
'things',either'things' presentin a subject or 'things' said of the subject. It seems to be the only way, forAristotleas well as forhis interpreter,to be consistent.
When doing so, the classificationwill be:
(1) secondary substances, i.e. thingsthat are said of individuals, yet
not are presentin them;
(2) particular accidents, i.e. thingsthat are presentin individuals,yet
not said of them;
(3) universalaccidents,i.e. thingsthat are said of a subject and present
in a subject;
(4) primarysubstances, i.e. thingsthat are neitherpresentin any subject nor said of any subject.
statusof the 'things'meant
3.2.6 The different
The two foci of the problem area covered by this classificationmay
be termed (a) the status of the 'things' mentionedin the fourformulas,
91
23:15:40 PM
and (b) the status of that which theyare in, resp. said of.Lloyd (p. 154)
is of the opinion that, in order to defend the categories,it had been
necessary after Aristotle to insist sometimes that they are concepts,
not things (i.e. real properties), sometimes that they are in re universais, not 'separate'. This might have been the way later Ancient
philosophersconsidered the problemsinvolved. However, it should be
borne in mind that such a wordingof the controversyis only possible
if one opposes somehow clear Aristotelian formulas to well-profiled
Platonic doctrine, e.g.: the categories are sometimes to be taken as
concepts, not things, and if as things,then sometimes as one kind of
thing,sometimes as quite another kind. Well, the author of the Categories (who presumably was the young Aristotle) does certainly not
seem somethinglike a muddle-head. So the historian's task is to find
a generic heading to cover all those kinds of 'things'. Of course, it is
of some use to repeat here what has been said before (p. 85, n. 4), that
the word 'thing'used in this connectionis to be taken as a blanket-term
foritems in any category (cf. Ackrill,71).
3.2.6.1
The firstitemof theclassification
Returningnow to the classificationthe followingpoints can be made.

The secondary substances meant here are 'universal things', i.e. universais in the categoryof substance,which are said ofindividual things,
(primarysubstances), such as man of Socrates etc. The real status of
the latter is beyond all doubt, while that of secondary substance is
only a dependent one : it is nothingbut a species or genus in which the
things primarilycalled substances are (2 a 14-16). Ackrill rightlyremarks (p. 82) that in distinguishingsecondary fromprimarysubstance
Aristotle is not suggesting that 'substance' is used in two different
senses; it would be difficultforhim to concede that withoutupsetting
his whole scheme of categorial classification.
However, it must be stressed (with Ackrill) that our author fails to
say clearlywhat type of distinctionit is. As a matterof fact,secondary
substance is a 'thing' said of a primarysubstance, yet notpresentin it.
From 2 a 19-27 two things are perfectlyclear: (a) the hath' hypokeimenouitselfis a 'thing', not a name nor a definition,since the author
speaks of 'something' the name and definitionof which are necessarily
; (b) this 'thing' as far as it is said of
predicated of the hypokeimenon
an hypokeimenonis virtuallyidentical with its own name (and definition, in a sense). However, it is only fromthe contemporaneouspoint
of view that modern interpreters(e.g. Lloyd, see above, p. 88) and
92
23:15:40 PM
Ackrill,above, p. 91) assert that Aristotleis 'careless' in using 'thing'

instead of 'name of a thing'. Doing so, they fail to stick to a consistent
interpretation,such as Ackrill,who calls (p. 75) Aristotle careless in
thisrespect (at 2 a 31-34) and stressesall the same (p. 82; quite rightly
indeed) that in discussing the relation 'said of ... a subject' our
author (at 2 a 19-27) shows that forhim this relation is one holding
between things,not words (names). Our conclusion must be that not
only the word 'thing' is used ratherloosely but that the same holds
good with regard to the word 'name'. Generallyspeaking the Ancients
did not take a thing's name apart fromthe thingnamed as strictlyas
modernswould do. So in our author's phrase 'both the name and the
definition'(kai tounomakai tonlogon; 2 a 20; 26), 'name' (onoma) does
not referto just a label to designatetilings; ratherhe is speaking of the
mentalstatus of a 'thing', i.e. the thing as named, i.e. as conceivedby
human thinking. Therefore Lloyd's assertion (p. 65) seems to lack
sufficientground that the doctrineof the Categories,by makinggenera
and species only secondary substances, could not possibly be accepted
by a Platonist. It is true, Aristotle's remark (at 2 b 6) that if the
primarysubstances did not exist,it would be impossible forany of the
other things to exist is destructive of the very core of Platonism.
However, this needs not prevent us fromrecognizingin the author's
descriptionof the firstitem of the classificationat hand just the same
kind of 'thing' as the one we have termed before (1977, I07ff.) 'the
mental status of the Platonic Form'.
Another question may be raised concerning the firstitem of the
classification: are secondary substances also said of other secondary
substances ? The answer is to be foundat 2 a 34-35, but at firstglance
it mightbe somewhat ambiguous. As a matter of fact, Ackrill (p. 82)
supposes that someone mightcounter the claim of this assertion ("all
the other things are eithersaid of the primarysubstances as subjects
or presentin them as subjects") by pointingout that animal is said of
man taken as species and that colourtaken as genus is presentin body
taken as genus, to the effectthat both man and body here are secondary, not primary substances. I think, there is no single reason to
raise doubts as to the author's claim at 2 a 34-35. As clearly appears
fromthe context, the author intends to assert that all things other
than primary substance are eitherpredicable of primary substance,
i.e. eithersaid directlyofit or via its being said of secondary substance,
or presentin it, the latter case being beyond the scope of this item of
the classification.The example given is in support of this view (2 a 3693
23:15:40 PM
b i) : "animal is predicated of man and thereforealso of the individual

man; for were it predicated of none of the individual men, it would
not be predicated of man at all". The argument is perfectlyclear:
whenever some thing other than primarysubstance is said of say X,
the latter either is primarysubstance or the predication implies another one in which the subject X' should be primary substance.9 It
should be noticed that in the example given 'man' is used in three
differentways:
(a) "animal is predicated of man" : the universal man
(b) "animai is predicated of the individual man" : the individual man
not of man at all" : 'man' is here used loosely for
(c) "animal
both the individuai man and the universal man.
As for the firstitem these conclusions can be drawn:
(a) the 'kath' hypokeimenou'meant here is to be equated with a 'thing'
in its mental status (whichmay be parallelled to the mental status
of the Platonic Form)
() the 'hypokeimenon'meant in this 'kath' hypokeimenou'eitheris a
primarysubstance (i.e. a real thing in our world; see also below,
3.2.6.5) or a secondary substance (c.q. just substance, 'substance'
taken here loosely forboth primaryand secondary substance).
3.2.6.2 The seconditemof theclassification
The second item of our author's scheme of categorial classification
concernsthe particular accidents, i.e. 'things' that are presentin individuals (viz. primarysubstances), yet not said of them. Thereforethey
are certainlynot to be equated with a 'thing' in its mental status. On
the other side, their ontological status is not as real as that of the
primarysubstances whichtheyare in, theirreal existencebeingentirely
dependent upon that of primary substance. The author's examples
(1 a 25-29) do not affordany reason to call in question that he means
an individual subject (see also above, p. 89). So the en hypokeimeni
is a particular accident which 'informs',in an accidental (i.e. nonsubstantial, or non-essential) way, primary substance. To take the
author's examples, the 'white' informsthe subject to be a 'white subinformsit to be a 'grammarject', whereas the knowledge-of-grammar
Ackrill'scomments
(82-83)seemto be beyondthequestion.
94
23:15:40 PM
10
knowingsubject' . But what exactly is to be understood by 'white'
here? We have to remember,again, that it is a 'thing' (as belonging
to the class of 'thingsthere are' mentionedin I a 20). Well, this 'thing'
cannot possibly be anything else but the (accidental) propertyembodied in matter (c.q. in the soul when it is embodied). Therefore,the
'thing' meant here may be parallelled with what I have termedbefore
(1977, X05-107)'the Platonic Form taken in its immanentstatus'.
As for the hypokeimenonitself (meant in the phrase 'en hypokeimeni'), it cannot be anythingbut primarysubstance; forthe latter's
status see below, 3.2.6.5.- For the rest, there are many questions to
be asked about this item, which are betterdiscussed in the framework
of the thirdone (3.2.6.3).
3.2.6.3 The thirditemof theclassification
The thirditem of the classificationis universal accident, i.e. 'things'
that are both said of a subject and present in a subject. To remind
our author's example: knowledge is present in a soul as its subject,
and it is also said of a subject, e.g. knowledge-of-grammar
(1 b 1-3).
As forthe correctmeaning of 'beingpresentin' the firstpoint to be
made is that Ackrfflis wrong in saying (p. 75) that our author means
that the inherenceofa propertyin a kind of substance is to be analysed
in terms of the coherence of individual instances of the propertyin
individual substances of that kind. He is quite rightin assertingthat
for a propertyto be in a kind of substance it is necessary that every
substance of that kind should have it. But this does not amount to
saying that it is impossible that every substance of that kind should
have this property.The problem at hand was made more explicit by
later thinkersin the stock examples of the 'white' of snow or leadpaint as contra-distinguishedwith that of cheese, the formerbeing an
'essential property',the latter only an accidental one.. Our author's
formulaat i a 29-b x does not exclude the inherenceof universal accidents in some species or genus,i.e. in secondarysubstance or in a genus
or species of a non-substancecategory.
Anotherremarkshould be made. In spite of his distinctionbetween
the second and third item of the classification,Aristotle apparently
fails in drawing a clear-cut distinction between particular and universal accident. As has been already noticed by Ackrffl(p. 82), our
10It shouldbe noticedthat
is notin thesoulas a place,suchlikea
knowledge
is virtue'
pencilis in a box. Comparethe Socratic(Platonic)adage 'knowledge
and Aristotle's
doctrineof 'the soul 'undergoing*
(paschousa
) knowledge'.
95
23:15:40 PM
author seems to introduce(in 2 a 27-34) the same thingin two different

11
ways. The passage makes it indisputably clear that, for Aristotle,
the items (2) and (3) ultimatelyare one group, viz. that of the proper:
ties, being 'things' that are presentin a subject (en hypokeimeni)
2 a 27-31: "As forthingsthat are in a subject,withthe mostof them12
is predicatedof
(epi mentnpleistn)neitherthe namenorthe definition
that in whichtheyare present.Withsomeof them(ep' enin),however,
to preventthenamefrombeingpredicated
ofthatinwhich
thereis nothing
can neverbe
theyare present,though,on the otherhand,the definition
predicated".
The gist of our problem is in that in Greek universal properties(such
'
as whiteness) may be signifiedby 4he white (toleukon), not only by
'
'
the abstract name whiteness (1leukots
).13 Well, the definitionof the
concerns
it
in
its
abstract
(formal)nature (e.g. whitepropertyalways
ness, not: the white). So it is quite obvious that, unlike the definition,
the name of the propertyhas an ambiguous semantic value. That is
why Aristotlecan say (2 a 31-34) : 'the white' (to leukon) being present
in a body as its subject is predicated of that subject, for a body is
called white (i.e. we signifya thing by '(the) white', leukon).14 One
result of this ambiguity, it should be stressed, is that the concrete
name ('the white*
i to leukon) in signifyingthe particular accident (e.g.
thisparticularwhite in thisparticularsubject) is linguisticallyidentical
with the abstract one signifyingthe formalproperty (whiteness).The
other, which was bound to follow from the firstone, was more influential, indeed. To say it in Lloyd's words (p. 61), there was in
Ancient thinking a widespread failure, or refusal, to distinguishbetween a particular white thing and the particular white colour it has.
Well, the failure (i.e. that which we moderns are inclined to call a
11For thatmatterAckrillseemsto be wrongin assumingthatmoderns
should
'
insteadof 'things'
, as if theywereto introducethesameconcept
say concepts'
in twodifferent
ways.I think,we haveto avoidanymodernusagein thisfield.
Cp. above,p. 84; 88; 92i.
12Moderntranslations
thecases
seemtoblurtheauthor'sphrasein distributing
So "it is generally
of the 'things'at handratherthanthe 'things'themselves.
thecase" (Oxfordtranslation)
; "la plupart"du temps"
; "in mostcases" (Ackrill)
(Rolfes).
(Tricot);"allermeist... nurmanchmal"(Gohlke);"meistens
do stressthelackofa clear-cut
I haveto agreeall thesamethattheirrenderings
foundin
betweenthe items(2) and (3).- The Latin translations
distinction
Latinusare moreaccurate;Boethiushas "in pluribusquidem";the
Aristoteles
translatiocompositaand Moerbeke"in plurimis
quidem".
13Edghill(Oxfordtranslation,
to English'evil'usedin the
notead locum)refers
one case as a substantive
noun,in theotheras an adjective.
14I shallcomebackto thispassage(especially2 a 32) below,whendealingwith
theproblemofwhatis called'predication'(oursection4).
96
23:15:40 PM
failure)was unavoidable forthose who tried (e.g. in the Middle Academy, such as Albinos) to combine Plato and Aristotle. However, we
would be wrongifwe were to look forthe starting-pointofthis development as late as in the Middle Academy (as Lloyd seems to do). As a
matter of fact it is found as early as in the Platonic dialogues, where
'to kalor sometimes stands for a beautiful particular (e.g. Sympos.,
210 5; Phaedrus, 274 A 8), sometimes for 'Beauty itself' (e.g. Sympos. , 211 A 6, Sophist, 257 D 11), sometimes for beautiful taken as a
universal accident (Grgias, 492 9 : tou kalou tou ts dikaiosuns kai
ts sphrosyns); as well as it appears not of Plato's concern to distinguish between a particular white thing and the particular white
colour it has, since his main point is to distinguishbetween a particular
white thing and the universal white colour it partakes in. Therefore,
the occurrenceof such ambiguitiesin the Categoriesneed not surprise
us at all, even though Aristotle,unlike Plato, clears part of the ambiguity away by using Ho ti leukor (so at 1 a 27 ; to be compared with
ho tis anthrpos
, i a 22ff.)to signifythe particular accident (the particular white colour). However, his use of 'to leukori at 2 a 31-33 still is
ambiguous: at 2 a 31 it means both theparticularwhitecolour (en hypokeimenion) and the universalaccident (kathgoreitai
touhypokeimenou)
and at 2 a 32 'leukori means ambiguouslywhiteas well as whitething,
which is due to the word's predicative position in the phrase leukon
gar soma legetai, where it may be rendered with "one calls a body
white" as well as "one names a body: 'a whitething' ". Therefore'to
leukori is found at 2 a 31 for both particular and universal accident,
and that in the same linguisticform,even with the same word numerically; at 2 a 33 leukon' stands ambiguously for universal property
and some primary substance denoted denominatively by the name
'white'
Especially the latter ambiguity was going to do its work in the
views of the Middle Academy: the phrase 'leukon ti' could mean (just
like with Plato himself)'an instance of whiteness'- equally in the sense
of substantive (substantiated) noun, to the effectthat it signifiesthe
composite (synolonor syndyasma), or of the adjective, i.e. signifying
'the whitesaid of something'- as well as 'the white being presentin it'.
Indeed the Categories
, too, have theirobscuritiesin this respect: there
is implied throughouta theoryof in re universais; neverthelessCh. 2
draws a distinction between universal and particular accidents as
separate items of the classification.
However, Lloyd (p. 62) seems to be definitelywrongin considering
97
23:15:40 PM
these differentviews as two doctrinesthat are hard to reconcile. It is

'
'
'
'
true, the distinctionmade in Ch. 2 between particular and universal
seems to be more in line with genuine Platonism, whereas the theory
of the in re universais would seem of Aristotle's own invention. But
putting things that way one risks to fall a victim to optical illusion.
Aristotle's turn of interestto the in re universais can certainlynot be
considered an unexpected one; rather it was quite obvious to insist
upon the immanentstatus of the 'universal' (Platonic Form). Therefore
Lloyd seems to be beside the mark when asserting(p. 64) that his very
closeness to Plato (especially in the Categories) makes Aristotle's
silence, or even equivocation, over the status of universal propertyas
distinct fromparticular properties (poia) seem unjustifiable.
The foregoingremarksdo definitelynot amount to arguing all obvious obscurityand ambiguityin the Categoriesaway. Their only aim
is to show that the ambiguitiesin the Categoriesare very close to those
discussed in section 2 (on the differentstatus of the Platonic Form),
to the extent,indeed, that theremay be foundquite a lot of continuity
in Plato's and Aristotle'sviews of the matter. The latter's rejection of
the doctrine of the Ideas was just a rejection of their separate (i.e.
transcendent) status and an attempt at yielding, accordingly, the
ontological monopolyto theirimmanentstatus. No doubt, when doing
so Aristotle was turningthe relative positions of the intelligibleand
the sensible worlds upside down, as may be shown most clearly from
2 b-6 (see above, p. 87). Nevertheless,he is opposing to Plato's view
of the eid, rather than to the eid themselves. Aristotle's position in
this respect may be compared with Plato's Neokantian interpretersof
the firsthalf of our century (Cohen, Natorp) : they, too, were of the
opinion that in explaining Plato in the sense of Critical Idealism (by a
reduction of the three status to just the logical one, which we have
termed the mental one) the true Plato was brought about.15 Well,
Aristotlehad to struggleagainst his master Plato in personain orderto
arrive at what he (wrongly)consideredto be the trueexplanation of the
eidos, i.e. the eidos (form)taken in its immanence,embodied in matter.
15Paul Natorp,PlatosIdeenlehre,
Leipzig1902,226-235;"Die Idee selbstwird
sich fortannur behauptenknnen,wofernsie sich auszuweisenvermagals
d.i. methodischen Begrndung von ErGrundlagezur "Mglichkeit",
und endderIdee grndlich
Damiterstwirddie falscheAbsonderung
fahrung.
nmlichvon der Aufgabe
sein: die von der Erfahrung,
gltigberwunden
Idee verlierteben durchdieseAbsondeihrer Ermglichung abgesonderte
auf die es doch uns zuletztnur
rungjedes InteressefrunsreErkenntnis,
ankommenkann" (235).
98
23:15:40 PM
3.2.6.4
'
'
The ontologicalstatus of the things meant in the items (2)
and (3)
On the groundof the foregoingdiscussions and referringto our conclusions (a) and () (which concern the status of the 'things' meant in
the firstitem; see above, p. 94) we may state the followingpoints:
() The en hypokeimenimeant in the second item of the scheme of
categorial classificationis an embodied accidental property,e.g.
the particularwhite colour being present in a primarysubstance.
It has a real status that is entirelydependent upon the existence
of that primarysubstance in which it is present
itselfmeant in the second item is nothing but
(8) The hypokeimenon
primarysubstance
for the third item of the classification; the en hypokeimeni
As
(e)
(universalaccident or universal property)has the weakest position
"
in the domain of thingsthere are" (tonontn,1 a 20), as they are
missngthe accidental way of being the particular accidents have
and, unlike secondary substance, they cannot enjoy some affinity
to primarysubstance nor, accordingly,borrow somehow fromthe
latter's full-fledgedbeing
() The. hypokeimenonitself which is meant in the phrase 'en hypo'
keimeni as used in the third item is either secondary substance
(i.e. a genus or species in the category of substance) or another
universal accident (i.e. a genus or species in a non-substancecategory)
'
(y)) The kath hypokeimenonmeant in the third item is the universal
accident, again; see sub (e)
(6) The hypokeimenonitselfwhich is meant in the phrase 'kath' hypo
'
keimenou as used in the thirditem is eitherprimaryor secondary
substance.
3.2.6.5 The fourthitemof theclassification
It is quite plain that this item is primarysubstance, i.e. the concrete
particular existingin the material world. It is not the embodied form
('immanent eidos') itselfbut the composite of matter and form,as it
is dealt with in Metaphysics Z and H, where is found a far more
fundamentaldiscussionofsubstance (ousia) than gives the fifthchapter
of the Categories.
It goes beyond all doubt that for Aristotle as early as in the Categories it is primarysubstance that has the exclusive prominence of
99
23:15:40 PM
being.One has only to remindof the passage discussed already (above,

p. 87 and 98) whereAristotleis most explicitin givingabsolute priority
to primarysubstance : "if the primarysubstances did not exist it would
be impossible for any of the other thingsto exist" (2 b 6). He is evidently intended here to mark out primary substance, i.e. the composite of embodiedform and matter, as somehow basic, against his
master Plato, who gave the most prominentposition to the Form in
its transcendent status. Aristotle's statement is to the effectthat
secondary substance, which is a universal, cannot exist without primary substance, to be sure, without a number of (i.e. more than one)
primary substances; primary substance, however, can exist without
secondary substance, i.e. apart from,and not dependent upon, the
universal called secondary substance. For that matter,in this passage
'
'secondarysubstance is loosely used including ambiguously both the
separate eidos rejected by Aristotleand the eidos in its mental status;
for this ambiguity, see also Ackrill, ad loc. However, this much is
beyond all doubt: primarysubstance has, for Aristotle,the absolute
primacy in the domain of being.
Concluding this section I give the result concerning the fourth
item:
(1) The oute en hypokeimenioute kath1hypokeimenouis primarysubstance.
'
'
3.2.7 The relationsbetweenthedifferentthings
The interrelationsof the differentitems may be described in terms
of opposition and affinity.The strongestoppositionsare found:
- between secondarysubstance {said of,notpresentin) and particular
accident (presentin, not said of)
- between primary substance (neither said of nor present in) and
universal accident (both said of and presentin).
On the otherside there is only partial opposition and some affinity:
-
between secondary substance and universal accident (both being

said of)
between primary substance and secondary substance (both not
being presentin) ; see especially Aristotle's argument at 3 b 10-23
wherehe insiststhat secondarysubstances are just kindsofprimary
substance
100
23:15:40 PM
between primary substance and particular accident (both being

not said, of); their affinityespecially concerns their real status (see
above, p. 94i.)
between particular and universal accident (both being presentin) ;
theiraffinityis most clear in 2 a 27-34, discussed above, p. 86f.
The followingscheme may be dressed:

secondarysubstance=
Reality
> primarysubstance <X
universal accident
*particular accident
There is an exclusive opposition between primarysubstance and universal accident as well as between secondary substance and particular
accident. On the other hand, there is some affinityand compatibility
between primaryand secondary substance as well as between particular and universal accident, and also between primarysubstance and
particular accident as well as between secondary substance and universal accident. Well, as to the latterpair it may be noticed that since,
to Aristotle's mind, the strongestposition in the domain of being as
taken over the whole scheme of categorial classification(i.e. both substance and the nine accidents),is held by primarysubstance and particular accident, it followsthat secondary substance and universal accident have theirlack of reality in common, (cp. above, pp. 95 ; 92ff.).
It is precisely that resemblance and indistinctnesswhich can be
considered the most fundamental weak spot in Aristotle's mapping
out the area of being. The difficultycomes best to lightin a discussion
of the relationshipbetween category and predicable.
Categoriesand predicables
My own view of the proper meaning of the Aristotelian categories
will be discussed in the next section. Let me confinemyself,now, to
state that theirpropermeaning is just to performthe rle of a general
heading (in fact, one of ten general headings) covering a class of
entities, to the effectthat they are classes of names rather than of
predicatesin the technicalsense of sentence-predicates.In dressingthis
list of headings Aristotlearrived at a classificationof the main typesof
'thing1involved in the whole of what presentsitselfto human thinking.
The doctrine of predicables, on the contrary,is linked up with a
special type of sentence-makingand based on the principle of con3.3
101
23:15:40 PM
vertible predication. In line with the Academic practice initiated by

Plato's Sophist it states the analytic propertiesof a classificationby
genus and species. As a matter of fact the doctrinewas given in the
Topics in the frameworkof findingdialectical arguments.To that end
Aristotle considers (Topics I 4, 101 b 13-28; cp. I, 5-9) the variety of
relationsbetween subject and predicate occurringin the premissesof a
dialectical argument or expressed in dialectical problems raised for
discussion. Any predicate turns out to be either convertiblewith the
subject or not. If convertible,it eitherstates the essence of the subject,
in which case it is its definition,or does not so, in which case it is a
property.If not convertible,it eitheris part of the definition,in which
case it is the genus (or difference)of the subject, or is not so, in which
case it is an accident.(Later on in the historyof logic a fifth(species)
and sixth predicable (individual) were added, by Porphyry and
Boethius respectively,not without a ratherconsiderablechange of the
originalpurportof the doctrine).So the predicablesmake up the logical
frameworkforthe whole treatment of problemsin that they provide
the appropriate subject and predicate terms of the commonplaces
which are indispensable to any dialectical discussion.
3.3.1
The oppositionof categoryand predicable
Lloyd (65) rightlypoints out a fundamentaldifficultywhich anyone

studying the Categoriesin connection with what elsewhere (especially
in the Topics) is said about the predicables cannot fail to notice. On
his view the categories(whatever else theymay be, he adds) are classes
of predicate and subject, which entails them to be absolute divisions
of terms,where the predicables were relative to a proposition (op. cit.,
65-66). They are, on this view, all presented as potential ingredients
(rather than as parts) of a proposition. This view can find some
support, indeed. To say it in Aristotle's own words, the categories are
the thingssaid without any combination (ta de aneu symploks,x a 17
or ta kata mdemiansymploknlegomena,1 b 25). The predicables, on
the other hand, which, especially in the context of the Topics, are the
five distinctmannersin which a predicate is said of a subject, describe
the relations between predicates and subjects and in doing so are
definitelynot classes of possible predicates and subjects, but rathera
list of predicative relationshipsbetween terms.
For that matter, Aristotle seems himselfto oppose category and
predicable in nearlythe same way (Top., I, 9). At 103 b 27-35 he clearly
states that startingfrom things in reality (coming up for dialectical
102
23:15:40 PM
discussion) and looking for their true nature (to ti esti) one always
arrives at one of the ten categories; when starting fromthe kinds of
predicate,on the otherhand, the predicables will come into the picture
(103 b 35-39).
However, the oppositionbetween categoryand predicablemay come
more clearlyto lightif one takes (more properly,I would presume; see
above p. (jiff,and our next section) the categories as classes of names,
rather than of sentence-elements(predicates and subjects). Therefore
the predicative relations between the names (terms) classifiedin the
differentcategoriesaxe really beyond the scope of the Categories.The
predicables, on the other side, are especially concernedwith a termin
respect of its being said in combination with other terms (kata symplokn, i a 16). The categories, accordingly,are absolute divisions of
terms,to use Lloyd's formula,and that to the extent that they are
arrangementsof terms (i.e. names) taken on their own, whereas predicables are divisionsof termstaken as possible sententialingredients,
accordingto theirmutual predicative relations.
3.3.2 The impact of theopposition
This opposition between categories and predicables may be considered only a formalone. Its material impact, however, can hardly
be overestimated.It is brought to light by two major problems concerningAristotle'sview of particular and universal being given in the
Categories.
First, thereis that remarkable relationshipbetween the genera and
species in the categoryof substance on the one hand and primarysubstance on the other. As is well-known,the decisive point here is that,
forAristotle,all strata higher than primarysubstance lack realityin
the full sense; that is why he could not help rejecting the Platonic
chrsmos.However, when it came to specifyingthe ontologicalstatus
of what was termed universal 'thing', Aristotle had no other choice
but to telescope all the genera and species of the substance category
into primarysubstance; see e.g. in Metaphysics990 b 34ff.; 997 b 5ff.;
1002 b 27-32; 1086 b 6-7. As is easily seen, such procedureof 'telescoping' universaisinto particularsis not only useful to deprive the former
fromall kind of real existence of their own but cannot help rendering
Aristotle'sconcept of 'thingsthere are' (1 a 20), to say the least of it,
ratherequivocal.
Another difficultyis about the genera and species in the nonsubstance categories. For that matter, Lloyd seems to be completely
103
23:15:40 PM
mistaken in thinking(66) that, since at Categ.,2 a 15-17 Aristotleexpressly restrictsgenera and species as secondary substances, to those
which contain (logically) firstsubstances as their particulars, genera,
such as science (episim) which is an instance of en hypokeimeni,are
excluded fromthe domain of genera and species as meant in the Categories. The non-substancegenera and species (i.e. all those 'universal
things' that are not secondary substances) really are included in the
doctrine of the categories, and make up the third item of the classification ("things both said of a subject and present in a subject";
i a 29-b 3, see above, p. 95-)- The real problem,however,is that, for
example, some kind of quality, such as white, that clearly belongs
to the non-substance category, Poion, whenever it is said of snow
is an essential property, but said of man an accidental one; well,
the formerone is much alike the differentiaof the category of substance. (Cp. Simplicius,In Arist. Categ.48, 1-11).
Our thirdproblemis closely connectedto the firstone : what exactly
is the position of the differentiae
(diaphorai, 1 b 17) ? The word 'genus'
is sometimesused by Aristotleratherloosely includingdifferentia(e.g.
Topics, I 4, xoi b 18-19; I 5 and 6; IV 2, 123 aii-19). The Categories,
too, seem sometimes to be rather careless on that score. Ackrill has
rightlypointed out (81-82) that in characterisingsecondary substances
(2 a 14-16) Aristotledoes not, as mighthave been expected, go on to
say that secondary substances are 'things said of a subject but not
present in any subject'. Instead he describes them as the species and
genera of primary substances and only later (3 a 9ff.) he makes the
point that they are said of primarysubstances but not presentin any
subject. He seems to be quite rightin assuming that this is only because Aristotle (3 a 2iff.) is going to say (surprisingly,Ackrill adds)
that the differentiaeof substance genera, though not themselvessubstances, are neverthelesssaid of the individualsand species subordinate
to the genus concerned,and are not presentin them. Ackrill (85) takes
Aristotle'sstatement (at 3 a 2iff.) that somethingthat is not substance
is neverthelesssaid of substance fora surprisingone, as it can hardly
be reconciled with the scheme of ideas so far developed; for Ackrill's
specific arguments,see his Notes, 85-86. He thinks it plausible (for
Aristotle) to take the differentiaas part of the 'what-is-it' (ti esti) of a
secondary substance and this provides (on his view) a strong motive
forassimilatingit to substance even while distinguishingit fromgenera
and species. True, Aristotle may have been influencedby such considerations as put forwardby ProfessorAckrill,yet the confusionstill
X04
23:15:40 PM
remains: forone thingthe differentiais on a line with secondary substance, for another it seems to be much alike somethingadventitious
('accidental') to it (viz. the genera); and that while it is essential to the
species concerned (which species may be a subordinate genus), to the
effectthat, within one and the same category, one and the same differentiais both accidental and essential. Thereforewe have to face the
intricate question of how to telescope such things as differentiae
? To
schematise our problem:
41
X
I
genus <5
^
'
'
1
^species
It seems not to be of much help to take the pair of opposite differentiae

(e.g. rational vs. not-rational)as one disjunctive essential attribute for
any genus to become real being (on the level of primarysubstance),
since a set of disjunctive (i.e. mutually exclusive) elements cannot
possiblybe telescopedtogetherwiththe genera into primarysubstance.
3.3.3 The obscureposition of thedifferentia
As a matter of fact the position of the differentiais rather obscure
throughoutthe Topics. Especially in Book IV the dissimilaritiesbetween genus (c.q. species) and differentiaare stressed. In IV 2, 122 b
I2ff.the author shows that differentiais not genus (122 b 12-17; I23 a
3-5), nor eidos (122 b 18-24)> the differentiashould neitherbe taken in
such a manner as to make the eidos a genus (122 b 37-123 a 1) nor in
such a manner as to go the other way around (123 a 1-2). Its own
position,which is also underlinedin Top . IV 6, 128 a 20-30 and VI 6,
144 a 5ff.,does place the differentiaoutside the genus-speciescolumn
that Aristotle intends (and has) to telescope into primarysubstance.
So much for the dissimilarities.Nevertheless 'genus* is used many
times rather loosely including the differentiaconcerned, as has been
pointedout above, (p. 104). There seem to be good reasonsforit, indeed.
105
23:15:40 PM
In Topics IV 6, 128 a 20-21 Aristotleintroduces Ms warning against

the confusionof genus and differentiawith the assertion that some
people think (dokeitisi) (it should be borne in mind that the author of
the Categoriesis among them; see Categ. 5, 3 a34-b 9) that the differentia,too, is an essential predicate of the various subordinate species. Another resemblance is stated in the Topics several times, although in an indirect way: a thing's differentianever signifiesits
essence, but rather some quality (Top. IV 2, 122 b 16-17; IV 6, 128 a
144 a. 18-19 an(i 21-22) ; well, the very same characteristic
26-29; VI
is given of secondary substance (genera and species) in Categ. 5, 3 b
15-21. It should be duly stressedhere (against Lloyd, 66) that thereis
not a shred of contradictionin Aristotle's saying that the differentia
does not signifya thing's essence but rather some 'quality' (tohvti,
best accented thus, as well as in Categ. 5, 3 b 15ft.; see Lloyd 66, n. 3).
Unfortunately,Lloyd failed to see the important differencebetween
this kind of 'quality' and the categoryof quality (forthe correctview,
see Ackrill,88-89) and did apparently not rememberthat secondary
substance, too, had been called a poion ti by Aristotle.For that matter,
it is exactly this meaning of 'quality' that is found later on in the Stoa
and with the grammarians;e.g. Apollonios Dyskolos, De pronominibus,
336: ousian smainousin hai antnymiai,ta onomata ousian meta
poiottos; Prscan, Inst, gramm. II 55, 6 has: proprium est nomnis
substantiamet qualitatemsignificare,a statementwhichwas to become
veryinfluentialin the developmentof Mediaeval logic and semantics.16
The unclear position of differentiaas occurringin the doctrine of
the categoriesmay also be revealed by assertingthat (a) the differentia
turns out to have somethingof an essential attribute of genus, once
the genus has specified itself into the subordinate species, whereas
(b) 'until then' it is quite alien to the genera-speciescolumn.The former
aspect (a) comes to light whereverdifferentiais included in genus (see
seems to be taken
above, p. 104) ; the latteroccurs wheneverdifferentia
This
in its own right.
amounts to acknowledgingthat differentiadoes
have an essential predicative relation to various subordinate species
128 a
(Aristotlehas : h diaphora en t&iti esti tn eidn katgoreisthai,
to
a
relation
the
it
has
not
such
unlike
genus concerned,
20-21), yet,
rational'
'man
is
to
the highergenera. For instance, one is allowed say
as well as 'man is an animal' ; but, unlike 'man is a body' and 'animal
18See myLogicamodernorum,
A Contribution
to theHistoryofEarlyTerminisi
Logic,vol. II, passim.
106
23:15:40 PM
is a body' the statement 'the 'rational' is a body' does not concernany

essential predication (since a 'rational' may be immaterial).
3.3.4 Conclusion
Before summing up the results of this section (3.3) it should be
noticed that when discussingdifferentiain the Topics Aristotledeals
not only with differentiain the substance category but with those in
the other categoriesas well; as a matter of fact, most of his examples
are taken fromthe other categories.
It may be clear fromthe foregoingdiscussions that Aristotlecould
only dispose of the separate Platonic Forms in telescopingthe genera
and species as secondarysubstances into the level of primarysubstance.
However, differentiaturns out to resist to this attempt in a peculiar
way. The (partial) miscarriageensuingupon this resistanceis the mor
regrettableconsideringthat the differentiaeseem to be as full-fledged
inhabitantsof the world of universais (Plato's Forms) as are the genera
and species. To say it in other words: in having some insurmountable
problems with the differentiaeAristotle could not help failingin his
attemptto get rid of Plato's Ideal World as a separate Realm of being,
in which,fromthe Sophist onwards, the Forms of the differentia-type
played a rle of utterimportancein the constitutionof Being. For that
matter, in resisting to Aristotle's effortsto telescope the Platonic
Forms into the material every-day world, it is the very differentiae
that set offthe hierarchicorder as a basic characteristicof the Ideal
World. Hierarchy and telescoping really are one another's natural
enemies.
To be continued
Leiden,
FilosofischInstituut
Witte Singel 71
107
23:15:40 PM
VivariumXVI, 2 (1978)
Master Guido and his View on Government:
On Twelfth Century Linguistic Thought
C. H. KNEEPKENS
0.
Introduction
his Studies on Priscian in the Eleventhand TwelfthCenturies

}
Dr
R.
W.
outlines
the
Hunt
of
of
In
development linguisticthought
that era. Starting with the late eleventhcenturyGlosule,the Note
Dunelmenses and Peter Helias' Summa super Priscianum, Dr Hunt
finisheswith the rediscoveryof the School of Ralph of Beauvais. Of
course, many details of this development still remain uninvestigated.
So in the course of my currentresearchinto the lifeand worksof Ralph
of Beauvais, I find that twelfthcenturytheorieson syntax in particular have still been only superficiallyexamined, although, according
to Dr Hunt "it is generallyrecognizedthat the principal achievement
of mediaeval grammarianswas the organizationof syntax; and it was
in this field that the grammarians of the second half of the twelfth
centuryworked most fruitfully".2
One of the key notions of mediaeval syntax, and particularlyin the
grammarof the pre-modistictype, was regimen,as has recentlyagain
been stressed by ProfessorJan Pinborg in his study Some Syntactical
Conceptsin Medieval Grammar.3 Up till now our acquaintance with the
reception of this notion in early twelfthcentury grammar has been
restrictedto the texts printed by Charles Thurot in his Notices et extraits4 more than a hundred years ago.
Thurot noticed an isolated use of the term regereas an equivalent
1 R. W. Hunt,StudiesonPriscianin theEleventh
I Petrus
and Twelfth
Centuries,
Helias and His Predecessors
, in: Mediaevaland RenaissanceStudies,I (19411943),pp. 194-231(quoted: Hunt I); id., Studieson Priscianin theTwelfth
II The Schoolof Ralph ofBeauvais,in: Mediaevaland Renaissance
Century,
Studies,II (1950),pp. 1-56(quoted:HuntII).
2 HuntII, pp. 35-6.
3 JanPinborg,SomeSyntactical
in MedievalGrammar
, in: Classicaet
Concepts
IX, 1973,pp. 496-509.
Mediaevalia,Dissertationes
4 Ch.Thurot,Noticesetextraits
dediversmanuscrits
latinspourservir Vhistoire
au moyenge,in: Noticeset extraits.. . . Vol. XXII,
desdoctrines
grammaticales
a.M. 1964](Quoted:Thurot).
2, Paris 1869[Repr.Frankfurt
108
23:16:00 PM
of the usual term exigerein grammatical writingsfrom the seventh

century onwards.5 At the end of the eleventh century the situation
altered drastically:e e.g. Peter Abailard and Hugh of St. Victor made
frequentuse of it. But the firstgrammarianwhose reflectionson this
topic were available in printforfurtherresearch,was Peter Helias(/Z.
1150).7 Yet it is obvious fromPeter's own words that he was certainly
not the firstgrammarian to speculate about regereand regimen,for
he argues against the opinions held by grammarians of the previous
period: the antiqui.8
Karin Fredborg has recentlydemonstratedthe dependence of Peter
Helias on William of Conches' Glose super Priscianum, and stresses
the debt of both of them to the works of their predecessors.9Among
these was, as Dr Hunt has pointed out, a certain Master G.10He must
have been one of the authoritiesofhis time,forhis views are frequently
referredto in the fifthsection of the Note Dunelmenses,11and William
of Conches quoted him togetherwith the famous Master Anselm in
the second and revised edition of his Glose.12This paper focusses on
this Master G. and on his view of government. But first,it will be
necessary to restoreto this master his gloss on the Priscianus minor,
the authorship of which had not been established up till now.
I.
I.i.
Master Guido and His Gloss

The Manuscripts
The gloss in question has come down to us spread over threemanuscripts:

I.i.i.
The MS London, BM Burney 238 (Sigi. ) 13
1.1.1.1. Parchm.; 37ff.; s. XIII (184.0 catalogue)
6 Thurot,
p. 523 (and p. 82).
Thurot,
p. 239.
7 Thurot,
pp. 240-246.
Thurot,
p. 240: Dubitatumest autemet quesitumab antiquisquid sit dic'
tionemregeredictionem'
TheDependence
; cf.KarinM. Fredborg,
ofPeterHelias
SummasuperPriscianumon WilliamofConches*
Glose superPriscianum,
in:
Cahiersde l'Institutdu moyen-ge
de Copenhague),
grecetlatin(del'Universit
" (1973),PP- 1-57,esp. p. ii.
9 Fredborg
1973.
10HuntI, p. 210.
11Ibid.
12E. Jeauneau,Deux rdactions
desglosesdeGuillaumede Conches
surPriscien,
in: Recherches
de thologieancienneet mdivale.27 (iq6o), pp. 213-247.
13Catalogueof Manuscripts
in The BritishMuseum, N. S., I, ii, The Burney
Manuscripts,
1840,p. 63.
109
23:16:00 PM
According to the catalogue description,the MS contains two texts:

1. An incomplete and anonymous gloss on the Priscianus minor:
"Anonymi Commentaries in Prisciani Caesarensis libros de constructione".
2. Scholastic notes on the visibles and invisibles: "Scholastica quaedam de visibilibus et nvsibibus".
The fly-leaves(ff.I and 2) have been added later. The modernpart
of the MS' historycan be established froma note in an eighteenthcen"
/In Priscianum /De la
turyhand on f. ir: FragmentaCommentationis
de
Chevalire
D*
la
Eon.
.
olim
/ /
Maffeianus". The MS was
Bibliothque
the
British
for
Government
and
deposited in the Library of
purchased
the British Museum together with the other Burney MSS in 1818;
Charles Burney no doubt acquired it togetherwith other D'Eon MSS
at the public auction of the greater part of D'Eon's possessions in
1791.14In all probability,the mysteriousFrench nobleman acquired
this MS togetherwith the other MaffeiMSS in his collection at the
sale of AnthonyAskew's libraryin 1785. Askew, in his turn,purchased
them from the collection by Raffaele Maffei,the Bishop of Aquino
15
(1516) and of Cavaillon (1527), who died in 1537. Its mediaeval
provenance remains unknown.
The ff 3ra-36vb
The MS falls into two main sections: A) glosses on Priscian, XVII
1-30 and B) glosses on Priscian, XVII 52-142. The originalorderwas
disturbedby faultybinding.
I.I.I.2.
A. The glosses on Priscian , XVII 1-30

The glosses on Priscian, XVII 1-30 are writtenon the ff.3ra-nvb,
30ra-35vband i2ra"vb,in this order, without gaps in the text. At the
end of f. I2vb, the text breaks offabruptly:
14P. Pinsseau,L'trangedestinedu ChevalierD'on (ly28-1810),deuxime
dition,Paris 1945,p. 247; cf.M.-T.Dougnac,Art.D'ON Charles- Genevive
- Andr- Timothe,
- Louis- Auguste
in: Dictionnairede bioChevalier,
t. XII (1970), 1329-34.
graphiefranaise,
16Cf. B. L. Ullman,Studiesin theItalian Renaissance,Roma 1955 P- 374
des deuxbiblioRecherches
]; J.Ruysschaert,
[ChapterXX: CodicesMaffeian
LX-LXI
Bibliofilia
La
in:
XVIe
XVe
et
des
romaines
sicles,
Maffei
thques
ofBooks &
pp. 306-355,esp. p. 323; S. De Ricci,EnglishCollectors
(1958-1959),
1930[Repr.i960],esp.pp. 52-53Cambridge
1530-1930,
Manuscripts,
IIO
23:16:00 PM
F. , inc. : QUONiAMin ante EXPOSITIS LiBRis proposuit/ auctor

tractare de litteris JUL modis /uariatis, idest ut per se consideratis, ut con/unctisin sllabas, ut in dictiones, ut in per/fectas
orationes. Expedita igtur secundum tres mo/dosprioresmateria
sua, ad quartum competenterac/cedit,ut uidelicet agat de litteris
eo modo / ad perfectamorationem,idest ad constructonemfac/
endam apte coniunguntur. . .
f. i2vb, des. : . . . aliam personam [Priscian, XVII 30, ed. Hertz II,
1258] quam obliqui significant;hoc ideo quia uerbum obliThe ff.3-10 are writtenin double columns,of 36 lines each; the script
is a regularearly gothic minuscule.The uprighthasta of the d and the
frequentoccurrenceof the e-caudatapoint to the middle of the twelfth
century.The ff.11, 30-35 and 12 are writtenin double columns,of 46
lines each, in a hand of the same period as the ff.3-10, but the script
is of a smaller type.
B. The glosses on Priscian , XVII 52-142
The glosses on Priscian, XVII 52-142 are written on ff.I3ra-29vb
The text starts in the middle of the gloss on XVII 52
and f. 6_.
and runswithoutany interruptiontill the gloss on XVII 142 at the end
of f. 6,where it breaks offabruptly:
F. 13ra,inc. : ostendendo constructioneseorum consequenter agit de
de construct/onepronomnum.Littera sic legitur . . .
f. ,des.: . . . Non fuit/congruum,ut possessua patria a singularibus sumuntur,cum res cum sngularis/numerinon esset possessor
uel possidens quod subntelligeretur,
si pos/sessuapatria a singulari numerodiriuentur.Si enim docenturipse
The ff.13-29 and 36 are writtenin the same hand and with the same
mise-en-pageas the ff.11, 30-35 and 12.
L1.1.3. The f. 37
This leaf is writtenon longueslignes in an entirelydifferenthand
fromthe previous leaves. Stains, faded ink and tight binding make
large sections of the texts almost illegible. It contains a fragmentof
extracts fromHugh of St. Victor's De operibus triumdierum[Migne,
PL 176, 811 C-838 D]. The fragmentcorrespondswith the text printed
in PL 829 D-838 D.16
16For the MSS. of thistreatise,cf.R. Goy,Die
derWerke
Hugos
berlieferung
vonSt. Viktor
, Stuttgart
1976,esp. pp. 98-115.
Ill
23:16:00 PM
1.1.2.
The MS London, BM Harley 2713 (Sigi. H) 17
Parchm.; 42.; s. X/XI (ff.1-34), s. XII (ff.35-42)

According to the descriptionin the 1808 catalogue this codex contains three texts:
I.I.2.X.
"1. Isidoi Hispalensis EtymologiarumLiber primus,sc. de Grammatica

2. Commentarusin incerticujusdam Grammaticam
3. A. Man. T. Boetii, in Porphyrii Isgagogen (!) Liber, imperfectus".
The manuscript falls into two parts: the ff. 1-34, probably dating
fromthe tenth century,are of no interestfor our inquiry.
1.1.2.2.
The ff. 35-42
This manuscript,or rather part of a manuscript,was acquired for

Lord Harley's Library by Humphrey Wanley, Lord Harley's librarian,
on 20 October, 1725. It was offeredby Zamboni, the Italian residentof
the Landgrave of Hesse-Darmstadt, who acted as an agent forBchels,
the Elector's librarianat Dsseldorf.18The formerownerwas probably
J. G. Graevius, whose library was acquired by the Elector JohannWilhelm in 1703.19I am not able to trace its historyback any further.
1.1.2.3. The ff. 3Sra-4irb
These leaves are writtenon in double columns, of 46 lines each in a
small and finegothic minuscule,that bears a great resemblanceto the
script of Burney 238, ff.13-29 + 36. They contain an uninterrupted
fragmentof a gloss on the PHscianus minor (viz. on XVII 142-XVIII
12). The text starts in the middle of a sentence and ends in the comment on XVIII 12, where the scribe stopped his work; the lower part
of f. 4irb is left blank:
F. 35ra, inc. : tuas uel meas ntelligereturquod unus esset patria uel
gens que debet / possidere ilium quem meas uel tuas extrnsecus
significent.Et hoc est quod dicit . . .
17A Catalogueof theHarleianManuscripts
in The BritishMuseum,4 vols.,
1808-1812;Vol. II (1808),p. 709.
le Cf. The Diary of HumfreyWanley,ed. . E. Wrightand RuthC. Wright,
2 vols., London 1966,p. 385-6and 467; and . E. Wright,FontesHarleiani,
London1972,p. 367.
19Cf. A.C.Clark, The libraryof J. G. Graevius,in: The ClassicalReview,V
(1891),pp. 365-72,esp. pp. 366 and 370.
1X2
23:16:00 PM
f. 4irb, des.: Alio non tantum quando ignoraturpossessio, utimurdatiuo in de/sgnationepossessoria coniungendo cum nominatiuo
significantiI possessionem. Sed etiam in alio sensu quam possessoris inue/nisdatiuum;
.1.2.4. The ff. 4iv-42v
These folia are writtenon longueslignes in a hand that differsentirelyfromthat of the previous leaves; it closely resembles the hand
that wroteBurney238, f. 37. This text is also a fragmentofthe extracts
fromHugh of St. Victor's De operibustriumdierum' it runs fromthe
beginning[PL 176, 811 C] to 829 C.
F. 41v, inc. : Inuisibilia autem Dei a creatura mundi per ea que facta
sunt intellecta conspiciuntur.Tra sunt inuisibilia Dei potentia7
sapientia benignitas. Ab his tribus procedunt omnia; in his consistunt omnia; per tra reguntur...
1.1.(1-2). The Reconstruction
of theOriginal Manuscript
I.i. (1-2).I. There is no doubt that the ff.H 41^42 and 37, in^this
order, originallyformedpart of the same manuscript; writtenin the
same hand, they contain continuousextractsfromthe same text in the
rightorder.20
1.1.(1-2).2. Since the firstpart of these extracts in is writtenon the
verso side of f. 41, and the Priscian gloss ends at f. 4irb, one should obviouslysubmitthe relationshipbetweenthe glosses in and H to closer
scrutiny.
contains two larger fragments(1. the glosses on Priscian, XVII
1-30, and 2. the glosses on Priscian, XVII 52-142) ; H contains a continuous gloss on Priscian, XVII 142-XVIII 12. In the firstlines of
Priscian, XVII 142, the question was asked as to why the pronomina
patria were only derived from plural pronouns. The text reads as
follows:
[Priscian,XVII 142,ed. HertzII, i7822-i795]
"Quaeriturcur 'nostras'et
'vestras',id est7)(i.e8a7r
et ufieSaTc,
solispluralibustamapud nosquam
et (j.s&xtu<;
derivantur?
ad quod dicendum,
apud Graecos7)(is$a7r
quod
ideo a solispluralibusfiunt,quia patriaad plurespertinent,
nonad unum,
et sciendum,
quod,quomodopronomina,
quae finitasunt,habentet patria
et possessiva,ut 'meus,tuus,noster,vester,nostras,vestras'".
20Hauraunoticedsimilarextractsfrom
in theMS Paris,
Hugh'sDe operibus
BN. lat. 18096 (s. XII), ff.82-86,cf.. Haurau,Noticesetextraits
de quelques
manuscrits
latinsde la Bibliothque
nationale
, t. VI, Paris 1893,p. 33.
23:16:00 PM
The last lemma in is: Queriturcur nostras,the firstwords of 142.

They are commentedon in a general way. ff starts with the text of a
gloss, the lemma of which was on the previous leaf; the firstlemma in
ff reads: nostraset uestras.When we join the texts together,we obtain
a perfectlymatched and continuous gloss on 142 :
ipseuero
significat
[ , f. ]"Cumautemintrinsecus
ipsumpossessorem,
possessornonpossetesse una res,sed multe,ut patriauel gens,non fuit
cum res [cum]
sumuntur,
congruumut possessiuapatria a singularibus
numerinonessetpossessoruel possidens
; quod subintelligeretur
singularis
numerodiriuentur.
si possessiuapatriaa singulari
Si enimdiceretur
[docentur MS] 'ipse [H, f.35ra]tuas uel meas',intelligeretur
quod unus esset
patriauel gensque debetpossidereiliumquem measuel tuasextrinsecus
Et hoc est quod dicit".
significent.
Furthermore,when we take into considerationthe similarityof the
hands, the same numberof lines, and moreoverthe same manus corrigens in the glosses on XVII 52-142 as in ff, a hand that also shows
the
extracts of Hugh's De
a great resemblanceto the hand that wrote
- , we are forcedto accept that B, ff.13-29 + 36, ff, ff.35-42
operibus
and B, f. 37 in this order,originallyformedone codex.
1.1.(1-2).3- The answer to the question as to whether the ff.1-11,
30-35 and 12 of also belonged to that MS, and whetherthese folios
contain the firstpart of those glosses on Priscian, would have caused
more problems,if a decisive solution were not given by a thirdmanuscript, also mutilated, that supplies the text of these glosses on
Priscian, XVII 1-83:
21
I.I.3. The MS Orlans,BibliothqueMunicipale, (olim87) (Sigi. )
1.1.3.1. Parchm.; i38ff., numbered: pp. 1-388 (the pp. 247-358 are
missing); s. XII.
Accordingto the descriptionmade by Ch. Cuissard,the MS contains:
1. glosses on Paul's Epistles (pp. 1-202)
2. glosses on the Apocalypse (pp. 203-246)
3. glosses on Priscian (pp. 359-388).
The whole MS formerlybelonged to the library of the famous
Benedictine monastery of Fleury, where it was already listed as a
composite codex in the catalogue of 1552 : 219. Glossulaesuperepistolas
21Catalogue
deFrance,Dpartedesmanuscrits
desbibliothques
publiques
gnral
ments,t. XII: Orlans,par M. Ch. Cuissard,Paris, 1889,p. 43; cf. L. Delisle,
in: Noticesetextraits. . . ,Vol.XXXI
manuscrits
Noticessurplusieurs
d'Orlans,
365.
(1884),p.
114
23:16:00 PM
Pauli , super Apocalypsin et super totumPriscianum.22In about 1841'

pages 247-358 were liftedfromthe originalcodex by Libri duringone
of his MSS raids, and became part of his collectionunder the signature
ig . In order to hide his theft,Libri labelled it with the false ex-libris
"Di S. Grmod. Pistoia".23 At present,they formthe MS Paris , Bibliothquenationale, nouv. acq. lat. 1623,24
1.1.3.2. The pp. 247-358
These pages contain an abbreviated versionofthe Glosulae,examined
by Dr Hunt in the firstpart of his study on Priscian in the eleventhand
twelfthcentury.25
1.1.3.3. The pp. 359-388
These pages are writtenon in double columnsin a microscopicgothic
minuscule.26They consist of two quires, the pp. 359-374 and 374-388
(the last leaf of this quire has been torn off). This part contains the
firstpart of the same gloss on the Priscianus minoras (and H) ; the
text breaks offabruptly in the commenton XVII 83 :
P. 359a, inc. : QUONiAMin ante EXPOSITIS.Proposuit auctor tractare
de litterisIIIIor mo/disuariatis, idest ut per se consideratis,ut
coniunctisin sillabas, ut in dic/tiones,ut in perfectasorationes ....
p. 88, des. : ut ego loquor ducentes (/) / similitudinemex ilia construction I in qua pronominisnomnatiuuscum conunctoneapponitur[= , f. 23va in fine]
As the gloss in in a continuously running comment covers the
text of both parts of , i.e. the glosses on XVII 1-30 and on XVII 5283, and supplies the comment on XVII 30-52 lacking in , we can
affirmthat the ff. 3-11, 30-35, and 12 of originally formed one
manuscriptwith the ff.13-29 and 36 of and with H .
I.I.4. Summing up, we now have at our disposal a continuous gloss
on Priscian, XVII i-XVIII 12 :
22Catalogue
Orlans,Introduction,
p. XV.
23Delisle1884,andL. Delisle,Catalogue
desmanuscrits
desfondsLibrietBarrois,
Paris 1888,p. 92.
24Catalogue
Orlans,
p. 43.
26HuntI, pp. 195sqq. ; however,
in note4, Dr Huntconfuses
thesignatures
of
Nouv.acq. lat.1621and 1623,theformer
beinga partofOrlans25g,thelatter
oftheMS inquestion(go); see also L. M. de Rijk,LogicaModernorum.
A ContritotheHistory
bution
II, 1, Assen1967,pp. 21 and 101.
ofEarlyTerminist
Logic,
28Cf.HuntI , p. 195.
115
23:16:00 PM
a)
b)
c)
d)
e)
XVII
: ,
1-30
XVII 30-52
: 0,
XVII 52-83
: 0,
XVII 83-142
: ,
XVII 142-XVIII 12: ,
1.2.
ff.3-11, 30-35, 12 + 0, pp. 359a"37*a

pp. 37ia-377a
pp. 377a-388b + , ff.13&-23
ff.23va-29vb,36
ff.35ra-4i*b
The Author
1.2.1. The authorship of the gloss remains anonymous in all three

MSS. No directmentionof the author has been made in eitherthe text
or the rubrics.
Apart fromclassical authors, five names occur in the glosses; four
of them in examples: Fulco , Herbertus
, Wido and PetrusP The fifth
name is well-knownin grammaticaltractsof the early twelfthcentury:
magisterAnselmus.28In the comment on the use of the pronoun ipse
[ad Priscian, XVII 204], the glossator takes a view contraryto that of
Master Anselm:
[H, f. 39rb]'et quamuis.Unitostenderealiamnturmhuiuspronominis
ipse, hanc scilicetquod alia pronominaterciepersonenon possuntiungi
cum primaet secundapersonauerbinisipronomenprimeuel pronomen
secundematerialiter
apponaturhoc modo: 'Ule egofacio', 'tu ille fads'.
Istud pronomen
terciepersonequod est ipse,iungiturcumprimapersona
uerbiet cumsecundaita quod[queMS] ipsumpronomen
primeuel secunde
secundum
sed tantumintelligitur
personenon apponeturmaterialiter,
nostram sententiam, quia secundummagistrm Anselmum nichil
Est enimeiussentenciaquod ipse estproprieprimeet
oportetsubintelligi.
secundepersonesicutet tercie'.
1.2.2. This note turnsout to be the key to an approximate identification of the masterwho provided the glosses. One of the twelfthcentury
grammarianswho also refersto Master Anselm is William of Conches.
27B, f.9rband , p. 82:'Quandoenimaudioproferri
illud
Fulcouel Herberte,
perse
intelligo:Fulco ueniuel audi'; B, f.20rband 0, p. 385a: 'Quandoquidem
ad absentesquod deberetsubaudiri,suntinfinita;ergo
posita[sc. pronomina]
nominaquibuspostpositapronomina
scribentes
] rationabiliter
preponunt[ur
ut si scriberem
alicuiabsenti: 'egotibimando,
habentsignificationem,
certiorem
Si autemdicis: 'Fulcofratrisuo W.
nullamhaberentpronomina
certitudinem.
ad presalutem'
, post ea subdas: 'egotibimando',tuncpronomina
respiciendo
; H, i. 8: 'Obliqui enim
posita [positiB] nominahabentdemonstrationem'
aliquo modo poterantconstruicum primaet secundapersonasic: 'misereor
in terciapersona.Sed nusquam
Widonis'
transitiue
, licet
Widonis','misereris
'
' sineadiunctione
, Widomisereris
pronominis';
possetrectedici: 'Widomisereor'
uocatiuaet substantiuanon
H, f.8: 'Uideturfalsumesse quod dicitquod
'
'
'
cumcongruedicatur:Petreadestouel Petreuocaris'
socianturnisinominatiuis,
''
uel 'uocare
28Cf.HuntI, pp. 205-6;209-10.
I16
23:16:00 PM
In the second and revised edition of his gloss on Priscian,he quotes the
views of Master Anselm three times.29In the comment on Priscian,
XII 6, he contrasts Anselm's view on ipse with that of a certain
Master G. :
'Nota secundumM. Anselmum
essetria:
[MSParis,BN lat.15130,f.82ra"b]
esseunumquod enimestprime,alteramtantumsecunde,terciumtantum
personetercie.SecundumM. G. unumsolumestipse,quod diuersisadiungiturpersonis.Et si quis querat[f.82rb]cumdicitur:'egoipse', cuiuspersone
est ibi pronomen[prepositione
MS], dicitquod est tercie;nam quamuis
illud[iliaMS] significat
tamenut terciamearnsignificat
primam
personam,
Nosuerodicimusquod ubi iungitur[tangitur
[significat].
MS] cumprima
persona,estprime;ubi cumsecunda,secunde'.
1.2.3.1. One comes across an almost literal quotation of the view held
by our glossatorin the fifthpart of the Note Dunelmenses, that contains
glosses on the Priscianus minor.Here it is also attributedto Master G. :
C. IV, f. i83vb]'Nota secundumM. G. ipse
[MS Durham,Cathedral
Library,
sempertercieessepersone,sed posseaddi primeet secundequemadmodum
et ille,
sed in hoc tam ille quam ceterapronomina
terciepersoneexcedere
quod potestuerbisin primaet secundapersonaconiungisinepronomine
materialiter
apposito,ut 'ipsefaci','ipsefads'. Sempertamensubintelligendumestegouel tu'.
1.2.3.2. Master G. is frequentlyreferredto in this section of the Note.
Dr Hunt counts thirty-three
referencesto him, whereas only two other
mediaeval grammarians are mentioned: M. A. twice and M. Guill.
once.30The compiler of the Note speaks about him as if he was still
alive, and even gives the impression that he is in touch with him.31
From a comparisonwith the correspondingpassages in the glosses, it
appears that only part of the referencescan be identified; some are
nearly literally reproduced, whereas others are expressed in rather
differentwording, e.g.:
NOTE :
[f.i66rb,ad Priscian,XVII 78]
. 'Cum duplicatasubstantiedemonicario.Ita enimexstratio,idestsignif
est secundumsententiam
primendum
illamM. G. qua dicithicagi de pronominibusin simplicisignificatione
substantie'.
GLOSE :
[, f.22;, . 387]
' duplicatademonstratio
substantienihilperfectum,
idestdeterminatum
uel certificatum
sinequalisignificaret
tate,idestcumnullamqualitatemsignificent,
per quam substantiedeterminentur'.
29Jeauneaui960, 227.
30HuntI, p. 210. p.
31Cf.HuntI, p. 209,whichrefers
to theMasterG. ofthefirstsection.However,
thesamearguments
holdfortheMasterG. ofthefifth
section.
II 7
23:16:00 PM
[f.I78vb,ad Priscian,XVII 136]

[, f.2Vb]
'ETGRECIQUIDEM
INUENERUNT
PRO'CorporaproprierefractiuauocanNOMINA
ETC. TRANSLATIUE
A CORPO- tur,quia inuicemin se refringuntur.
RIBUSQUE IN SESE REFRINGUNTUR,
Uerbi [uerba MS] gratia. Si aliquis
idesta quatuorcorporibus
huiusmun- percutiet
lapidemde alio lapide,uterdi, que in se texunturet retexuntur. que lapis et conteritet teritur:ita
Ita placetM. G. quod mirumuidetur. inuicemrefringuntur.
Uel secundum
Nihilenimuideturaliud esse refringi phisicamipsa corporadicuntur
refraccorporain se quam replicali,ut fitin tiua, quia unumin aliud refringitur.
in aerem,
uirgula quasi in circulumrefracta, Uerbigratia.Ignisuertitur
idest replicataetc/.
aer in aquam,aqua in terramet similiterde terrain aqua<m> et sic de
aliis'.
ad Priscian,XVII 137]
[f.I79ra"b,
'Ecce non semelnam dixitsui non
posse construiin extrnsecapersona
nisiper retransitionem,
licetsuperius
in quodameiusdemprincipiicumisto
sit secundumquandic-[f.i79rb]-tum
dam sententiam
posse fieri.Sed nondum illius constructionsexemplum
inuenimusuel inueniemussecundum
M. G.
[S, f.36]
'Quamuisdicamquodsui accipiatur
loco simplicis
cumextrinpronominis,
secus transitactus in eum, tamen
sciendumest quod nonquoquo modo
potesthoc fieri'.
1.2.4. The compilerof the Note and William of Conchesboth attribute

the same view of ipse to Master G. ; this view, a ratheridiosyncratic
one forthose days, is put forwardby the author of the glosses with the
words 'secundumnostranisentenciam'.Furthermore,part of the views
attributed to a certain Master G. in the Note V can be traced in the
glosses. Thereforewe may, I surmise, reasonably assume that both
Note V and William of Conches referto the same master G., who is the
Master of the glosses in question. The lack of counterpartsto some of
the referencesin the glosses can be accounted forby the fact that the
compiler of the Note did not have the text of the glosses to hand, but
took his quotations fromlectures or frommarginal glosses.
Since the name of W(IDO) occurs in two of the examples, the full
name of the Master G. will have been GUIDO or WIDO.
1.2.5. MasterGuido must have belonged to the generationofgrammarians that were productivein the firstdecennia of the twelfthcentury.
The primitiveversion of the Glosule, or at least the one closest to it
supplies the terminuspost,32since, accordingto the Note,Master Guido
rejected an opinion that was put forwardin the Glosule:
32For thisversion,
cf.De Rij, LogicaModern.,II, i, p. 105.
118
23:16:00 PM
[NoteV, f. I70vb]'DicitM.G.se nullomodorecordari

unquamlegisseiuxta
quod in Glosulishabetur.Sed forsitan
aliquisex sua parteita admiscui.
His connectionwith the interpolated version of the Glosule that is
preserved in the MS Chartres,Bibliothquemunicipale, 20g (deleted
33and, under an abbreviated form,in the MS Paris ,
duringW.W. II)
BN nouv. acq. lat. 1623; 34 will be more difficultto determine. The
discussions on the differencein significationsof verbs and nomina
36
sumpta,zsand on the substantive verb clearly show that there must
33Cf.Cataloguegnraldes manuscrits
des bibliothques
publiquesde France,t.
LUI. Manuscritsdes bibliothques
de IQ40 IQ44, Paris 1962,p. 11.
sinistres
34Cf.HuntI, pp. 195-6.
35For thetextofthediscussionon this
, I referto HuntI,
topicin theGlosule
in the argumentation
p. 219. In the Chartresand Fleurycopies,a refinement
was appendedthatis missingin the moreprimitive
Cologneversion: 'Iterum
uerbasic: uel
quandodicituruerbumsignificare
ipsuminherere,
perscrutentur
inherere
istamuocem,quod nichilesset,uel significatimi
illius,quod iterumnon
potestesse proptersupradictasrationes'[HuntI, p. 219,n. i]. This notedoes
not belongat the end of the discussion,but oughtto be introducedat the
In Guido'sdiscussionof thistopic,it occursagainbut thistimein
beginning.
therightplace:B, f.3ir*-b
j 0, p. 365^: 'Sed obiciturhic: Si propriasignificatio
uerbiest significare
remsuam,idest actionemuel passionemut inherentem,
a significatane
adiectiuinominisquod similiter
remsuam
quid [quodB] differt
utinherentem.
Lectorenimuel albussignificai
lectionem
uel albedinem
significat
ut inherentem.
nominiset uerbi.
Itaque confundi
[confusa
] uidetursignificatio
Ad quamquestionem
soluendamquidamdistinguere
conantur
significationem
uerbi[ad . . . uerbi. ] et nominisadiectiuihoc modo: Albedo,inquiunt,
eandemut inherensignificat
ipsamalbedinemin essentia; albusuerosignificat
tem; albetautemuerbumsignificat
earninherere.
Et ita [ut] expressius
significationemuerbi[f.3irb]distinguent
[-at ]. Dicuntquia albetipsuminherere
idestipsaminherentiam
albedinis.
significet,
Sed hii qui hoc dicunt,minusattendereuidenturquod dicunt:Si enimalbet
idestquod albedoinhereat,
aut significabit
albedinem,
significat
ipsaminherere
has uoces,scilicetipsaminherere
aut significata
harumuoeum.De
albedinem,
uoeibusnondicunt.Si autemsignificat
harumuocum,aut hoctotum
significata
scilicetinherere
aut alterumtantum,scilicetuel albedinem
albedinem,
significat,
uel inherere.
Si hoc totum,scilicetquod albedo inhereat,iam uidebituralbet
habere[albereMSS] sensumpropositionset significare
uerumuel falsumintellectum.Sinautemalterumtantum hocdicerepoterunt
quia nequealbedinem
Albedinem
nonsignificat
nequeinherere.
tantum,quia nequesecundumessentiamnequesecundumadiacentiam;secundumessentiamnonsignificat,
quia si
secundumessentiam
nichildifferret
uerbiet substantiui
significaret,
significatio
] nominis,
idestalbetuerbiet albedonominis;secundumadiacen[substantia
tiam autem non significat
albetalbedinem,quia [et ] tunc nichildifferret
uerbiet adiectiuinominisquod ipsi quam [. ] maximeuitare
significatio
uolunt.Inherereautemnondiceturtantumsignificare,
quia eadem[eandem]
de singulisratione[-nemO] oporteret
eos dicereet ita omniauerbaidemsignificarent,scilicetinherere.
Sed fortead hocconfugient,
ut dicantquod albetsignificet
[-at] inherentiam
albediniset amaiinherentiam
amationiset quodlibet[quod-B]uerbumsignificet
119
23:16:00 PM
have been some connection. However, whetherMaster Guido was influenced by the interpolated version or whether the interpolatorappended parts of Guido's argumentationin the respectivesections,must
sic [sicut] determinatam.
inherentiam
quodammodo
Quodsiita est,iamomnia
nonactionemnelpassionem,
nerbasignificabunt
sed qualitatem,idestinherentiam.
nos supradictobiectioni.Dicimus
Sed iam his pretermissis
respondeamus
non difquia reuerauerbumet nomenadiectiuumin adiectiuasignificatione
ferunt.Utrumqueenimsignificat
ut albuset
[-antB] remsuamut adiacentem,
nominiset uerbi.Si quis subtiliter
significatio
legit.Nontarnenideoconfunditur
consideret
nominiset principlem
[-iter] significationem
significaprinciplem
tionemuerbi,aperteilla uidebitdifferre.
Propriaenimet principalis
significatio
acsubstantiam
nominisest significare
cumqualitate.Uerbiautemsignificare
tionemuel passioneminherere.
Ecce,in principali
significatione
apertedifferensuamaliquid[quod] nominiaccitia. Quodsipostprinciplem
significationem
nonestinconueniens'
dit quod uerboconuenit,
36For thetextofthediscussionon thesubstantive
verbin theGlosule,I refer
to the textsprintedby Dr Hunt in Appendix,2 (HuntI, pp. 225-228).The
discussion
onthistopicintheCologneversionstopsat line21 ofthetextprinted
ofthesubstantive
on p. 226. The Chartres
MS suppliesan extensivetreatment
forcewithits
its substantival
verb,in whichan attemptis madeto harmonize
see De Rijk,LogicaModern.II, 1, pp. 102-4.)
verbalforce.(Fora fullcomment,
A similaropinionis encountered
in Guido'sglosses: [0, p. 372a] 'Hoc modo
duasdiuersashabetnaturas,alteramex hoc
uidendumest: estistaparsorationis
Ex hoc quod est
quod est uerbum,alteramuero ex hoc quod substantiuum.
est
uerbumactionemuel passioneminherentem
; in qua significatione
significat
equiuocumad omnesactionesuel passiones.Non enimunamtantumactionum
uel passionum,sed omnessub disiunctione,
prediquando est estprincipaliter
'
catum,ut cumdico Socrateses, omnesactionesuel passionessub disiunctione
uel aliquid aHorum.Et hoc
predico,idestuel legituel ambulatuel percutitur
rarousi suntauctores.
uerboin hac significatione
Et hoc
omnesresin essentiasignificat.
Ex hoc autemquod estsubstantiuum
modoacceptumitemest equiuocum,quia ita estacceptumprincipaliter
predicatum omnesessentiasut essentiassub disiunctione
prdicat,ut cum dico
uel homouel asinusuel etiamalbedo.Et ita de omnibus;
'Socrateses, intelligo
maximeagiturab actoribusde est.
secundumquamsignificationem
Et notandumest quod estquandofitterciumadiacens,ex naturauerbinon
actionemuel passionem,sed hanc solam proprietatem
habet tunc significare
estuerbis.
quod unamremalii copulatet retinetex uerboquod solumcopulare
'istaresest
:
in
ut
dicatur
suo
alii
ita
unam
essentiam
esse
uero
coniungit
Quod
uerbum.
ilia*, hoc habetex hoc quod est substantiuum
estalbus' et albedopredicetur
'Socrates
Uidendumestcumin hac propositione
de Socrateet ponaturibi estsubstantiuum
uerbum,utrumaccipiaturin ui subinui substantiui
Si enimaccipiatur
uerbi,
stantia uerbiquantumad predicatum.
cumuerbumsubstantiuum
coniungatresin essentiasua, dicetur:'Socratesest
albedo'
, quod falsumest.
'Socratesestalbus' ex ui predicaSed nota aliumsensumhuiuspropositionis
hocsolumintendithec
uerbi.Ex ui predicationis
tionis,aliumex ui substantiui
hoc dicitquod
ui
substantiui
ex
uero
albedo
inhereat
Socrati,
propositioquod
ilia res que est Socrates,est res alba. Hoc enimsolumdicithec propositioin
sensuquantumad uimsubstantiui:'Socratesestresalba' '
120
23:16:00 PM
be left undecided for the moment. To go furtherinto these doubtless

highlyinterestingquestions- Master Guido appears to be an adherent
of an opinion refutedby Abailard in his gloss on the De inter
pretatione
of the Logica 'Ingredientibus'37- would be to go beyond the scope of
this paper.
The same holds forhis view on the significationof the noun : here,
we can only point to the fact that Master Guido does not yet labour
under William of Conches' misconceptionof qualitas in the definition
of the noun; nor does he show any acquaintance with it.38
One can say that froma doctrinal point of view, Master Guido fits
well into the grammaticalclimate of the early twelfthcentury,with its
typical entanglementoflogic and grammar,as alrady has been pointed
out by Dr Hunt.39
The same can be said of his technique of glossing: here, too, one
finds the usual image of early twelfthcentury grammar. More than
once, Master Guido submits Priscian's words to the syllogisticanalysis
of propositio,assumptio and conclusion Furthermore,we do not find
A conciseversionof thisviewappearsin B, f.22ra/, p. 8: 'Uerbasubstantia exhocquodsuntuerbahabentcopulare,
etexhocquodsuntsubstantiua,
resinsua essentia.Et propter
hocduo[. ] habentunamessentiam
significant
alii copulare.Ex hoc autemquod unamessentiamalii copulant,habenthunc
sensumquod dicunt[-turI?]: 'istaresestilia' [ista]. Et proptertalemsensum
benepossuntponiin eademparteorationissubstantiuum
uerbumet nomensine
..
pronomine
37In hiscomment
ontheBoethiandefinition
and
oftheverb,Abailardmentions
of theverbto denote
rejectsthe viewthatit is the fundamental
signification
actioor passio. The adherentsofthisopinionassertthatverbslikesedeo
, vivo
verbdenoteactioand passio,at leastwhenacetc.,as wellas thesubstantive
verbin a doubleforce:
ceptedin vi verbi.Moreover,
theyacceptthesubstantive
in vi verbiand in vi substantivi,
and in bothcasesit is acceptedaequivoce
[Logica
Fromthetextsquotedabove,pp. 119-20
, ed. Geyer,p. 3461*24].
'Ingredientibus'
it appearsthatMasterGuidowas also an adherent
ofthisopinion.
38Guido's theoryofthe
ofthenounrunsparallelwiththatin the
signification
Glosule,cf. e Rijk, LogicaModem., 1, pp. 522-3; forWilliam'stheoryof
see De Riik,LogicaModern
. II. 1. dd. 221-228.
signification,
39HuntI, pp. 214-5.
40I will onlyquote one
exampleof thismethod,but thereis any amountof
themin thegloss:[., f. 17,, p. 38ib] 'diffiniunturautem.[Priscian,XVII
60] Dixit quod pronominadiuersashabentuoces ad diuersasterciaspersonas
discernandas
XVII 58].Et quareipsa hochabeant,nullamadhucinde
[Priscian,
dedit[et quare . . . dedit. ] rationemquod modouultfacere.Summacuius
hoc est: Pronomen
in omnibuspersoniscertomodosignificat,
argumentationis
sed hocin terciapersonanonpotestfacereuna uoce; ergodiuerseuocesinuente
suntpro pronomine.
Huius sillogismi
hauel propositionem
argumentationem
bemusin principio
uersushisuerbis: Diffiniuntur
; assumppersone
pronominum
tioest: Ne una uocesignificentur
nonponitassumendiuerse,
quia assumptionem
do, sed subiungitearnconclusioni
uelutcausamunde sequaturconclusio'.
121
23:16:00 PM
in Ms commentsthe abundance ofillustrativequotations fromclassical

and contemporaryauthors we do in some grammatical tracts of the
second part of twelfthcentury.41The quotations used are taken from
Priscian's Institutiones,and they play a negligiblerle.42
I.2.6. At present,any identificationofthis Master Guido witha grammarian of that name known to us fromelsewhereremains speculative.
There is a Master Guido who taught grammarat Chartresin the early
twelfthcentury,but nothingfurtheris known about him.43Dr Hunt
has very tentativelysuggested Master Guy of Langres as one of the
M. G.s referredto in the Note (esp. in the firstsection of the Note).**
However, an essential divergence in opinion on the substantive verb
excludes this ratherobscure master fromfurtherconsideration.45
A serious candidate would, I surmise,be Master Guy d'tampes, who
was a teacher and magisterscholarumin several schools in England and
France for many years, before he succeeded his master Hildebert de
Lavardin to the see of Le Mans in 1125. He was a famous teacher in
his time, and as is evident froma quotation in the Glose 'Promisimus',
he belonged to the intellectual "in-crowd" of the time.46
II.
Twelfth Century Theories on Government
It is generallyaccepted that it was particularlythe grammariansof

the twelfthcenturywho developed a systematicdoctrineon regimen.47
41HuntII, p. 29.
42I have onlycomeacrossone quotationfroma mediaevalauthor,but I could
it or its author.Ad Priscian,XVII 46 (the discussionon the difnot identify
ferencesbetweenaliquis/ullus,alicubi/usquametc.),MasterGuidoquotesthe
exampleof the ringof Lord Landric: [O, p. 377a] 'sin autem dcam ullus
numero
idestex omnibusqu numerantur
pendet ex uno (!) infinitoomnium,
dicoinfinito,quasidicit:Omnesresaccipiotamhomines
quamdoslapidesetc.
ullusdeorum,ulluslapidumnonsedetin anulodominiLandrici
Ullushominum,
tam congruesicutiacinctus'
43Cf.A. Clerval,Les colesde Chartres
au moyen-ge
, Paris 1895[Repr.Frankfurta.M. 1965],PP-174-544HuntI, p. 210; A certainMasterGuidois referred
to
to m the ouppletwnes
in theMS Orlans,Bibliothque
262a,
266,
theDe interpretation,
p.
municipale,
and 'locus difThe Doctrine 'maximapropostilo*
cf. N. J.Green-Pedersen,
"
*in Commentaries the12th of onBoethius'
Topics", in: Studia
Century
from
ferentia
18 (1977)*PP-125-163,esp. p. 125,n. 3.
Mediewistyczne,
45 For Guyde Langres'viewsofthesubstantive
verb,cf.HuntI, p. 224.
46Cf.HuntII, p. 41 and n. 4.
47 Cf. J.Golling,Einleitungin die Geschichte
der lateinischen
Syntax,m: G.
Vol. III, Leipzig1903,
derlateinischen
Grammatik
Historische
Sprache,
Landgraff,
Syntaxund Stylistik,
pp. 239-46:J.B. Hofmann,Lateinische
pp. 28-9; Thurot,
neu bearb. von H. Szantyr,Mnchen1965, pp. 8-9; and G. L. Bursill-Hall,
oftheMiddleAges,The Hague i972>p. 36.
Speculative
grammars
122
23:16:00 PM
This doctrinewas not merelya descriptiveor normative one, but was

based upon a theory. On several occasions, twelfthcenturygrammarians emphasized the differencebetween the treatmentof regimenon
the elementarylevel, called the puerilis disciplina / doctrina, and their
reflectionson this linguisticphenomenon,which would result in a real
,48
theoryof regimen
As has already been pointed out in the introduction,many aspects
of the development of this theorystill remain obscure, and our knowledge of it is mainly dependent on the texts printed by Thurot.49The
rediscoveryof Master Guido's glosses, although in mutilated form,
on the Priscianus minor supplies us with an important text in the
development of linguistic thought in the early twelfthcentury. The
views he held on regimenwill particularlyattract our attention: they
representan early stage in the reflectionson governmentby the mediaeval grammarians.
In the followingsections, I intend to trace brieflythe position of
Master Guido in this development by analysing his own views on this
topic, and by comparingthem with the opinions held by other twelfth
centurygrammarians:William of Conches,the anonymousglossatorof
the gloss 'Licet Multi in Arte*
, Peter Helias, Ralph of Beauvais and
Robert Blund.
II. I.
Master Guido
Master Guido attached his views on regimento the comment on

Prscian, XVIII 4-5, where Priscian treats the three kinds of constructions according to the doctrineof the Stoics.50Guido starts his reflection with the statement that the use of the term regeredoes not rest
on any auctoritas, but that it is in fact a metaphorical usage ; it equals
the exigereof Priscian's Institutiones.In order to explain this usage,
Guido uses the simile of a man who is lost and needs guidance from
48Cf.HuntI, p. 214; Guido
[H, f.40va]:'Cumque etiam Priscianusdicat[ur]
quod nominatiui
exiguntobliquoset interexemplaponat'bonussumanimam*
,
, quare pocius concedunt,quando dicitur'filiussum . . . .'
'fortissum uirtute'
uel et 'commodus
sumduci*
, quod hic nominatiui
regantobliquosquam ubi est
nominatiuus
[accusatiuusMS] cum accusatiuo[nominatiuo
MS] uel ablatiuo
nullamutique uidemusrationemnisi quod sequnturusumpuerorum';[H, f.
40vb]: 'Et sicutin expositoria
regi<tur> ab esseinfinito,
quod estcontrausum
Nos autemet ab
puerorum,
qui dicuntnominatiuos
regia uerbispersonalibus.
infinitiuis
eos quandoregiconcedimus,
sensumquam usum*.
pociusattendentes
See also thetextquotedbelow,p. 133.
49 Thurt,
pp. 230-46.
60Ed. Hertz,II, 21119-27.
p.
123
23:16:00 PM
somebody else in order to returnto the rightpath. In the same way,

Guido continues,most words used in a constructionneed to be joined
to other words which they are governed by, in order to remove the
uncertaintythat exists about theirmeaning:
[H, f.40ra"b]'Nota quando in usu nostrodicimusquod dictioregitaliam
uel regiturab alia, persimileloquimur,
dictionem
nonuerbumauctoritatis
Sicutenimaliquiserransinuiaeget
ut ad uiam
alicuiusregimine
proferimus.
redeatet peruiamcertuseat,sic dictiones
posite
plerequein constructione
per aliquamincertitudine<m>quam habentcircasuam significationem,
dictionum
idestab
a quibusregantur,
[f. 40rb]egentaliarumconiunctionem
illa incertitudine
et certum[-am MS] quid
quam habent,remoueantur
in
monstrentur.
Auctoritas
quidem,ut dixi,nec regitnecregitur
significare
hac [lacuna],sed equipollenter
dicitquod una [uia MS] dictioexigitaliam
ut hichabemus,etunaconstruitur
cumalia etrefertur
ad aliam'.
dictionem,
Next, Guido examines which words can govern, and which can be
governed:
A. The nominative and vocative cases are intransitivelygoverned by
the verb;
- an objection that it is the nominative case that governs the
verb, is refuted-;
B. The oblique cases can be governed by:
1. the transitiveverbs and theirparticiples;
- an objection that it is the oblique case which governsthe verb,
is refuted-;
2. the nominativecase of a noun or participleand by theiroblique
cases;
- an objection that it is ratherthe oblique case which governsthe
nominative is refuted-;
3. an adverb; master Guido uses the adverb similiterand the ex'
. The class of adverbs that can
ample similiterhis loquitur*
govern is restrictedto those adverbs that derive the property
of governingfrombeing a derivativeof a noun that governs a
particular case, e.g. similis + dat. ;
4. the prepositions;
- here,too, the objection is made that it is the oblique cases which
determinethe prepositions,but this view is also refuted.
A separate category is formed by the casus absolti;51 they are
61For a modernlinguistic
cf.L. Hjelmslev,
viewon thecasusrectiand absolti,
La catgorie
des cas, Copenhague2
1935-37[Repr.Mnchen1972],p. 22: 'Mais,
des Anciensest rellement
dimension
faitedu vocatif,la premire
abstraction
124
23:16:00 PM
"
absoltia regimine*, and thereforenot incorporatedin the system of

governing.Their most importantrepresentativeis the ablativusabso
lutus, but the Middle Ages also accepted a nominativusabsolutus,a
genitivusabsolutusyetc., althoughalways with the meaning of absolutus
52
a regimine.
Master Guido appends some special cases or dubitabilia to the dis. We will confine ourselves to the firstone:53
cussion on the regimen
The question arises as to which part of speech governs the accusative
'
*
nos in the construction bonumestnos hic esse (Mt. 17,4). Four answers
are proposed:
1. Some people argue that the accusative nos is used in an absolute
way, so that it is not governed by any part at all; this solution is
rejected, since it is not sustained by any authority:
accusatiuum,
[H, f. 40vb]'Dicuntquidama nulloregiibi,sed esseabsolutum
sicutablatiuiinueniuntur
sit,
Quodlicettolerabile
sepeabsoltia regimine.
tarnenquia auctoritasnonhabet,hoc nos recipimus'.
2. In the second answer,bonumis proposed as the governingpart, as it
'
'
is in the construction bonus sum animam ; however,Master Guido
does not understandhow nos would assume more "certainty" from
the nominativebonum:
ut supra
? A bonumnominatiuo,
'H, f. 40vb]'Dicunt alii: A quo regitur
'. Sed non
animama bonus,scilicetubi [ut MS] dicitur*bonussumanimam
eumaliquo
illerespicere
uideturnominatiuus
nosaccusatiuumcertificando
modo'.
3. Other grammarians accept nos hic esse as an aggregate; but the
problem still remains, since the accusative case is not an absolute
one, and still requires a part to govern it :
[H , f.40vb]'Finguntaliinoshicessehoctotumquasi ibi esseinilia oratione.
Sed siue ita sit siue non,resttquestioeademcumpredicta.Nos enimet
esseconiunctasuntoratioinperfecta
et utrumqueest dictio.Nos, scilicet
obliquus,uerocumnonsit absolutus,requiritaquida quo regatur'.
au pointque toutcas est conucomme indiffremment
dpendant
supprime,
Toutcas peuttrerectus
alexandre de villedieu
ou indpendant.
ou absolutus.
et l'ablatifpeuvent
le gnitif
nousditexpressment
l'accusatif,
que le nominatif,
c'estreconse trouver l'tatrgiou l'tat absolu.Faire cetteconstatation,
et
natrele faitcapitalque toutlmentflexionnel
est la foissyntagmatique
See also below,p. 131,. 63.
paradigmatique'.
62H, f.40va: 'dicuntquidama nulloregiibi,sed esse absolutumaccusatiuum
sicutablatiuiinueniuntur
sepe absoltia regimine';cf. Thurot,
pp. 246-8and
below,p. 131,n. 63.
63The otherquestionsconcern'uideoressebonus'and 'egofiorethoricus'.
125
23:16:00 PM
4- Finally, the opinion is expressed that nos is governed by the infinitive esse. This solution cannot be accepted without some modification, since the objection was made that it would be in conflictwith
the general rule that every mood of a verb has to govern the same
case as its indicative; the indicative mood of'esse, i.e. sum, governs
the nominative case, so esse has to govern the nominative case.
Guido refutes this objection with the argument that this rule is
only applicable to transitiveverbs. The infinitivemood of absolute
verbs oftengoverns cases differentfromtheir personal verbs:
nos accusatiuumibi regiab
[.H, f.40vl)]'His de causisquidamconfirmant
sed occurrit
ubi
hecregulaquodnullusmodusuerbi[metrus
esseinfinituo,
MS] regitaliumcasumnisieum qui regitura capiteeius. Sum ueroquod
est caputesse,nonregitaccusatiuum;ita essenec debetnecpotestregere
eundem.Ad quod dicimusillamregulamdatamesse de uerbistransitiuis,
non de absolutis.Quem enimcasumregitlego(transitiuum
est), eundem
et cumlgat.Nam sicutdicimus:'legolibrum'ita 'legeet legere
legeet legere
casus
uolo librum'et e conuerso.Infinitaueroabsolutorum
' sepe exigunt
quos non reguntpersonaliauerba eorum <ut> uiuere Uirgilium'et
similia'.
It is interestingto note that Peter Abailard was also acquainted with
thisproblem.He dealt withit in connectionwiththe modal proposition:
'
', ed. Geyer,pp. 492-3]'Soletetiamquaeride materia
[LogicaIngredientibus
cum dicitur'possibileest Socratemesse vel sedere',a
subiectaeorationis,
quo videlicetaccusativusille regatur.Nullusquippe moduscasumregit,
nisi quem regitindicativus,a quo nascitur[noscitur
Geyer].'Sedeo' vero
sive 'sum' accusativumnon regit,quarenec sederevel esse. Sed profecto
sciendumestillamregulamessede casibustransitive
iunctis,nonintransitive, ut si dicamus: 'do tibi librum','do' transitivesubiunctiscasibus
coniungitur
ideoque ceteriomnesiisdemcopulantur.At vero 'esse' vel
'sedere'cumSocrateintransitive
iunguntur
ideoquenilad regulam'.54
54For thistext,see also L. Minio-Paluello,
Century
Logic, II AbaelarTwelfth
diana indita,1. SuperPeriermeneias
XII-XIV, Roma 1958,[32] p. 21. It is
ofthesecond
in theirrefutation
to noticethatthelatergrammarians,
important
'noshic
madeuse ofthe construction
solution(i.e. nosis governedby bonum),
'Si
essepossibilees, cf.WilliamofConches[MS Paris,BN lat. 15130,f. 127]
dicent: 'bonumestnos hic esse' huneaccusatiuumquis regit?, dicunt: bonum
' Sed nobisnon
si dicamus:'isteestbonusanimam
regitiliumquemadmodum
nisiubi aliquidpersuam
non
soletregereaccusatiuum,
adiectiuum
placet,quia
hie dicerent:'noshic
toti.Sed <si> hoc ita soluerent,
quid
partematribuitur
'
essepossibilees ?'; and thegloss'LicetMultiin Arte
, Bodl.Canon.
[MS Oxford
ab hocnominebonum,
misc.281,f.64r]'Quidamdicuntquod regitur
quod alias
non
esse
'iste
animam
autem
ut
est
bonus
'.
Hoc
accusatiuum,
posse
regit
Priscianusaperteostendit,ubi dicit a nominehuiusmoditunc tantumregi
accusatiuumcasum,quando per nomenadiectiuumquod toti conuenitper
attribuitur
totiperpartemquod inestparti.Si tamenquocumappellationem,
ubi nos
que mododicaturibi esse,nullomodopossetin hac alia assignatione,
est accusatiuicasus 'nos hic esseestpossibile''.
similiter
126
23:16:00 PM
Guido's theory about regimenclearly illustrates the entanglementof

logic and grammar in early twelfthcentury linguistic thought. His
point of departureconsists of observationson a syntacticphenomenon,
namely the occurrence of several cases in a construction,and he intends to give an explanation and justification for their appearance
together.From a syntactic point of view, regeredenotes for Guido a
sort of dependency relationship or in modern linguistic terminology,
a syntagmatic relationship55 between two words or aggregates in a
construction(the regens: id quod regit,and the rectum: id quod regitur),
in which the rectumis put in a particular case (nominative or oblique
case) by virtue of the regensin orderto denote the relationshipto that
regens, and the kind of the relationship.He does not express this view
absolutely explicitly,but we may derive it fromhis text:
a uerbistransitiuis
interdum
et ab
[H, f.40rb]'Obliquietiamregidicuntur
eorumparticipiis,
tunc scilicetquando in constructione
positisignificant
resin quas transitactus f unumquemque
| illatus,uel quas respicitactus
uel possessio[passioMS, sedcorr.MSC]sinetransitione;
quod fitin adquisitiuaconstructione
et intransitiua. .
However, in order to expound this phenomenon and to develop his
theory, Guido resorted to logic, and introduced a semantic-logical
componentinto his reflections: the determination
The materialfora mediaeval theoryabout determinatio
was supplied
in
Boethius
his
De
divisionibus.In this tract, Boethius points out
by
that thereare three separate ways of dividingthe voice into its significations:
divisionetrac[De div.,PL 64, 888D] 'Resttut de vocisin significations
temus.Fit autemvocis divisiotribusmodis.Dividiturenimin significationesplures,ut aequivocavel ambigua. . . Alio autemmodosecundum
sed multismodis. . . Alitervero
modum;haec enimnonplurasignificane,
modussecundumdeterminationem
According to Boethius, the functionof determinatiois to remove the
doubt {dubietas),that exists in the mind of the hearer as to the meaning
of a word. He implicitlyintroduces two kinds of determination: one
concernsonly one word, the other concernsimperfectsentences.57
56Cf.Hjelmslev,op. cit.,p. 22.
6e Forthenotionof"determination"
inmodern
see H. Happ,
research,
linguistic
einerDependenz-Grammatik
des Lateinischen,
Grundfragen
Gttingen1976,pp.
130-131.
67PL 64, 889A-B: 'Quotiesenimsinedeterminatione
diciturvox ulla,facitin
intellectu
ut esthomo,haecenimvox multasignificai
. Nullaenim
dubitationem,
diffinitione
conclusaaudientisintelligentia
multisraptaturfluctibus,
erroribus127
23:16:00 PM
Boethius' discussion of determinationinspiredAbailard to deal with

this matterin the fifthtract of his Dialctica in a rathercomprehensive
way. Abailard explicitly makes the distinction in the determinatio
dictioorationumand the determinationes
between the determinationes
num, The firstcategory can be divided into determinationsof uncertain sentences and determinationsof both uncertain and ambiguous
sentences:
[Dial. V, ed. De Rijk2,pp. 586-7]'Sunt autem alie (sc. determinationes)
alie dictionum.
Orationum
autemut earumque dubiesunt,sive
orationum,
sivenon.equeenimomnedubiumestambiguum,
scilicetambiguefuerint,
sed omneambiguumdubium. . . Ea namquedubia tantumambiguasunt
diversosintellectus
ut nichilextrinsecus
plenecontinet
quorumconstructio
. . . Sunt autemet alie dubiein
supplendumsit ad sensusperfectionem
extrinsecus
ad sententie
sueperfectionem
quibusipsumproferentem
' aliquid
'
oportetsupplere,ut si dicam da michi','currecito quid tribuipostulem
ut
iubeamdeterminandum
aut quo currere
relinquitur,
aliquosupplemento
'
videlicetet illudapponamde quo fieridonumvolebamdicens'librum
dictionum
The second category of determination,the determinationes
,
in
occur
have
a
that
words
concerns
They
multiplexsignificatio.
only
several forms.Abailard sums up the list given by Boethius : secundum
casus, secundumaccentum
, per adiectionemgeneris/numeri/
diffinitionem
. These determinationsremove the uncerand secundumortographiam
an uncertaintythat was caused by equia
from
word,
single
tainty
vocation or by a shared externalform.In fact,they could be considered
as explanations of the determinedwords.
dictiNext, Abailard passes on to another kind of determinationes
onum. They take two forms: the determinantis joined to the determined either transitivelyor intransitively:
[Dial. V,ed. De Rijk,pp. 589-90]'Suntetiamunarumvocumquedamdeteralie intransitive
minationes,
quarumalie transitivehis que determinant,
'domusSocratis ad aliud
ut
'video
transitive
'
lupum
quidem
iunguntur;
transitquam determinadeterminatio
enimdesignandum
que subiungitur
demonstraAlianamquepersonaper'video', alia per'lupum*
tumsignificet.
'
'
turatque alia resper 'domus'
, alia per Socratissignificatur.
autemdeterminatio
Intransitive
quando circaeiusdemrei
coniungitur,
tunc enimnon dicitur
determinatur
ei
coniungitur;
significationemquod
. . . Atquehec quidemdetertransircumin eademremanetsignificatione
consisti.
sed in quadamrestrictione
minationonin expositione,
ubi id quoddicensloquitur
Quodenimquisqueauditorintelligat,
que traducitur.
dicens: Omnishomo
? Nisienimquisita diffiniat
concludit
nulladeterminatione
ita
nomine
si
hoc
et
homo
certe
aut
continget,
designet,
ambulat,
ambulat,
quidam
audientisquod rationabiliter
intellectus
intelligatnon habet.Suntetiamaliae
nisi
ut si quis dicat,detmihi,quod daredebetnullusintelligit
determinationes,
et certaratiodeterminationis
intellectus
addatur,vel si quis dicat,ad me veni,
non cognoscitur.'.
quo veniatvel quando veniatnisideterminatione
128
23:16:00 PM
In Guido's discussion of regimenthe notions of uncertainty[incertitudo) and determinarealso occur, but the positions of the determinant
and the determinedare interchangedwith respect to Abailars theory.
Whereas in Abailars view lupum is the determinateof the verb video,
Guido states that the accusative lupum is determined and removed
from uncertainty by its regens, i.e. video: in his theory regens and
coincide.
determinans,rectumand determinatum
This transposition was doubtless suggested by the metaphorical
usage of regereinstead of exigere. In its usual meaning, regereassigns
the leading position to its subject, the position of being led to its
object. From this point of view, if thereis talk of any uncertainty/itis
the object rather than the subject (the regens)of regerethat is uncertain. This way of thinkingis illustrated by Guido's explanation or
ratherjustificationof the metaphor.58
Guido saw the objections that immediately resulted fromthis particularadaptation. Thus he was forcedto admit some determinationby
- in fact the ordinarymethod of determination
the rectumon the regens
in Abailars system ; the governingposition of the regenscould only
be maintained by a less convincingappeal to a higherdegree of determination by the regens:
determinan
uerbaperobliquos,quia nisi
[H, f.40rb]'Si quis dicatsimiliter
opponantur
obliqui,incerta[etiamcertaMS] suntuerbain quid transeat
actussuus,concedimus
ita [utiMS] quodammodoesse. Sed quia agentis
personafrequentius
significata
peruerbum,digniorestpacientesignificata
per obliquum,ideo pleniuset certiusuerbumdeterminat
ipsumobliquum
quam obliquumuerbum;quare recte uerba exigereet regeredicuntur
obliquos';
esse incertosnisi additione
'H, f.40va]'Sed dicitaliquisita nominatiuos
certificentur,
obliquorum
quodconcedimus
quodammodo.Sed magisincerti
sunt obliquiet magisegentdeterminatione
quam nominatiui;illi et res
dsignantobliquomodo,istidecreto,ideo [ut MS] meliusdicuntur
regere
obliquosquam obliquinominatiuos1.
Of course, Master Guido could not have relied upon the text of Abailars Dialctica, the firstversion of which appeared in about 1118.59
We may assume that Guido's glosses were already composed at that
time. Moreover,such an elaborated theoryof determination,in which
the perfectiosententie60is already introduced, is in fact incompatible
with Guido's problematic adaptation of the theory of determination,
58 For thetext,see above,p. 124.
59PetrusAbelardus,Dialctica
, ed. L. M. de Riik, Assen2IQ70,p. XXII.
60Cf.above,p. 128,and below,p. 136.
129
23:16:00 PM
probably due to the fact that his theory was still in a relativelyearly
and primitivestage.
II. 2.
William of Conches
A comparison of Guido's discussion of regimenwith those of other

twelfthcenturygrammariansreveals a strikingresemblanceas well as
essential differences.
William of Conches, the famous grammarian,who taught grammar
at Chartresand Paris ( ?) in the second quarter of the twelfthcentury,61
also discussed the problemsof governmentin his commenton Priscian,
XVIII 4-5, too.62The arrangementof the text is, generallyspeaking,
the same as in Guido's glosses:
A.
An introductionto a more recent use of the termregere:

auctoremobli[MS Paris,BN lat.15130, f. I27ra]'Hic habemusPriscianum
quimimodoa nomine,modoa uerboexigi.Sed hoc quod modernidicunt
a dictioneregi,equehicequealibiinuenimus.
dictionem
Quoniamtamen
- nonenimmeliores
sumus
nostri,
quibusnonfasestcontradicere
magisti
quam patresnostri- , hoc dixerunt,quid nobis de regiminedictionum
ostendamus'.
uideatur,breuiter
B. The division of the words into governingand non-governingwith

an explanation and justification:
alie nonregunt,sed
[Ibid.] 'Dictionumigituralie reguntet nonreguntur,
et regunt,
alie
sed reguntMS], alie et reguntur
[nonreguntur
reguntur
regunt.
reguntur
Soli
Nominatiuiet uocatiuiet uerbumita reguntquod non reguntur.
obliquiregihabent,qui perse predicali subicihabent,sed ex ui et
enim
naturaaliorumhabentponiin orationeet nonperse. Quemadmodum
semi dicunturregi a domino,non dominusab eis, ita obliqui casus a
dicuntur
et uerbotranslatiue
regi.
partibus,idesta nominatiuo
dignioribus
non
Sed nominatiuus
quia per se et non alio exigenteuenitin orationem,
et uerbaet aduerbia.Obliquiuerocasuset reguntur
diciturregi;similiter
ab aliis semperet quandoqueregunt.Aduerbiaet prepositiones
regunt;
regunt reguntur.
coniunctiones
Solisigiturobliquisconuenitregi.Ab ilio autemregitur
obliquus,natura
uel ponitur.Sed obliquus
cuiusexigentetalis obliquusuenitin orationem
aliquandoregitura nomine,ut 'isteestfiliusSocratis',quia pronomen
uerbumestcausa quareistegenitiuus
ponaturin orationehac; aliquan61 ForWilliamofConches,
Medieval
cf.Jeauneauio ; seealso R. W. Southern,
Humanismand otherStudies
, Oxford1970,pp. 71-3 and PeterDronke,New
6 (1969),
in: Anuariode estudiosmedievales,
totheSchoolofChartres,
Approaches
pp. 117-140.
62For thetwoversionsoftheseglosses,seeJeauneaui6o. Thecomment
on the
in MS Paris,
Priscianusminoris onlyextantin thesecondversion,preserved
BN lat. 15130.
130
23:16:00 PM
do regitura uerbo,ut 'uidetSocratem'',

aliquandoa participio,ut 'uidens
Socratem
'; aliquandoab aduerbio,ut 'similiter
his') aliquando[f.I27rb]a
ut 'isteamaturab ilio'.Sed obliquialiquandoreguntsecundum
prepositione,
hancregulam:Quemcumqueobliquumregitnominatiuus,
eundemregunt
omnessui obliquitamsingulares
nullum
quamplurales.Sed si nominatiuus
sed nonregen.
regit,sui obliquiregantur,
. A section on dubitabilia.William confineshimselfto the construction 'bonum est nos hic esse'. He lists the four answers, and adds an
interestingrefinementto the last solution. Nos , he argues, is governed
by the infinitiveesse by virtue of the natura omniuminfinitiuorumin
modalipropositione:
ab hocinfinito
esse.Sed illisopponitur
[Ibid.]'Aliidicuntquod regitur
quod
dicitPriscianus
cumquo et suus
cumeo construitur
quod uerbuminfinitum
esse.
indicatiuus;sed sumnon regitaccusatiuumnec ergoistudinfinitum
Sed dicuntquod Priscianusnon dicitquod infinitum
non construitur
nisi
cumeo cumquo et suus indicatiuus,
uerumest igiturquod infinitus
construitur
cumilioobliquocumquo et suusindicatiuus,
sedposteaex communi naturaomniuminfinitiuorum
in modalipropositione
habethuneregere
accusatiuum
etinfinitum
absoltiuerbi,ut 'Socratem
uerumest'.
legere
An importantimprovementon the theoryput forwardby Guido is
the mutual separation of regimen and determination. Whereas in
Guido's theorythe regenshas to determinethe meaning of the rectum,
the notion of determininghas totally disappeared in William's theory.
In his opinion the regens requires that particular oblique case ui et
natura sui exigente
. In consequence, one does not come across Guido's
simile; William uses the notion of the master and his slaves, a simile
which,unburdenedby determination,fitshis theorywell.
By inserting the nominative and vocative cases into the class of
words that can govern, but cannot be governed, William, probably
inspired by Priscian, XVIII 2,63 took a retrogradestep, for, from a
63Ed. Hertz,II, p. 21011-3:
'Nominativus
et vocativusabsoltisunt,id estper
unam personamintransitive
ut 'ego Priscianusambulo,tu
possuntproferri,
Platophilosopharis,
Aristoteles
oftheabsolutecase
disputt'.The phenomenon
is dealtwithby A. D. Scaglionein hisstudyArsgrammatica,
The Hague - Paris
AlbericofM ontecassino
and theAblativeabsolute.To the two
PPeleventh
witnesses
ofthetermablativus
a masterJohannes
century
absolutus,
( ?)
in MS Paris,BN lat.7505,f.3V[Thurot
one
, p. 87] and AlbericofMontecassino,
couldadd AnselmofBesate,whoin hisRhetorimachia
[ed.. Manitius,Weimar
et dativmet absolutum
1958 2, 4 p. I495"8]uses this term:'per genitivum
ablativumhis a te discedentibus'(cf.theMittellateinisches
I A-B,
Wrterbuch,
needssomecorrection.
Absolutus
5472).However,Scaglione'sviewon absolutus
is not onlythe translationof the Greekterm7cXuTo
, p. 135, cf.
[Scaglione
withanother",
Thurot,
p. 246]in itsgrammatical
meaningof"notin opposition
"initselfsufficient"
(cf.Apoll.Dysc.,G.G.II2, p. 21-2adn.),butalsoofaTOTeXyj
131
23:16:00 PM
modern point of view, he now holds a more conservativeposition that

mars the perfectionof his theory.64The relations between subject
nominative<-> verb, and verb predicate nominative are now no
longer considered: the nominative case (and vocative) enters, as
William argues, the constructionin its own rightlike the verb, adverb,
preposition and conjunction.
'
'
II.3. The Gloss Licet Multi in Arte
The glossator of the anonymous glosses on the Priscianus minor,
inc. : 'Licet multi in arte', which is preservedin the MS Oxford
, Bodl.
Canon. mise. 281, may be placed in the same grammaticaltraditionas
William of Conches.65
In this gloss, the reflectionsof the glossatoron regimenare appended
to the commenton Priscian, XVIII 4-5, too. The discussionstartswith
the traditional remarkson the recent use of the term regere:
Bodl.Canon,misc.281,f.6]'Sciendumest itaque <quod>
[MS Oxford,
alia<m> exigerequod a magisperidemappellato a Priscianodictionem
trisnostrisdiciturdictionemaliam regere.Unde nichilaliud est regimen
a Prisciano
dictionum
Licetigiturhocuerbumregere
quamearumexigentia.
non sit nobistraditum,tamenquia equipollens
in huiusmodiproprietate
in hoc loco inuenitur,
etiamquia a patribus
eius in eademsignificatione
nobisexstiterunt,
doctrinehoc uernostrisqui meliores
causa manifestions
de hac prousu in proprietate
hac nobisest traditum,
bum in frequenti
hoc uerboearndesignantes,
agamus. .
prietatedictionum
. In his opinion, a
Next, the glossator expounds his view on regimen
word is said to govern anotherword,when the nature of the governing
word requires the word governedto be put in the constructionin order
; cf.
, notneeding
any helpfromextrinsecus)
by perfect
(whichmaybe translated
Prisciandidnotyethave
Apoll.Dysc.,G.G.II2, pp. 509 and 510. Furthermore,
=
Thus the conceptabsolutus
any acquaintancewiththe notionof regimen.
is ofmedievalorigin.
a regimine
absolutus
64Cf.Hjelmslev,op. cit.,p. 22: 'Dans ce domainecommedansbeaucoupd'autres,le moyen-gemarqueun progrsdcisif.En tirantparti avec plus de
des faitssyntagmatiques,
on largitl'idede la rection
consquencequ'autrefois
etreconnat
prendsa placevritable
grecque,le nominatif
que,malgrla doctrine
dans la catgoriede la dpendanceet non dans cellede l'indpendance.
(Voir
Dans ses deux rlesprincipauxde sujet et de
Princ. 159,avec bibliographie).
au mmetitre
estrgipar les autrestermesdu syntagme
prdicat,le nominatif
que les cas ditsobliques'.
66 For a description
Tracts
oftheMS,see L. M. de Rijk,SomeThirteenth
Century
Canon,
in
II. TheObligationes
ontheGameofObligation,
Oxford,
found
parisienses
misc.281,in: vivarium,XIII (1975),pp. 22-3; forthepositionoftheseglosses
"
in the grammatical
see mypaper MulierQuaeDamnavit,Salvavit
tradition,
in: vivarium,XIV (1976),
A NoteontheEarlyDevelopment
oftheRelatiosimplex,
-13-5.
132
23:16:00 PM
to explain the meaningof the regens. For, by its very nature,the regens,
used alone in the construction,would keep the listener in suspense.
'
The glossator illustrateshis words with the example accuso socium'.
An objection gives him the opportunityto introduce a distinctionin
the term exigere: a noun requires a verb (or another governingword)
in order to be governed, a verb requires an oblique case in order to
govern. While both the verb and the oblique case require each other,
the verb is said to govern the oblique case, since the verb has precedence in the construction:
[Ibid.]'Illa itaque dictiodicituraliamregerecuiusnaturaexigentealia in
orationeponitur,que regiturab ea, quoniamad eius exigentiam
ponitur.
Natura autemdictionisappellatureius institutio. Cum enimtalis nature
est aliqua dictio,ut non ponaturin orationenisi cum alia dictioneque
eius explanetet per se autem prolata suspensumreddit
significationem
cumea
aliam dictionem
animumauditorisex naturasua ut exspectantem
quam ipsa exigit,talis,inquam,dictiodiciturtalemregere.
proferri
ut nonponaturin
Uerbigratia.Hoc uerbumacusotalisestinstitutionis,
in quem transitauctus eius. Per se
orationeabsque obliquosignificante
ex naredditanimumauditorisexspectantem
autemprolatumsuspensum
tura uerbiobliquumquem desiderathoc uerbum.Similiteromnetransisicqueritur:
ut 'acusosocium
'. Undein puerilibus
tiuumuerbum,
disciplinis
Sociumque parsregit? Acuso. Quare? Quia talis est natureuerbumquod
exigituel regitacusatiuumcasum.Si quis autemdicat: Nonnesic casus
casum| Necesseestenimaliquidponiad explaexigituerbum? f contingit
nationemsue significationis.
Sociumenimnec acuso plane potestaliquid
significare.
: Nomenexigituerbumuel aliamdictionem
sibiadiungi,
Ad quod dicimus
ut regatur;uerbumobliquumexigit,ut regat.Sicut ergoseruusindiget
exigit
regiminedominiet dominusseruitioserui,sic nomenfrequenter
uerbum,uerbumueroex naturasui exigitobliquum.Item.Cumutrumalterumsic exigat,uerbumtamendignitate
constructionis
interpartesprincitamenquia sineipsoperfectus
nonpotest,
sensusgenerari
ptmoptinens,
ex ipso absquealiishabeturperfectus
intellectus
. . . uerbumtanquamrex
et dominuspartiumregerepotiuscasusdiciturquam regiab eis. Uel quod
meliusest: Possumusdicerequoduerbumperse prolatumsuspensum
reddit
animumauditoris
respectucasus,quoniamex naturasui exigit.Et similiter
omnisdictioalia que [quamMS] regitaliquemcasum.Dictio autemque
nonita. Cumenimdicoacuso, exspectatur
accusatiuusab auditore,
regitur,
ei. Si enimaccusatiuusperse proferatur,
nonredditanimum
qui supponatur
auditorissuspensum
respectuuerbi,cumpossitregia participioet a uerbo
sicuta uerbo'.
et prepositione
In a theory that ascribes precedence in the constructionto the verb,
the verb must govern both the nominative case and the oblique case
as well. These considerationsmust have led the glossator to obliterate
the differencebetween the nominative and vocative cases and the
oblique cases in the division of words that can govern and that can be
governed; they are replaced by the parts of speech categories:
133
23:16:00 PM
1.
2.
3.
4.
alia
alia
alia
alia
regit et non regatur: uerbum, aduerbium, prepositio;

regituret non regit: pronomenet nomen proprium;
regitet regitur:nomen (non tamen omne) et participium;
nec regit regitur: coniunctio et interiectio.6
The question of the regensof the accusative nos in the construction

'bonumestnos hic esse' is also discussed by the glossator.67He supplies
the four traditional solutions, and turns out to be an adherent of the
last one as are Master Guido and William of Conches. The obligatory
objection he refuteswith a still more refinedargument,as William did.
A threefolddistinctionis made in the way a verb can govern: 1. ex
natura uerbi,2. ex natura modi,3. ex naturapersone.The firsttwo were
already present in William's discussion. The last one, the regimenex
natura persone,is needed by the glossator in order to re-establishthe
relation of governmentbetween nominative and verb which William
denied. An instructiveapplication of this distinctionto a modal proposition is given:
esse hominem est uerum
diligo Socratem quem
ego
1I
t
t
II
i
modi
modi
persone verbi
dicuntnosregiab infinitiuo
esse.Si autem
[.[bid., f.4r]'Qui meliussentiunt,
opponaturcontrahoc quod Priscianusdicitnullummodumuerbiregere
dicendumest: Priscianus
aliquemcasumnisiquem regiteius indicatiuus,
de ilioregimine
casuumquemhabentuerbaex natura
hoctotumintellexisse
uerbipreterquamuerbareguntcasustamex naturamodiquam ex natura
casusa uerbocumquo uerbumconstruite
persone.Ex naturauerbiregitur
illa quam contrahunt
ex sua institutione.
Undein omnisuo
ex proprietate
ex hac promodo talemcasumregit.Cum nulloitaque casu construitur
ex eadem
nisiindicatiuuseius cumeodemconstruatur
prietateinfinitiuus
Ex naturapersoneregituerbumcasumillumquemomneuerproprietate.
bumeiusdempersonein omnimodohabetregere.
Item.Ex naturamodiregitillumcasumquemomneuerbumilliusmodi
ex naturamodiregereaccusat temporef regit.Habet itaqueinfinitiuus
et absolutis.Licet enimcum
tiuumquod maximeapparetin substantiuis
ex naturasua construantur,
infinitiui
tameneorumaccusasolisnominatiuis
tiuos[auctsMS] reguntut in modalibuspropositionibus
apparet,ut 'me
' Omneshii modiin hac constructione
esse9 uel 'uiuereestbonum
possunt
9'
estuerum
quemessehominem
assignari'egodiligoSocratem
II.4. Peter Helias
As has already been suggestedby Dr Hunt 68and recentlydecisively
MS Oxford
, Bodl.Canon,misc.281, f.63v.
67 Otherquestionsdealtwithconcerntheconstructions
: 'uideoressebonusand
'mediocribus
essepoetis*.
68Hunt[, p. 214.
134
23:16:00 PM
proved by Karin M. Fredborg in her detailed study of William of

Conches, Peter Helias and their relations,69Peter Helias was highly
indebted to William, but he also proved not to be a mere slavish
imitator.This also holds good for Peter's theoryon regimen.
At the end of his Summa on the Priscianus minor, Peter appended a
,70 although like the other previously mentioned
chapter on regimen
grammarians,he also touches upon this matter in other parts of his
work.
The chapter starts with the usual explanation as to the use of the
termregere; therefollowsan interestingnote on the views held by the
"antiqui". These grammarians held the opinion that dictionemregere
iictionemmeans that a word takes anotherword with it into the constructionin orderto determineits own meaning. In opposition to this
view, which he considered would lead to absurd consequences, Peter
defended the belief that the final cause of regimenwas the perfectio
constructionis.
The regenskeeps the listenerin suspense; this feelingis
removedby the rectum
, not by determiningthe meaning of the regens,
however,but by completing the perfectionof the construction.71
We have to bear in mind the differencebetween the constructio
perfecta and the constructiosensumperfectum
generanshere. According to
'
Peter, the locution Socrates accusa is a constructio
, but it is
perfecta
not a constructio
perfectumsensumgenerans.That predicate is reserved
'
forthe sentence: Socrates accusai + object':
[MS Paris, Arsenal,yn, f. 135],'Antiquienimaliud addideruntin dedicentessic 72: Constructio
est congruadictionum
scriptionconstructionis
ordinatio.
Sed quoniam[quomodoMS] hocconuenit
de
locutioni
imperfecte,
hicagere,sedtantumdeperfecta,
sensum
adduntperfectum
qua nonintendit
in animoauditoris.Sed quia hoc nonpotestconuenire
omnipergenerans
'
sensumin
fectlocutioni,
quia hec Socratesaccusanongnrtperfectum
- queriturenimquemaccuset- , et tarnenperfectaest,
animoauditoris
ideoadditurquantum
in ipsa est'.
In his discussion of regimen, Peter in fact alludes to this difference.
69Fredborg
1973.
70 Thurot1869,pp. 239-43containsextensiveextractsfromPeterHelias' disIn theMS Paris,Arsenaly11 (- Thurot'sMS N, cf.Thurot
cussionon regimen.
fillstheff.I77va-i78vb.
1869,p. 22),thechapteron regimen
71Thurot,
pp. 241-3.
72Cf.MasterGuido'sdefinition
ofconstructio
[B, f. ira; O, p. 369a]'Constructio
est congruaordinatiodictionumperfectum
sensumgeneransquantumin ipsa
es; cf.alsothedefinitions
givenby WilliamofConches[MS cit.,f.85rb]'Conin compositione
structio
and
dictionum
orationis
coniunctio',
igiturestregularis
inthegloss'LicetMulti'[MScit.,f. iv] 'constructio
regularis
igiturestdictionum
in constructione
orationisconiunctio'.
135
23:16:00 PM
First, he only brieflydescribes regereas: traheresecum earn [sc. the

ad constructionsperfectionem.
word that is governed] in constructione
But, some lines furtheron, he mentions a distinctionin the perfectio
constructionis
between the simpleperfectioconstructionis
and the maior
.
constructionis
Peter
work
out this
does not
perfectio
Unfortunately,
distinction,which could have led to a discussion on regimenrelations
at a primaryand a secondary level. For, when we closely examine the
text in question, we can see that in the case of the regimenrelation
between verb and mere subject nominative, Peter speaks of the per; in the cases of the other regimenrelations, the
fectio constructionis
notion of maior constructionis
perfectiois used:
'Sed ut breuiteret ueriusdicam: dictionemregere
[Ibid.,ff.I77vb-i78rb]
nichilaliud est quam traheresecumeam in constructione
aliam dictionem
ad constructionis
... [f.178]. . . Undeuerbumexigitnomiperfectionem.
natiuumcasum,quia ad perfectionem
constructionis
trahitsecumin conut cum dico 'lego',designoactumlegendialicui inesseet ita
structione,
sermonem
fieride aliquo. Nominatiuus
id de quo
uerosignificat
significo
fitsermo.Quareilluduerbumtrahitsecumnominatiuum
in constructione
...
ad perfectionem
constructio
constructionis,
quia aliternoneritperfecta
[f.I78rb]Cumenimdico 'homoalbuscurri,nontrahithomosecumalbus
ad perfectionem
immoad determinationem
constructionis,
significationis,
huiusnominishomonimisestuaga. Ideoque determinate
quia significatio
illiusadiectiui;et hocuerbumcurrit
adiunctione
exigitalbus,quia si dicereturtantum'homocurri
, uaga essetlocutio.Ideoque ad determinationem
[Thurot
] [maioremconstructionis
MS] necesse
significationis
perfectionem
est addereipsi substantiuoadiectiuum.Uerbumautemnon tantumregit
nominatiuum
casum,sed etiamobliquum.Quandoenimuerbumsignificat
actumtranseuntem
in alium,tuncuerbumin constructionem
trahitobli'
quum casum,ut cumdico Socrateslegi,ostendoactuma Socratetransir.
Sed semperpendetanimusauditorisex re ut significatur
per obliquum.
Queritenimsemperquid legit?, et ideo ad maioremconstructionis
perfectionemtrahithoc uerbumsecum accusatiuumcasum,ut 'Socrateslegit
Et tarnenperfectaeritconstructio,
si dicam'Socratesleginon
Uirgilium'.
addendoobliquum,ut legitponaturabsolute
From this passage, it followsthat Peter's maiorperfectioconstructionis
includes Abailars perfectiosententiein the discussion on the determinano dubiarumorationum: A word is required extrinsecusin orderto
: {da michi + librum' This kind of
accomplish the perfectiosententie
determination does not imply any restriction,but only removes a
dubietas and supplies a clarificationof the deter
minatum
The differencebetween Abailard and Peter Helias must be sought
in the position each takes up with regardto this phenomenon.Abailard
is workingfroma logical point of view and tends to favour a theoryof
73For thetext,see above p. 126.
136
23:16:00 PM
determinationthat is in its turna part of a more comprehensivetheory

of signification.Peter Helias confneshimselfto grammar. He tries to
arrangethe several featuresofregimeninto grammaticalcategoriesand
to renderthem by means of grammatical terminology.
Peter finishes his chapter on regimenwith a discussion of some
iubitabilia. The firstone is the well-knownquestion of the regensof nos
in the proposition'bonumestnos hic esse'.The fourby now traditional
answers are given; Peter, too, is an adherent of the last one:
cum hoc uerbo
[MS Paris, Arsenalyii, f. 178]'I taque nos construitur
esse.Ideoquenecesseestdicerequodilleinfinitiuus
regitiliumaccusatiuum,
hac scilicetregula: Quod omneinfinitiuum
ex ui infinitiui
preterconstructionemsui uerbiregitaccusatiuumcasum'.
II.5. Ralph ofBeauvais
The grammarian Ralph of Beauvais, who taught grammar in the
second part of the twelfthcentury,75must also have composed glosses
on Priscian, but this work has not yet been identifiedin the manuscripts.76We are forcedto deduce his views on governmentfrompassing
remarks in his Liber tan, Glose super Donatum and the Verba preceptiva.
Ralph appears to belong to the group of grammarianswho were so
. For
severely reproached by Peter Helias for their concept of regimen
he held the view that it is the significationof the regensthat requires
determiningby the rectum:
[LiberTytan,MS London,M Add. 16380,f. ii5ra] 'Uerbumsubstantiuum
ex naturasignificationis
sue regitnominatiuum
post se positum.Habet
enimtalemsignificationem
que exigitdeterminan
per nomen'.
As to the insertionof the nominative case into the regimensystem,
Ralph took a positionmidway between William of Conches on the one
hand and the gloss 'Licet Multi in Arte' and Peter Helias on the other.
According to Ralph, the verb requires the subject nominative, and
thereforeit governsthis nominative; the same holds forthe predicate
nominative, as we have seen in the text quoted above. An exception
to this rule occurs in the case of the nominative in subject position to
the substantive verb: in such cases Ralph does not accept government
by the verb of the subject nominative:
'uideoresse bonus '
74Otherquestionsconcernthe constructions
' fio bonus*,
' etc.
'misereor
hominis
uolentis
esseepiscopi(epis)' . 'mediocribus
essepoetis
76For his lifeand works,see HuntII; am
his
of
critical
edition
a
preparing
writings.
78HuntII, p. 45.
137
23:16:00 PM
, M Add. 16380,f. I24va"b]'Ad quod

[GlosesuperDonatum,MS London
sic respondemus
rem agentemhoc ipsum
quod omnenomensignificans
trahita uerbo,scilicetpersonali,quod poniturin nominatiuocasu. Hoc
enimexigiturnaturaipsiusuerbipersonali.... Sic ergouerbumconfert
tam nominatiuo
quam obliquoponiin tali casu. Meritoigiturregitutrumque, licetalia et alia ratione.
Non tarnendicoomnisnominatiuus
regatura uerbo,sed solusille nominatiuusqui a uerbotrahitponiin constructione,
ut cumdicitur:'homoest
animai*
homoa nulloregitur,
aliunde
, istenominatiuus
quia noncontrahit
etuerbumei accommodatur.
poniinconstructione,
quia naturaliter
pecedit
Sed hic nominatiuus
animalregitur,
ab ipso uerboponi in
quia contrahit
constructione'.
ibid.,f. I2ivb and MS Tarragona
[Verba
preceptiva,
, Proveno.54, f. 128]
' 'Homo
estanimal'.In hac constructione
que parsregithuncnominatiuum
homo? Nullapars.Quare? Quia hicnominatiuus
noncontrahit
aliundeponi
animai?
[deponiTarrag.]inconstructione.
Queparsregithuncnominatiuum
Hoc uerbumest.Quare? Quia ex hoc uerboestcontrahit
hic nominatiuus
poniin constructione'.
IL6.
RobertBlund
According to the table of contents of his Summa in artegramatica,

preserved only in the MS London, BM Royal 2 D XXX, ff,79ra-94vb
and I03va-i06rb,the English grammarianRobert Blund, who was active in the last quarter of the twelfthcentury,77
devoted his ninth and
last chapter to the regimen.Unfortunatelythe text in the MS ends
abruptly in the sixth chapter De appositione.However, we can catch a
glimpse of Robert's theory on regimenfromthe firstchapter De constructioneet eius speciebus.
Robert interpretsthe term constructioin three ways :
1. constructiodictioniscum dictione;
def.*est congrua ordinatio dictionis cum dictione in oratione cum
exigentia uel determinatione;78
2. constructiodictionum;
def.: est congrua ordinatio dictionumin oratione quas exigentia uel
determinatioconcilit;79
constructio
orationis;
3.
def.: a. oratio constructa
b. congrua ordinatio orationis ex dictionibus.80
77For thelifeand theworksofthismaster,see L. M. de Rij, LogicaModem.
II, i, pp. 22-5 and 255-7; paper The Relatiosimplexin theGrammatical
andEarlyThirteenth
TractsoftheLateTwelfth
in: vivarium,XV (1977),
Century,
pp. 4-9.
78MS dt., f.7Qrb.
79MS cit.,f.79vb.
80MS cit.,. 80.
138
23:16:00 PM
In the firsttwo definitionsof constructio

, which only concernthe junction of two parts of speech, Robert introducesthe notions of exigentia
and determinatio.
Thereforehe feels compelled to explain them.
First he states that there cannot be any constructionbetween two
words without exigentia or determination
in this context exigere and
regereare accepted as being identical:
, BM Royal,o D XXX, f.79rb]'Noneniminteraliquasdictiones
[MSLondon
estconstructio,
nisiintereasdemsitexigentiauel determinatio.
Quidautem
sit exigere,postea dicetur.Interimpro indifferenti
accipiaturexigereet
'.
regere
Regereis definedas : conferrealieni dictioniponi in tali casu et esse eius
patronus. Determinaremeans: aliquam dictionemmodificareet quodam
modo restringere
.81 So regereconcerns a syntactic relationship,determinarea semantic relationshipbetween the words in the construction.
The limits of the relationshipmay be the same, but the relationships
themselvesare on a differentlevel; they can run parallel, but also in
the opposite direction.Moreover,determinare
is a more general notion.
It also concernsthe relationshipsbetween verb and adverb,substantive
noun and adjective, conjunction and conjoined:82
mediante
uerbo
1
i
1
Determinatio
^
!
J*
1- I
J
Socrates est homo i Socrates
1
' 1
L*j
Regimen.
!
1
N
1
I
i
l
et Plato sunt homines
1
IL*J
*
'
[Ibid.] <D>e exigentiaautemdicimusquod quotienscumque
aliqua dictio
aliamexigit,uel earndeterminat
uel ab ea determinatur,
sed
non
e conuerso.
Uerbigratia.Cumdicitur'Socratesesthomo'
, hoc uerbumestexigitnominatiuumprecedentem
et eum modificat
et quodammodorestringit.
Uaga
enimest suppositiohuiusnominisSocratesad quodlibetappositum.
Item.Hoc nomenhomodeterminat
illuduerbumquod uage se habetad
Non autemconuertitur.
Non enimquotienscumquemlibetnominatiuum.
earnregituel ab ea regitur.
Cumenim
que aliqua dictioaliamdeterminat,
dicitur:'homoalbuscurri
hoc nomensub, hoc nomenalbusdeterminat
81MS cit.,f.79rb.
82MS cit.,. 79vb:'Possuntenimtresdictionesita
ut earumquelibet
coniungi,
cumaha construatur,
ut hic 'Socratesesthomo'
. Construitur
enimuerbumcum
ilionominatiuo
et ilium[-udMS] determinat.
Construitur
etiamcum
precedente
Determinatur
enimetiamhoc nomenSocrates
sequenteet ab eo determinatur.
hocnominehomomedianteuerbosubstantiuo'.
139
23:16:00 PM
sed ipsumregit ab e regitur.

stantiuum
:
cumdicitur
Similiter
'Socratesethomo,
Plato sunthomines',
illa coniunctio
hoc
determinat
interposita
nomenSocrates
etipsadeterminatur
ab hocnominePlato.Cumenimdicitur
:
et ad singulari'Socrates',ille nominatiuus
eque se habetad pluralitatem
tatem.PotestenimSocratessolusintelligi
et nonsolusintelligi.
Hoc autem
confusiodeterminatur
:
cum dicitur
coniunctionem,
per
copulatiuam
'Socrateset Plato' . . . Neutramtamenillarumdictionum
[coniunctionum
et a neutraregitur'.
MS] regitilla coniunctio
In Robert Bluns theory of construction,determinatici
, which was
removed by William of Conches and Peter Helias fromregimen
, is reintroduced, but this time not in order to define or explain regimen
(Master Guido) nor as a requirementon the part of the regens(Ralph
of Beauvais) : Regimenand determinatio
have theirown distinctrealm,
save forthe fact that regimenpresupposesdeterminatio
between the two
.
constructibles,and not e converso
III.
Summary
If one takes an overall look at the development of the theory of

, what emergesis a fairlytraditionalframework,viz. the use of
regimen
Priscian, XVIII 4-5 as a starting point, the apologies for the recent
, the introductionof a similein orderto explain
usage of the termregere
this phenomenon,the distinctionbetween governing,being governed
and non-governingwords, and the discussions on the dubitabilia, esp.
'
on the regensof the accusative case nos in the construction bonumest
'
nos hic esse togetherwith the fourrecurringsolutions.
On the otherhand, the theoryitselfwas continuallyunder scrutiny.
Master Guido representsa quite early and even primitivestage in this
development.The notions were not yet clearlydefined,and the way in
which logical matterwas adapted led to trouble and contradiction.
William of Conches freed the theoryof regimenfromthe misuse of
determinatio
, and established the regensas the part that requires the
The nominative and vocative cases were removed
relation.
regimen
fromthe class of words that are governed.
'
'
In the anonymousgloss Licet Multi in Arte the nominativecase was
put back in its place among the words governed. Peter Helias' contribution consistsmainly of his effortsto preventthe re-entanglementof
, and to introduce a more grammatical
regimen and determinatio
of
Beauvais
is a representativeof the antiqui,
terminology. Ralph
who held an opinion that was so stronglyattacked by Peter Helias.
Finally, Robert Blund represents the stage in which regimenand
140
23:16:00 PM
determinatiowere incorporatedinto a more general theoryof construction.

However, the most originalanswer to a regimenquestion was given
by a simple rural chaplain in 1222 :
'Item.Probatusfuitde canonemissae,"Te igitur,clementissime
etc.
Pater*',
Nescivitcujuscasusesset"Te", neca qua parteregeretur.
Et cumdictum
essetei,ut diligenter
inspiceret
regere"Te",
quae parspossetcompetentius
dixitquod "Pater",quia omniaregi.83
Njmegen
Katholieke Universiteit
Erasmuslaan 40-8
83 VetusRegistrum
ed. W. H. Rieh Jones,London1883,p. 305;
Sarisberiense,
cf.N. Orme,EnglishSchoolsin theMiddleAges,London1973,p. 20.
Thanksare dueto E. KellermanB. Ed. M. A. and Drs J. Blomforhelpwith
thetranslation
ofthistext.
141
23:16:00 PM
Vivarium
XVI, 2 (1978)
Wisdom and Eloquence in Nicholas of Cusa* s
Idiota de sapientia and de mente *
M. L. FUEHRER
and eloquence were concepts widely discussed duringthe

WisdomRenaissance. By the fifteenthcentury, the classical idea of
wisdom was rediscoveredand was in the process of being adto
scholastic philosophy.1At the same time, the rhetoricalideal
apted
of eloquence received sustained attention in the oratorical culture of
the humanists.2While these two concepts were undergoingchange in
the cultures of the humanists and the schoolmen,there arose a third
culture which was unifiedin its opposition to both the secularization
of the medieval idea of wisdom and the oratorical emphasis upon eloquence.3 This third culture, which apparently had its roots in the
northerncountries of fourteenthand fifteenthcentury Europe, was
marked by its insistence upon the thesis that wisdom was attainable
and that eloquence was unnecessary
by the layman or non-professional,
in both the pursuit and transmissionof this wisdom. The movement
known as the devotiomoderna, the literarytradition which produced
works like the Ackermannaus Bhmen, and the Rhineland Mysticsare
examples of this third culture.
Nicholas of Cusa, born 1401 at Cues on the Mosel, was exposed to
both scholastic culture,while he attended the UniversityofHeidelberg,
and humanisticoratoricalculture,while he pursued the study of canon
law duringhis stay at the Universityof Padua.4 Despite the fact that
* This paperis based upon an earlierpaper whichwas read at the Second
on MediaevalStudiesheldat JohnCarrollUniversity.
AnnualOhioConference
1 E. F. Rice, TheRenaissance
Idea ofWisdom
, Cambridge
1958,Ch. 1. Cf.E. F.
, in: Traditio,XIII (1957)P-345: "In
Rice,NicholasofCusa*s Idea of Wisdom
and popularidea in flux".
Cusanus*ownlifetime,
sapientiawas an important
2 Cf.J.E. Seigel,Rhetoric
Princeton
in Renaissance
and Philosophy
Humanism,
1968,Ch. i.
3 For an analysisof someof the popularGermanliteraturedealingwiththis
"thirdculture",cf. F. P. Pickering,Noteson Late MedievalGermanTales in
Praise ofDocta Ignorantia,in: Bulletinof the John RylandsLibrary,XXIV
(1940),pp. 121-137.
4 Seigel,p. 217 "Literarycirclesin Padua, Florence,and Naples all included
es. Perhapsthemostadvancedofthethreegroupswas the
notariesand dictator
142
23:16:09 PM
his writingsshow strongtraces of both scholasticismand humanism,5

withrespectto the question ofwisdom and eloquence he took a position
which, as we shall try to argue, amounted to a rejection of both
scholasticism and humanism. Perhaps the cause of his reaction was
partly due to exposure to some of the ideals of the devotiomoderna
which he possibly might have encountered during his early years of
studyor at the UniversityofHeidelberg.6It is certain,however,that he
was deeply influencedby the mystical writingsof Raymond Lull, the
Rhineland Mystics, and especially Meister Eckhart.7 The concept of
doctaignorantiaor 'learned ignorance", which Cusa developed, seems
to be the resultof the influenceof both mysticismand the ideals of the
devotiomoderna*Whetherhe meant to advance the concept of learned
ignorance as a conscious alternative to the scholastic method or oratoricaleloquence in his early study,De doctaignorantia, would be hard
to establish. In 1450, however, Cusa wrote a series of four dialogues,
entitledthe Idiota , which, when carefullyanalysed, indicate his rejection of the humanistic amalgam of secular wisdom and eloquence as
well as the scholastic method of pursuingwisdom.9
It will be the task of this paper, therefore,to argue that Cusa,
through the figure of the idiota or "layman" who functionsin the
dialogues as an embodiment of docta ignorantia, was attacking both
humanism and scholasticismat theirvery core. It will be shown that
Cusa's conception of docta ignorantiawas not an ideal Cusa developed
in an intellectual vacuum, but indeed was an alternativehe proposed
to the two leading intellectualinfluencesof his day.
With respect to the Idiota , only the firstthreedialogues in the series
need be considered.They are the Idiota de sapientia, in two books, and
oneat Padua,wherefairly
well-known
likeAlbertino
MussatoandLovato
figures
Lovati wereactive in both the writingof poetryand the studyof classical
literature".
Cf.RobertWeiss,Lineamenti
peruna StoriadelPrimoUmanesimo
in: RevistaStoricaItaliana,LX (1948),pp. 349-366.
Fiorentino,
5 Cusa'slanguageis thatofscholasticism.
in
Manyofhislaterworksarewritten
dialogueform,a favoredhumanistgenre.
e R. R. Post in his studyof the devotiomoderna
: Con, The ModernDevotion
withReformation
and Humanism,
Leiden1968,arguesthatCusa was
frontation
mostlikelynotexposedto thedevotio
moderna
whilehe was a studentat Deventer.
7 Ernst Cassirer,The Individualand theCosmosin Renaissance
,
Philosophy
NewYork 1963,p. 33.
8 Rice,NicholasofCusa,p. 346.
9 Cassirer,
of boththe
p. 56. "He [Cusa]attacksthefundamental
assumptions
- but he himself
Scholasticand Humanisticconceptsof knowledge
stillspeaks
theLatin oftheSchools".Cassireradvancesno evidenceforthisthesis.
143
23:16:09 PM
the Idiota de mente. The remainingdialogue, the Idiota de staticisex, deals with philosophyof nature and contributesnothingto
perimentis
the problem of eloquence or wisdom. The structureof these dialogues
is simple. There are three protagonists: a Roman orator, a Roman
layman (the idiota), and a philosopherofindeterminatenational origin.
There is also the author, who never takes part in the actual dialogue
but provides editorial transition.10In the firsttwo dialogues only the
orator and the layman appear. In the third dialogue all three appear.
The scene is set in Rome, and although the date of the first two
dialogues is uncertain,mentionis made of a "jubilee at Rome" in the
third,and presumably this refersto the jubilee of 1450, the year Cusa
produced the work.
The action of the work, while it is simple, undramatic and at times
even clumsy, tends to indicate quite definitelythe role each of the
protagonistsis supposed to take. The orator for Cusa seems to represent humanism.11Oratory was always an integral part of Renaissance
humanism.12The reason forthis lies not in any discoveryor recovery
of the principlesof classical oratoryby the early humanists,but rather
is due to the continuityofthe oratoricaltraditionin medieval rhetoric.13
Since the humanists were the successors of medieval rhetoric,it would
followthat Cusa's orator representshumanism. But when we examine
the orator'sconcept ofwisdom as it unfoldsin Cusa's dialogues,we shall
see that it is one no longer typical of the medieval rhetor,but is more
representative of early humanism. Where the medieval rhetoricians
tended to disassociate the question of wisdom from the pursuit of
eloquence, the Renaissance humanistswere commonlymarkedby their
commitmentto synthesizingwisdom and eloquence.14
10The function
ofthedialogue
oftheauthoris clumsy.The variousmanuscripts
into
his awkwardinterpolation
oftenconfusehimwiththe "orator",indicating
thetext.
11H. Gray,RenaissanceHumanism:The Pursuitof Eloquence
, reprintedin
P. P. Wiener,NewYork1968,p. 202:
Renaissance
Essays,eds. P. O. Kristeller,
werereferthehumanists
"Beforetheword'humanist'gainedgeneralcurrency,
and to theircolleaguesbyothernames- sometimes
'philosringto themselves
'orahowever,theycalledthemselves
ophers',often'poets'. Mostfrequently,
tors'".
12P. O. Kristeller,
: theClassic, Scholastic
RenaissanceThought
, and Humanist
Strains,New York 1961,p. 11. Cf.Seigel,p. 173.
13RichardMcKeon,Rhetoric
in theMiddleAges,in: Speculum,XVII (1942),
op. cit.,pp. 12-13.
pp. 1-32.Cf.Kristeller,
14Seigel,p. 179.Cf.Gray,p. 200: "Trueeloquence,according
to thehumanists,
unionbetweenwisdomand style".
couldariseonlyout ofa harmonious
144
23:16:09 PM
The philosopherin the dialogues representsthe peripateticschool,15

and fromthe mannerin whichhe argues,it is clear that Cusa wishes to
have his readersconsiderthe philosopherto be a schoolman. The philosopher stands in clear opposition to the layman, who presumablybeing
unlettered,cannot read the philosophical authorities. When asked by
the philosopherabout a distinctionin the usage of a certain term,the
layman replies flatlythat being a layman, he "does not much regard
words1'.16
Both the philosopherand the oratorbelieve they know what wisdom
is, and they are proud of their knowledge. The layman, on the other
hand, hesitates to declare that he knows what it is. He is sceptical and
humble about his abilities.17One way to interpretthe layman's scepticismis to see it as a representationof the principleof doctaignorantia.
The layman, as will presentlybe seen, serves four basic functions
in the dialogues : i) he proposes a definitionof wisdom usingthemethod
of learned ignorance: 2) he develops this idea of wisdom in such a way
that he reflectsCusa's mysticaltheology; 3) he offersobjections to the
concept of wisdom embodiedin the culture of the humanists as well as
to theiruse of eloquence as the method of pursuingwisdom; and 4) he
shows that the methodologyof scholastic philosophy is inadequate for
the attainmentof true wisdom.
Cusa makes it clear in his reaction to the oratorical tradition that
the search forwisdom requireshumilityon the part of the subject. The
orator, with his pride in eloquent expression, can never realize true
wisdom. And so Cusa has his layman reply to the orator early in the
firstdialogue:
You shouldknowthat wisdomis not in the art of oratoryor in great
fromthesesensiblethings
volumes,butintheseparation
; andintheturning
to themostsimpleand infinite
form.18
In this passage Cusa gives a clue about the nature of the object of
truewisdom: it is the most simple {simplicssima)and the infiniteform
(infinitaforma). In the De doctaignorantia, Cusa uses these two terms
to designate God.19 It is apparent therefore,that the humilitywhich
15NicholasofCusa,Idiotademente
, Berlin1967,Vol. I, p. 237: "qui peripateticus videris".
16De mente,
p. 257.
17Rice,Nicholasof Cusa, p.
and reticence
347. Rice arguesthatthe humility
on thepartof thelaymanis an exemplification
of mysticism
and the Socratic
commonin earlyItalianhumanism.
scepticism
18Idiotadesaentia,I, pp. 224-226.
19Nicholasof Cusa,De docta
ignorantia,
Hamburg1964,Vol. I, p. 18.
145
23:16:09 PM
is a requirementforthe realizationof truewisdom,is a religiousvirtue.

Hence the layman explains to the orator:
The knowledgeof thisworld,in whichyou thinkyou excelall others,is
beforeGod,andthereupon
merefoolishness
puffsmenup,whiletrueknowledgehumblesthem.20
Paraphrasing St. Paul in this passage,21 Cusa seems to be tryingto
establish an objection to the oratoricaltraditionupon religiousgrounds.
The orator, for his part, understands his discipline as dealing with
wisdom as its main objective. Here he is on common ground with the
layman. Only the layman does not understand the orator's objective
and so the orator accuses him of total ignorance of the oratorical
tradition.
Whatpresumption
is thisofyours,poorand completely
layman,
ignorant
thatyouso despisethestudyoflearning(litter
ae), withoutwhichno onecan
22
advance?
With this response Cusa develops a dialectical tensionbetween his own
position and that of the oratoricaltradition: the pursuit of wisdom as
an experience in humilityversus the same pursuit via the study of
literature.
One might raise the objection, however, that Cusa misunderstands
the rhetoricaltradition completely.The orator aims at effectivelanguage, i.e., eloquent expression,and nothingmore. And so to perceive
the orator as one who seeks wisdom is to misunderstandhim entirely.
The trouble with this analysis of rhetoricis that it is gleaned froma
modern and popular understandingof the discipline. In the Classical
tradition as well as during the Renaissance, rhetoricwas not understood as dealing merelywith effectiveexpression. Cicero,forexample,
regarded the ideal of eloquence as being a combination of rhetoricand
the love of wisdom which is philosophy.23The Renaissance humanist
likewise understood rhetoricas a union of wisdom and style.24Thus it
20De sapientia,I, p. 216.
21 i Cor. 3:19. "Sapientiahuiusmundistultitiaest apud Deum". 1 Cor.8:1.
"Scientiainflat".
22De sapientia
, I, p. 216.
23Seigel,p. 7. Cf.Cicero,De naturadeorum,
Mass. 1967,
Loeb ed., Cambridge,
p. 123.
24Gray,p. 200: "True eloquence,according
couldariseonly
to the humanists,
unionbetweenwisdomand style; itsaimwas to guidemen
out ofa harmonious
towardvirtueand worth-while
goals,notto misleadthemforviciousor trivial
purposes".
146
23:16:09 PM
seems that Cusa has a correctunderstandingof the oratoricaltradition

common among the humanists of his period.
What is it then which he finds objectionable in the oratorical ideal
of eloquence ? A carefulexamination ofthe books comprisingthe Idiota
reveals that he finds fault with this tradition on two counts: i) the
oratoricalmethod of arguingfromauthorityis unacceptable, and 2) the
concept of 'wisdom" in the oratorical culture is incorrect.
With respect to method, Cusa firmlygrasps the reliance of oratory
upon the use of authorities.The orator tells the layman that it might
be possible to know some thingswithout the study of letters,but not
thingswhich are difficultand of great importance [difficilesetgrandes)
since knowledgegrowsthroughaddition (additamenta
).25And this addition to knowledgeis not acquired as it is according to the modernideal
of "science", i.e., through repeated critical exposure to experience.
Rather it is the study of the "authors", the auctores.As E. R. Curtius
explains in his European Literatureand theLatin Middle Ages,
NotonlyfortheentireMiddleAgesbutalso on intothesixteenth
century,
Thereis as yetno modern
] are thetechnicalauthorities.
they[theauctores
science.Medicineis learnedfromGalen,as universalhistory
is learnedfrom
Orosius.. . . Theauctores
arenotonlysourcesoftechnicalinformation,
they
arealso a treasury
ofworldly
wisdom
andgeneralphilosophy.
In theantique
andthousandsoflineswhichputa psychological
poetstherewerehundreds
or ruleoflifein thebriefest
form.26
experience
We shall see how Cusa responds to this ideal of "worldlywisdom" in a
moment,but it is important to firstunderstand what it is about the
methodof auctoreswhich he objects to so strongly.
One mightdescribe the classical Ciceronianorator as one who would
ideally have a wide selection of philosophical opinion to draw upon in
order to support any case he might advocate.27 It is easy to see why
rhetoricas a method for pursuing wisdom would have to be merely
eclectic. Cusa attacks the orator for being blind and easily misled in
yieldingso readily to authority. "You are led by authority (auctoritate)", the layman tells the orator,"and so you are deceived; someone
writes somethingdown, and you believe it!" 28 But besides the cre25De sapierUia,I, p. 216. "Quamvisfortesine litterarum
studioaliqua sciri
et grandesnequaquam: cumscientiecreverint
possunt,tarnenresdifficiles
per
'
' is foundin Cicero'sOratio
additamenta".
The termadditamenta
proSestio,31,
68.
26E. R. Curtius,
andtheLatinMiddleAges, NewYork1953,
EuropeanLiterature
PP-57-58.Italicsare myown.
27Seigel,p. 18.
147
23:16:09 PM
dulity of the method of auctores,there is a much more serious charge

which Cusa raises. Given that one of his fundamental theses is that
being is unified in the One (which he identifieswith God), it must
follow that ultimate knowledge or wisdom must be unified as well.29
The oratorical method, therefore,cannot ever hope to achieve this
unity of wisdom because it is based upon a divergenceof opinion. And
so Cusa has his layman informthe orator: "The opinion of authority
has distracted you, . . . and your understanding,being bound to the
authorityof writersis fed with food which is alien and not natural".30
While the oratorical method can never hope to presentman with a
unified concept of wisdom, because it is based upon a divergence of
opinion, it also must fail to achieve unity because its source is located
in the wrong place. Cusa, followingPlato's condemnationof rhetoric's
dependence upon opinion, has the layman proclaim that true wisdom
is not found by canvassing the opinion of the ancients, or of anyone
else for that matter. Amalgamating two passages fromthe sapiential
literatureof the Old Testament, the layman declares, "Wisdom cries
out in the streets, and her cry is how she dwells in the highest".31
These passages are incorporatedinto the text of the layman's argument
in order to draw attention to two aspects of true wisdom. The first
aspect stresses the accessibility of wisdom to the ordinaryman. True
wisdom can be obtained without recourse to the studiumlitterarum.
This implies that thereis no need forthe humanists' ideal of eloquence.
The second aspect of wisdom which Cusa means to draw attentionto is
its transcendentnature. This point, as will soon be apparent,is directed
against the scholastic philosophers.
When the layman states that "Wisdom cries out in the streets", he
is indicatingthat the starting-pointforthe pursuitof absolute wisdom
is foundin the observationsofthe senses.32He proceedsto demonstrate
to the orator that the activity of the market-placeprovides an exemplar of the humble origin of all human wisdom. The activity of the
market-place,the buying and selling, the weighingand measuring of
things,all indicate that comparison (comparatici)and proportion(pro
29Cf.De doctaignorantia,
Vol. I, Ch. 5.
30De sapientia,1, p. 216.
31Ibid., p. 217: "Sapientiaforisclamatin plateiset est clamoreius quomodo
ipsa habitatin altissimis".Cf.Prov.1:2o, "Sapientiaforispraedicat;in plateis
habitavi".
dat vocemsuam". AndEcclus.24:7, "Ego in altissimis
32Rice,NicholasofCusa,p. 346: "Cusa emphasizes
thatits [wisdom's]
pointof
and itsmessagemoreaccesis thesimplestobjectofsenseexperience
departure
sibleto thesimplethanto thelearned".
148
23:16:09 PM
portio) are basic to human judgment (which Cusa calls discretio).It

would be an errorto attempt to defendthe thesis that Cusa anticipates
modernscience and empiricalphilosophywith his emphasis upon comparison and proportionin sense experience and judgment.33He does
not reject the methods of humanism and scholasticism in order to
develop a proto-scientificmethod. For, as will be seen in a moment,
sense experienceis only a starting-point,a terminusa quo. All of Cusa's
thinkingmoves away fromit to a terminusad quernwhich is absolute
and infinite.The appeal to sensation and judgment is similarto Plato's,
where sensation implies comparison and proportion. These in turn
imply an absolute standard which transcends the finite order.34
Between the transcendentrealm of being and the finiteorder there is
. "Between the finite and the infinite/1Cusa
separation, a chorismos
argues, "there is no proportion". The purpose of starting with an
examination of sensation in the search forwisdom, is to conditionman
to his finitude.This humiliatingexperienceprepares man forthe Absolute. The attention paid to the empirical in the Idiota thus does not
point toward a nascent scientific method, but rather to Platonic
Idealism.
Aftercriticizingthe rhetoricalmethod of humanism,Cusa deals with
the humanist conceptionof wisdom as he understoodit. What he seems
to find objectionable in the humanist ideal of wisdom is its secular
nature and its advocation of a sophistical "art of living" which ties
man to the finiteorderto such an extent that he is distractedfromthe
The layman points out to the orator that the
pursuit of the infinite.35
true object of wisdom so moves man to his proper life (propria sua
vita),36 that once he accedes to it, he begins to move away fromthe
finiteorder to the infinite.Cusa rejects thereforethe tendency of the
humanist to definewisdom as a moral virtue.37
33AlexandreKoyr,FromtheClosedWorldto the
, Baltimore
InfiniteUniverse
of Nicholasof Cusa is not based upon a
*97> P- 8: "The world-conception
criticism
of contemporary
astronomical
or cosmological
and does not
theories,
to a revolution
lead, at least in his own thinking,
in science".Cf. . Jaspers,
Anselmand NicholasofCusa, New York 1974,fora comprehensive
analysisof
Cusa and modernscience.
34De doctaignorantia,
Vol. I, p. 18: "Hoc autemomnemnostrumintellectum
in suo principiocombinare
transcendit,
via rationis".
qui nequitcontradictoria
35Cicero,De finibus
, Loeb ed., Cambridge
Mass. 1971,p. 46: "Wisdommustbe
considered
as theartofliving(ars vivendi)
. If it effected
no result,it wouldnot
be desired
; but as it is, it is desiredbecauseit is theartificer
thatprocuresand
producespleasure".
37Rice,Nicholasof
Cusa, p. 364.
149
23:16:09 PM
We shall examine Cusa's idea of wisdom afterwe have analyzed his

treatment of scholasticism,but it must be pointed out now that the
rejection of humanisteloquence is based upon a personal,perhaps even
a mystical conception of wisdom. This "experience" of wisdom precludes all academic appeal to authorityand is based upon something
purely subjective, though not necessarilyirrational.Thus eloquence is
undercut once and forall in Cusa's system. The layman concludes the
attack on the orator by observingthat,
This wisdom,whichno man tastesby merehearsay(auditum),
is tasted
onlybythemanwhotastesitwithinhimself
(ininterno
gustu).Andhebears
notofthingswhichhehasheard,butofwhathehasexperimentally
witness,
tastedin himself.38
Having dealt with humanisticeloquence, Cusa turnshis attentionto
scholasticism in the thirddialogue of the Idiota , the De mente.His interest in scholasticismin this dialogue focuses on demonstratingtwo
things. First, he wishes to show that the object of true wisdom is
transcendent.And second, he wants to demonstratethat the rationalistic method of scholastic argumentationis not strongenough to deal
with this object. The philosopherin the dialogue helps to establish this
last point when he admits to the humanisthis admirationforthe faith
of common laymen. "Certainly", he admits, "it must be a giftof God
{dei donum), that laymen see more clearly by faith,than philosophers
by reason".39 Here we see Cusa exalting lay wisdom, which will turn
out to be equivalent to faith, and at the same time condemningthe
attempt to use reason as a philosophical tool forpenetratingthis wisdom. Thus Cusa is attacking scholasticism in the very heart of its
program.
The layman begins his criticismof scholasticism by rejecting the
scholasticuse ofauthorityin argumentation.Cusa describesthe methodological use of authorityas being almost as eclectic as that of humanistic eloquence. The philosopherseeks wisdom, he writes,"by comparing the writingsof the learned with each other, and seeking out their
meaning".40 Since he feels he has already adequately dealt with this
procedure, Cusa goes on to distinguish a second form of scholastic
authority which he regards as defective, i.e., the adherence to the
authority of a school. This formof authority,which to some extent
gives scholasticism its name, Cusa rejects because it submits man's
39De mente
, p. 236.
40Ibid.,p. 237.
150
23:16:09 PM
search forabsolute wisdom to mere human authority.And so the layman argues with the philosopher:
WhetherI am a "Pythagorean"
ornot,I do notknow.But thisI do know:
thatno man'sauthority
movesme.11
[nulliusauctoritas)
But what Cusa is most concerned with in scholasticism is what he
regards as the inappropriate extension of reason (ratio) in the vain
attemptto reach the object of true wisdom. There are two presuppositions involved here: i) that the object of true wisdom is transcendent
and 2) that mere reason is too weak a functionof the apparatus of
human knowledgeto reach this object. Cusa has argued in some of his
earlierworksthat human reason is confinedto finiteand denumerable
reality.42At best reason can achieve secular wisdom, i.e., wisdom concerningthingswhich are imperfect.Cusa proposes that the criterionof
this secular wisdom is that it judges of thingswhich admit of a "more
or a less". Finite thingscan be increased in the sense that they can be
morethan they are ; they can be decreased by being less than they are.
This increase and decrease can occur in many cases without loss of
identity. In fact, Cusa argues, finitethings are always increasing or
decreasing in some manner. The Object of true wisdom, however, is
absolute and unchangeable. Hence the followingsection of dialogue:
Orator:In thosethingswhichadmitof a moreor a less (magiset minus)
thereis no conception
ofGod whichcan be formed.
is correct.
SinceGodis infinite,
thosethingswhich
Layman: Yourinference
admitofa moreor a less are notlikeHim.43
This part of the dialogue generates a discussion of what accounts for
the inabilityto forma concept of God fromthe realm of the "more or
the less". The layman argues in a somewhat Platonic fashionthat since
God is the "precisionof all things(cuiuscunquereiprecisio)" 44and since
the concept of the "more or the less" measures the world of the imprecise, it would be a great error to attempt to measure God with
concepts which are opposed to His nature. And this is just what
scholastic philosophyattempts to do as far as Cusa understandsit.
The scholastic concept of wisdom which Cusa is attacking is based
41Ibid.,p. 249.
42De doctaignorantia,
Vol. I, p. 98: "Nominaquidempermotumrationis,
quae
multoinferior
intellectu
est,ad rerumdiscretionem
imponuntur.
Quoniamautem
ratiocontradictoria
transilire
nequit,hincnonestnomen,cui aliudnonopponatursecundummotumrationis".
43De sapientia
, II, p. 230.
44De mente
in Philebus,56a.
, p. 242. Cf.Plato conceptofdxpieta
151
23:16:09 PM
upon the classical Ciceronianand Augustinindefinitionwhichdeclares

wisdom to be of thingsboth human and divine.45The layman,however,
rejects this traditional definitionof wisdom by restrictingits object
solely to the divine nature. Knowledge of such an object, Cusa seems
to argue, is not naturally acquired by any process of abstraction or
analogy.
The concept of analogy presupposes proportion,and Cusa has already argued in the De docta ignorantiathat there is no proportion
betweenthe finiteand the infinite.In the Idiota he tries to show that
proportionserves to draw the conscious subject's attentionto its own
limitations.It can establish no directrelationto the infinite;all it can
do is provide the necessary condition for a kind of transcendental
"leap" of the subject into its proper object, a leap which is achieved
by negation. This is not to say that Cusa's layman is advocating a kind
of irrational mysticismbased upon mere will. Cusa proceeds with the
object of true wisdom in much the same fashion that Kant does in
establishingnoumenal being.46
The scholastics err, according to Cusa, in tryingto apply only one
aspect of finitereality to the infiniteobject. That is, they apply the
perfectibilityof the "more" of the finiteorder to God, or deny the
"less" of Him. But they fail or refuse to predicate the perfectionof
both the "more and the less" of finiterealityof Him because they feel
such predication must violate the law of contradiction.Cusa asserts,
however, that laying aside the law of contradictionis the key to the
correct method of pursuing true wisdom. For predicating both the
perfectionof the "more" and the "less" absolutely of the infiniteobject generates a contradiction. This contradiction,Cusa claims, accomplishes two ends: i) It cancels out the application of any aspect
45Cicero,De officiis,
Loeb ed.,Cambridge,
Mass.1951,p. 172:"Sapientiaautem
et humanarum
est,rerumdivinarum
est,ut a veteribusphilosophisdefinitum
Contraacascientia".St. Augustine,
causarumque,quibuseae res continentur,
Vol. LXIII,
Ecclesiasticorum
dmicosincludedin CorpusScriptorum
Latinorum,
New York 1962,p. 16. "Non enimnuncprimoauditissapientiamesse rerum
scientiam".
humanarum
divinarumque
46The epistemological
similarities
betweenKant and Cusa are quite striking.
Justas Kant arguesthattherecan be no realunityin theworldofappearances
unlessan orderofrealityis postulatedwhichprovidestheunityin themanifold
theappearancescannotsupply,Cusa argues
ofappearanceswhichbythemselves
thattherealmofthe "moreor theless" whenjudgedby themindindicatesa
the"moreortheless" northemindcouldprovide.Thus
whichneither
precision
whichis identified
withneithertheworld(= therealmofthemore
something
or theless) northefinitemindmustbe postulated.Withoutthispostulatethe
finitemindcouldnotthinktheconceptsof"more"or "less".
152
23:16:09 PM
of finitereality to the infiniteobject by means of analogy or proportionality.2) It demonstratesthat the object of truewisdomcomplicates
{complicare)dalectcally opposed aspects of finitereality.The infinite
object, Cusa observes in De doctaignorantia, is the absolute maximum
and the absolute minimum. That is to say, it is the "more and the
less" made absolute and predicated simultaneouslyof itself.The pursuit of such an object cannot be conducted with the standard logical
techniques of the schoolmen.
We have seen how Cusa has reacted stronglyto both humanisticand
scholastic culturewith respect to the problem of wisdom. It remainsto
be seen, however, how Cusa proposes to replace these ideals. It was
stated earlierthat the layman in the dialogues of the Idiota serves as a
means for Cusa to refute the ideologies and methodologies of both
humanism and scholasticism. But the layman also representsan embodimentof what Cusa calls 'learned ignorance". As we already noted,
he points to the activity of the market-placeas a key to wisdom. This
,47From this observation the layactivity takes place per discretionem
man argues that since discretionimplies number and number implies
unity,the activityfundamentalto human perceptioninvolves unity.48
But the layman goes on to point out that while unity or the "one"
{unum) is the necessary condition for the act of discretion and for
number,unity itselfis not subject to discretionnor is it denumerable
in any way. If it were, it would cease to be one. How can it be known
then, since man perceives by discretion? Reason {ratio), Cusa argues,
operates by discretion; there is, however,another aspect of the mind
which grasps objects by a kind of intuition.This intuitiveability Cusa
calls the intellect {intellectus)
.49It is the intellect which comprehends
that there must be somethingbeyond the realm of the "more or the
less" which is not subject to discretion.Man does not knowthis "one"
which makes all of his knowledgepossible, but he does have what Cusa
calls a "foretaste" {pregustatio)of it. It is throughthis foretastethat
desire for the infiniteis awakened in the soul. The layman tries to
explain this to the reader in somewhat metaphorical language:
The eternaland infinitewisdomwhichshinesin all things,invitesus,
a certainforetaste
ofitseffects,
to be carriedto it outofa wonderthrough
fuldesire.50
47De sapientia,X,p. 217.
48Loc. cit: "Per que autemdiscretio:nonne unumnumerat".Cf.De docta
per
Vol. I, p. 22.
ignorantia,
49De doctaignorantia
, Vol. I, p. 98.
50De sapientia
, I, p. 220.
153
23:16:09 PM
The fundamentalconditionof man with respectto this infinitewisdom

is one of ignorance.This does not mean that the human conditionimplies merehelplessness.'There is nothingdesired", Cusa argues,"which
is utterly unknown".51It is possible for the intellect to become informed or learned with respect to the humbling experience of its
ignorance. It can know the infiniteobject of human desire,the goal of
all knowledge,by means of negation. It can know what eternalwisdom
is not. And this, Cusa claims, is a kind of wisdom. This sort of wisdom
he says "knows how that which is inattainable may be attained
inattainably".52
Cusa's language appears to be similar to that of the via negativa,
i.e., the language of mystical theology. But what 'is interestingabout
Cusa' s use of the via negativa is his viewing it as a way of avoiding
auctoritasin the pursuitof wisdom. It is a method of approaching God
which is open to the commonlayman. The only prerequisitesare grace
and freedomfromsin. "The spirit of wisdom", Cusa tells us, "dwells
not in a body subject to sin, nor in an evil-willingsoul".53 F. P.
Pickering in his article, "Notes on Late Medieval German Tales in
Praise of Docta Ignorantia", observes that there is a distinct type of
"popular literature"at the close of the Middle Ages in Germanywhich
connects lay piety and mystical insight.54He restrictshis study to the
relation of this new piety to Churchauthority.It is apparent, however,
that this upsurgenceof lay piety extended to other areas of tradition
as well. An interestingtract from the fourteenthcentury, entitled
MeesterEggaart en de onbekendeleek {MeisterEckhartand the Unknown
Layman) is an example of the thesis that an ordinarybut pius layman
could teach theology to the most learned masters.55Another tract,
SchwesterKatrei, perhaps writtenunder the influenceof Meister Eckhart, develops an idiota-motifnot unlike that of Cusa's and stressesa
personal mysticismwhich transcends traditional liturgicalpiety. It is
certainly possible that Cusa knew of these writingsand was deeply
influencedby them. Anotherexample of lay literatureis Johannesvon
51Ibid.,p. 221.
62Ibid.,p. 218.
53Ibid.,p. 223.
54Pickering,
p. 122: "This typeoftale mustbe familiarto anyoneacquainted
But itis
ofthelaterMiddleAgesin Germany.
withthepopularmysticliterature
: ... a simplelaymanmay,in respectto
naturallyto be foundin othercontexts
truepiety,excelonein holyorders".
65C. G. N. de Vooys,ed. Meester
leek,in: Nederlandsch
Eggaarten de onbekende
voorKerkgeschiedenis,
n.s. VII (1910)pp. 166-226.
Archief
154
23:16:09 PM
Saaz's Der Ackermannaus Bhmenwhich develops the theme of a pius

ploughman challengingthe power of death to a debate. The overtones
of the dialogue are cosmic in their dimensions. Cusa, operating in a
genre similarto these, seems to challenge neitherthe ecclesiastical nor
the cosmic hierarchyof the Late Middle Ages, but the auctoritasof the
traditional,philosophical culture of scholasticism as well as the new
authoritativecultureof humanisticeloquence and wisdom. Though he
found much that he valued in both, he came to realize that both presented a dead end for the new line of thought which he sought to
capture. Between these two powerfulcultures which were so much a
part of intellectuallifein his day, he has his layman steer his craftof
mysticalpiety and learned ignoranceas Homer had his Odysseus steer
between Scylla and Charybdis.
Minneapolis,Minn., U.S.A.
AugsburgCollege
155
23:16:09 PM
Reviews
theorieen
van DanielHeinsius.Een onderzoek
naarde
J. H. Meter,De literaire
klassiekeen humanistische
bronnenvan De TragoediaeConstitutione
en
anderegeschriften
in English),Amsterdam
(witha summary
(A. M. Hakkert)1975.
Das Interessean Daniel Heinsiushat in denletztenJahrenstarkzugenommen.Biographische,
undtheologische
Arbeitenzu Personund
bibliographische
Werkverliessenin rascherFolge die Druckereien.
Das Augenmerk
galt vor
allemderniederlndischen
Dichtung,die z.T. in kritischen
AusgabenundreproNeudrucken
wiedererschienen
ist.1EineAusgabederNeder
duytschen
grafischen
Poematawird,wennich rechtunterrichtet
Die
bin, demnchsterscheinen.2
was
Schriften
standendagegenbisherim Hintergrund,
dichtungstheoretischen
nichtmit einerschlechtenVerfgbarkeit
der Texte begrndetwerdenkann.
und
Sie sindaufdenbedeutenden
Bibliotheken
durchbersetzung
vorhanden,
OffenNeudruckwurdensie z.T. einemweitenLeserkreis
gemacht.3
zugnglich
hatmandieMhegescheut,
sichtlich
dieOriginaltexte
zu lesenundsystematisch
zu analysieren.
dieserLage wirdmanes uneingeschrnkt
Angesichts
begrssen,
in einerVerlagsdass nun J.H. Meterslangeerwartete
Utrechter
Dissertation
publikationvorliegt.Um es gleichvorwegzu sagen: Es handeltsichum eine
in der zum erstenMal die Anfngeder Heinsiusschen
Poetik
Pionierarbeit,
werden.
sowie seine Horaz- und Aristoteles-Rezeption
umfassenddargestellt
istdamiteinegrosseLckegeschlossen
In derNeerlandistik
unddarberhinaus
frdietiefergehende
Literaturdes
eineGrundlage
dereuropischen
Erforschung
MeterstelltsichzurAufgabe,denspeziellenCha17. Jahrhunderts
geschaffen.
von Heinsius1
Ansichten
anhandvon
rakterunddie Entwicklung
theoretischen
Vorworten
undphilologischen
bis 1611,
Gedichten,
Briefen,
Verffentlichungen
Trotz
von De Tragoediae
dem Erscheinungsjahr
zu untersuchen.
Constitutione,
Band eine
derzeitlichen
enthltderfast650 Seitenumfassende
Beschrnkung
und neuerErkenntnisse,
Flle erstmaliger
die den Rezensenten
Darstellungen
vor das Problemder Auswahlstellen:Er muss Schwerpunkte
unweigerlich
setzenund luftGefahr,wichtigeTeile unerwhnt
zu lassen.
Meterstrebteinepsychologische
an,dievoneinerWechselBetrachtungsweise
dass
zwischenLebenund Werkausgeht.Daherist es nurfolgerichtig,
wirkung
das einleitende
der Persnlichkeit
Kapitelsich mit der geistigenEntwicklung
Wennman weiss,welchenSchadenterHorstsHeinsius-Monograbeschftigt.
dass Sellin ein
hat, und wenn man dem Urteilzustimmt,
phie angerichtet
flachesundblassesBild des Charakters
dannwirdmandie einleitenzeichnet,4
den Seitenwillkommen
heissen.Das giltauch,wennmanMetersPrmisse,der
"
Grundzugvon HeinsiusCharaktersei die eerzucht" gewesen,nichtin allen
missbilliHinsichtenfolgenkann.Denn was hierals individuelle
Eigenschaft
der Zeit. In
wird,gehrtzu den typischenCharakterzgen
gend angefhrt
1^6
Vivarium
XVI, 2 (1978)
23:16:16 PM
unterschiedlichen
Graden der Verwirklichung
bewegt sich die "eerzucht"
zwischenden Polendes Willenszur Reprsentation
und derGestederBescheidenheit.Meterentwirft
ein Persnlichkeitsbild,
das sich wohltuendvon dem
was bisherberHeinsiusgeschrieben
wurde.Da stehtnichtmehr
unterscheidet,
derreizbare,
da istnichtsmehr
MannimVordergrund,
launische,
trunkschtige
zu lesenvom Klischeeinesweitabgewandten
und pedantischen
Altphilologen
den Achimvon Arnimin einerNovellein der Figurdes Professors
Gelehrten,
Hemkengriper
parodierthat.6 Aufgrundeines grndlichen
Quellenstudiums
kann Meter Heinsius als einen Altertumswissenschaftler
darstellen,dessen
Interessenichtauf den Vergleichvon Lesarteneingeengt
und der nichtausschliesslich
von antiquarischer
Neugierdebesessenwar.Heinsiuslenkteseinen
Blickauch aufdie Gegenwart.
Sein humanistisches
enthielt
Wissenschaftsideal
nebensthetischen
und wissenschaftlichen
auch pdagogischeZiele. Die Bemitden AutorenderAntikewar ihmkeinSelbstzweck,
sie hatte
schftigung
auch der Formungdes humanistisch
gebildetenMenschenzu dienen.Diese
hat Heinsiusvon J. J.Scaligerempfangen,
der,jeglicherVertrockEinstellung
und grammatischer
Studienabgeneigt,aus einemleidennung lexikalischer
schaftlichbegeisterten,
jedoch noch wahllossuchendenSchlereinen festHumanistenformte.Im Verhltnisbeider,so der Vf., waltete
gegrndeten
ein pdagogischer
Krfteim Lehrlingweckte.
Eros, der die schlummernden
Das meiste,das Heinsiusin den frhenJahrenschrieb,verdankter den AnseinesVorbildes,
und alles,was er verffentlichte,
regungen
gingdurchdessen
Zensur.Die DarstellungdiesesLehrer-Schler-Verhltnisses
ist das Kernstck
des erstenKapitels.Was MeterdurchbehutsamesInterpretieren
der Quellen
istdas weitausBeste,das bisherzu diesemThemageschrieben
wurde.
erarbeitet,
Heinsius*
frheliterarische
wurdenvon derGrundauffassung
Anschauungen
dass der Dichterals Weisheitslehrer
Vermittler
zwischenGottund
bestimmt,
Menschist. Hieraus ergibtsich die prinzipielle
der Dichtungin
Zweiteilung
einenphilosophisch-sthetischen
und einenphilosophisch-moralischen
Bereich.
Sicherlich
bedarfdieTheseMeters,
dassdieseZweiteilung
einSpiegeldesinneren
Streiteszwischenden sthetischen
Idealen der Renaissanceund der calvinistischenEthiksei, einerweitergehenden
Diskussion.Dass Heinsiussich hier
durchausim RahmenderTraditionbewegt,stelltauchMeterausdrcklich
fest.
Er siehtin den frhenAnschauungen
eine Synthesezahlreicher
Bildungseinderpythagoreischen,
und stoischen
Lehren.Er
flsse,namentlich
platonischen
bleibtabernichtbei derblosspositivistischen
derQuellenstehen,
Entdeckung
sondernzeigtdie Entwicklung
derAneignung
auf.Es wirdberzeugend
argudass Heinsiusum 1609am Scheideweg
undDichtung
von Philosophie
mentiert,
stand,und es zeigtsich,dass er die bisherverfolgte
Richtungeinesplatonisthetizismus
und
sierenden,metaphysischen
zugunsteneiner rhetorischen
strukturellen
Betrachtungsweise
aufgab.Dieser Weg wurdeihm durchdas
Studiumder Schriftenvon Aristoteles,
Horaz, G. I. Vossiusund wiederum
durchScaligergewiesen.Das machtdas Buch so wertvoll:Die Wandlungen
werdennichtnurkonstatiert,
sondernSchrittfrSchrittverfolgt
und aus der
Methode
jeweiligenSituationherauserklrt.Die entwicklungsgeschichtliche
Meterszeigtkeinengradlinigen
Weg aufundfhrtnichtzu eindeutigen
Ergebeinzuebnen.
nissen,dennderAutorerliegtnichtderVersuchung,
Widersprche
Er lsstsie vielmehr
in ihrerKomplexitt
bestehen.Das zeigtsichzumBeispiel
deutlichin der Behandlungvon Heinsius'Stellungzur Personund zum Werk
des NonnusPanapolitanus.In der JugendfhltesichHeinsiusdem Geistder
undes bedurfte
einesstarkenEinsatzesScaligers,
verwandt,
Dionysiacainnerlich
den Schlervon dieserVorliebeabzubringen.
In Heinsius'Dissertatio
de Nonni
Dionysiacavon 161 findensich denn auch die EinwndeScaligersgegen
157
23:16:16 PM
es handlesich im Grundeum ein

Nonnusausfhrlich
belegtund bekrftigt:
das aus einerReiheversatzstckartig
WerkohneklarenHandlungsaufbau,
gebrauchterEpisodenbestehe.Einige JahrespterjedochwendetsichHeinsius
Nonnuswiederzu - diesmalin der muttersprachlichen
Dichtung.Fr seinen
LofsanckvanBacchushat er das Werkvon Nonnuszu Rate gezogenund ihm
auchimmer
So grossdieWirkung
desLobgesanges
einigeEpisodenentnommen.
Vorwrfen
gegenden
gewesenseinmag,es ist bekannt,dass er zu zahlreichen
In
des Gtzendienstes
AutorAnlassgab, die in der Beschuldigung
gipfelten.
fhrtHeinsiusNonnusals Kronzeugen
dafran, dass die
seinerRechtfertigung
des
ThemaskeinenMastabfrdie Glubigkeit
Behandlungeinesheidnischen
Autorsdarstelle.
war
Constitutione
Heinsiusdichtungstheoretisches
HauptwerkDe Tragoediae
Sein Wertliegtnichtso
als LehrbuchfrangehendeDramatikerkonzipiert.
wie
sehr in der neuen,originellen
Gedankengutes,
Deutung aristotelischen
InterBecker-Cantarino
schreibt,als in dem Bemhenum eine werkgerechte
Es istwohldas Werkvon Heinsius,das in ganzEuropadiestrkste
pretation.
vor,von denennur
Wirkungausgebthat. Dazu liegeneinigeUntersuchungen
sichmitdemEinfluss
die wichtigsten
genanntwerdensollen: Kernbeschftigt
auf die franzsische
Literatur,Zerbstmit dem auf Lessing,und in jngerer
Metersah sichalso
Zeit hat Sellindie Spurenim "StuartEngland"verfolgt.7
die Artund
: Heinsius'Text zu analysieren,
voreinedreifache
Aufgabegestellt
zu berprfen.
unddiebisherige
WeisederRezeptionzu untersuchen
Forschung
Um mit dem letztenPunktzu beginnen:Wichtigsind die Richtigstellungen
in dernachdes Verfassers
zurSekundrliteratur,
MeinungHeinsius'Katharsisweilallzu einseitig
behandeltwerde.Meterhingegen
unzureichend,
Auffassung
denermit"Nuttigineinemgrsseren
dieKatharsis
betrachtet
Zusammenhang,
unddener in die folgenden
heiden doeivan de tragedie"berschreibt
Aspekte
Heinund den ethischen.
densthetischen
: denpsychotherapeutischen,
gliedert
derKatharsis
siusgehtvomursprnglichen,
Begriff
medizinisch-therapeutischen
des Menschen.Das Ziel
ihn auf die psychische
aus und bertrgt
Verfassung
HierinsiehtMeterwohl zu
der Reinigungist das emotionaleGleichgewicht.
rechtdie Spurenderpythagoreischen
Lehre,wennHeinsiusihrauch nichtin
dass jede Emotioneinerunio
letzterKonsequenzfolgt.Denn die Auffassung,
Aristoteles
Er interpretiert
mysticaim Wegestehe,hat er nichtbernommen.
werdenmssen,whrend
in dem Sinne,dass nur die schdlicheneliminiert
frdie AusbungderWissenschaften
anderedurchausals Triebfeder
notwendig
des Zuschauers
sind. Der sthetische
Aspektberuhtauf einerDistanzierung
und
Geschehen.Der Vergleichvon fiktiver
vom auf der Bhnedargestellten
Situationfhrtzu einemsthetischen
tatschlicher
Genuss,der nichtals ein
werdendarf.Er wirdvielmehrals die VoraussetSelbstzweckmissverstanden
zung zum ethischenEffektangesehen,dessenZiel die Strkungder Widerdes Lebensist.Er beruhtaufdemallgemeinstandskraft
gegendie Wechselflle
Emotionenden Menschennureinebedass heftige
gltigenVerhaltensmuster,
Aufdie
eintritt.
grenzteZeit in Anspruchnehmen,da bald eine Gewhnung
heisstdas: das hufigeLesenoderAnschauenvon
Theatersituation
bertragen,
Tragdienfeitden Menschengegendie Unbilldes Daseins .Die sich aufdrngendeFrage,in welchemVerhltnisKatharsisund ethischeWirkungstehen,
UnterMeterwiefolgt:Zwischenbeidenbestehenureingradueller
beantwortet
diese trage zu seiner
schied. Jene bewirkeein emotionalesGleichgewicht,
bleibtHeinsiusderfrheren
Forschung
Festigungbei.MitdieserInterpretation
Wissenschaftlichkeit
dessennchterne
Robortello
z.B.,
verpflichtet,
weitgehend
ummnzt.In derBeschreibung
er aberzuweilenin einekraftvolle
Bildsprache
Prozessesjedochgehter unabhngige
des psychotherapeutischen
Wege.
158
23:16:16 PM
abergerechtmitZerbstund Sellin.Er wirft

Meterverfhrt
ihnenvor,
streng,
vomNutzenderTragdieallzu einseitig
HeinsiusAuffassung
als rhetorisch
und
zu haben. Seine hiervorgetragene
moralistisch
interpretiert
Deutungist ein
Zwar glaubteauch Heinsiusan die moralische
berzeugender
Gegenbeweis.
AufgabederTragdie,aberdiesezeigesichnichtin ersterLiniein derKatharsis,
sondernim Vorfhren
vorbildlicher
undnachahmenswerter
Charaktere
LebensDie dreigenannten
weisheiten.
ber
Aufgaben,die sich bei der Besprechung
denNutzenderTragdiestellten,
sindvoll und ganz erfllt
worden.Fr dieses
dass derVerfasser
ber
Kapitelwiefrdas gesamteBuchgiltdie Feststellung,
einebewundernswerte
Kenntnisder antikenund humanistischen
Quellenverin der Rezeptionsforschung
Sicherheit
besitzt,die
fgtund eine methodische
beidein die AnalysederHeinsiusschen
werden.
Texteeingebracht
Meterhat keinleserfreundliches
Buch geschrieben.
Die immenseFlle des
Wissensunddiedamitverbundene
lsstnurausdauernde
BreitederDarstellung
Leserstandhalten,
die aber dann frihreMhereich.belohntwerden.Einige
Massnahmenerleichtern
die Lesbarkeit.ZahlreicheQuerverweise
ermglichen
die Orientierung;
ausfhrliche
Registererlaubenes, das Buch auch als Nachzu benutzen;die Zitatein den altenSprachensindbersetztoder
schlagewerk
Die Erfahrung
Werkein
lehrt,dass literaturwissenschaftliche
paraphrasiert.
niederlndischer
SpracheimAuslandnureinenkleinenKreisvonFachgelehrten
erreicht.
Auch die frDissertationen
Zusammenungewhnlich
umfangreiche
fassungin Englischwirddarannichtviel ndern.Der Mehrzahlder deutschen
Germanisten
bereitetdie Lektrehollndischer
Bcherein unberwindbares
Hindernis.Und MeterscheintderAnsichtzu sein,dass sichdie Lektreseiner
Arbeitfrsie auch nichtlohne,denndie deutschenDichterund Gelehrten
des
httendietheoretischen
Schriften
vonHeinsiusnichtbeachtet.
17.Jahrhunderts
mchteich mit allem Nachdruckwidersprechen.
DieserVermutung
Wennes
bishernur verstreute
zur Rezeptionder Dichtungstheorie
von
Bemerkungen
Heinsiusin Deutschlandgibt,so sagt das nichtsberden tatschlichen
Austauschaus. Eingehende
dieauchdas 18. Jahrhundert
umfassen
Untersuchungen,
wrdenzu einemanderenErgebnisfhren.
Meterselbsthat mitseiner
mssten,
Dissertation
dafrdie Voraussetzung
Es ist daher
ausgezeichneten
geschaffen.
zu wnschen,
dass sie auf welcheWeiseauch immereinemdes Hollndischen
nichtkundigen
Leserkreis
gemachtwird.
zugnglich
UlrichBornemann
1 Bacchusen Christus.
TweeLofzangenvanDaniel Heinsius.Opnieuwuitgegeven
doorL. P. Rank, J. D. P. Warnersen F. L. Zwaan, Zwolle 1965 (= Zwolse
Nr.53); Emblemata
Drukkenen Herdrukken
Amatoria1608.Introductory
Note
EmblemBooks Nr. 10); Moerby . N. Smith,Menston1973 ( = Continental
* en
mansLeidenerDissertation
Daniel Heinsius.Zijn 'Spiegel
in de
spiegeling
liter
atuurgeschiedschrij
ving,Leiden 1974, enthltS. 10-33 * kommentierte
Editiondes SpiegelvandeDoorluchtige,
endevereerlicke,
Cloucke,Deuchtsame
vrouwen
(1606).
standege
2 Sie wirdvon BaerbelBecker-Cantarino,
of Texas at Austin,vorUniversity
bereitet.
3 Daniel Heinsius,On Plot in Tragedy.Translatedby P. R. Sellin and J. J.
MacManmon.WithIntroduction
and Notesby P. R. Sellin,Northridge
(CaliDe PoeticaLiberDaniel Heinsiusrecensuit
. . . Accedit
fornia),1971; Aristoteles
Daniel HeinsiusDe Tragoedia(sic!) constitutione
. . . reprografischer
Neudruck
derAusgabeLeiden1611, Hildesheim1976.
159
23:16:16 PM
4 D. J. H. terHorst,Daniel Heinsius, Diss. Leiden

1934; P- R- Sellin,Daniel
Heinsiusand StuartEngland,Leiden1968.
6 Und zwarin derNovelleHollndische
Liebhabereien.
e Daphnis4 (1975),S. nof.
7 E. G. Kern, The Influenceof Heinsiusand Vossius
upon FrenchDramatic
Theory,Baltimore1949; M. Zerbst,Ein Vorlufer
Lessingsin derAristotelesJena1878; zu Sellins. das in Anm.4 genannteBuch.
interpretation
StudiumUpsalense.Specimensof the Oldest LectureNotes Taken in the
of Uppsala ed. by AndersPiltz. Acta Universitatis
MediaevalUniversity
C: Organisationoch
Upsaliensis.SkrifterrrandeUppsala Universitet.
historia,Uppsala 1977(382pp.; Sw. kr. 91.- ).
of Uppsala celebratesits 5th centenaryin 1977. For this
The University
occasionAndersPiltzpresents
thelecturenotes,takenmainlybyOlausJohannis
in theyears1477-86,conserved
Gutho,fromcoursesat the Uppsala University
in theMss. ofthe University
Libraryof Uppsalaunderthesigns 195, 242,
599, , 6oi, 602 and 629. This materialoffers
a uniquepossibility
oftheintellectual
ofgivinga cross-section
activitiesat a late-medieval
university.The studyby Piltz openswitha preciseand informative
surveyof the
personsand the formsof teachinginvolved.The comparisonsbetweenthe
of Uppsala and thoseof Greifswald
curriculum
and Leipzig are illuminating
(For LeipzigPiltz can drawon newmaterialin theformof lecturenotesand
in theMs. oftheMunicipal
cedulaeby FabianWchterform1486-89,conserved
in scope,
comments
arerestricted
426fol.).The doctrinal
LibraryofNorrkping,
but balancedand well-informed.
and accuratedesThe secondpart ofthe volume(pp. 67-135)is a thorough
Uppsala-Mss.and ofa Ms. oftheCathecriptionofthesevenabove-mentioned
dralLibraryofStrngns,
contentsas 195
Q 16 (IX), withsimilartheological
doesnotonlyinform
abouttheworksincluded,but
and 242.Thisdescription
forthe exercitiaon Aristotleand the Parva Logicaliait also enumeratesall
questionswitha shortindicationof thepositiontaken.This sectionwillbe of
in 15thcentury
thegreatesthelpto scholarsinterested
philosophy.
61 selectedtextsfromtheMss.described.
Thethirdsection(pp. 137-313)offers
and the
fromtheintroductions
havebeenchosenpredominantly
The specimens
openingquestionsof the individualworks,so as to revealthe mostgeneral
of the restof the
whichhave a bearingon the interpretation
considerations
willbe
and ofphilosophy
texts.Hencestudentsofboththehistoryoftheology
ofthedoctrines
and thecommonpointsofview
able to forma firstimpression
whichis probablyfairlytypicalof late 15thcentury
oftheUppsalauniversity,
fromlate 15th
Thisis so muchmoreimportant
sincetext-editions
universities.
centuryphilosophy(exceptforthe Polishsources)are stillrare.- The edition
well-done.
Theremightbe someminorerrors
but that is unavoidis extremely
material.Each textis providedwitha doubleapparaable withsuch difficult
and one ofcriticalnotes.
tus: one ofreferences
foundin theMss.but
The volumeis roundedoffwitha listofabbreviations
in Capelli(pp. 315-41)whichwillbe veryhelpfulto studentsof
notmentioned
scholasticMss.Alsoincludedarea listofLatinwords(pp. 342-50),
15thcentury
ofthereferences
and quotations(pp. 351-64),a Bibliography
an Indexlocorum
an
rerum
et
and
Index
(pp. 376-81).
personarum
(pp.365-75)
to late 15th century
All this makesthe book a veryusefulintroduction
thoughtand a worthyhomageto theAlmamaterUpsalensis.
university
JanPinborg
j5o
Vivarium
XVI, 2 (1978)
23:16:25 PM

Vivarium, Vol. 16, Nos. 1-2, 1978 PDF

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

Vivarium, Vol. 16, Nos. 1-2, 1978 PDF

Hochgeladen von

Copyright:

Verfügbare Formate

Vivarium

Reprintedwiththe permissionofthe originalpublisher

OF VOLUME XVI (1978)

Boethius and Nicomachus: An Essay

Consequencesetinconsequencesde la suppositionvide dans la logique d'Ockham.

Mental Verbsin TerministLogic {John

Olga WEIj ers

Contribution l'histoiredes termes*na'

On Ancient and Mediaeval Semantics

Master Guido and his View on Govern

Wisdom and Eloquence in Nicholas of

was principally a translatorwhen putting togetherthe De institutione

Pattern of citation in the mathematical works

VI. Ncomachus and Ptolemy

Patternof citationin themathematicalworks

An initial step in determiningthe relationship between Boethius

14For editionsee IngemarDring,Die Harmonielehre

tions (JanS. 260, i2f.), as well as 3) a division of the monochord

Book II is such a logical and necessaryoutgrowthof Book I, and since

Citations of authors and works in Book III are as follows:

The general tone of Book III is completely Pythagorean, and no

No student of De institutionemusica can deny that Book IV is the

13 is based on Nicomachus' lost musical treatise;30and chapters 14-18

which mingle,while dissonant sounds axe thosewhichdo not. Boethius'

35The languageusedby Nicomachuswithregardto consonanceand highand

other than a diagram of the notation. The presence of Latin names

tion? 3) What are possible sources of the notation? 4) What is the

musica, and, given Boethius' manner of citingother writers,Boethius

was already changed in Ms Greek source. The immediate response to

clearly: Ptyhagoras firstdeterminedthe division of the diatonic genus,

vation of Boethius' division fromNicomachus is impossible.52Pizzani

two divisions mentionedby Nicomachus in the Enchiridion,those of

semitone. The division of the Timaeus was followed by Plato up to

monic parhypate (or trite).But mathematicalnecessityshould not lead

institutionemusica.**Even Pizzan must admit that this chapter of

chapters withoutexploringeach theoreticaldetail, but here I will only

corpus of Boethius' writingsrefutesthis opinion; formost of his works

But Ptolemy considers alpha to be the highest note while Boethius

Boethius includes the diapason system and synemmenon

of Boethius' discussion of the modes. An argument that Ptolemy is

Primamigiturdiximusessespeciemdiapasonearn,quae est AH.

threads representedtherein.In the firstplace Boethius bases his early

vicissim.EK. et .KF. plectroadhibitopellantur,

individual parts, one might,like Pizzani, view this book as a "wild

100Thus the conservatism of

are foundin the firstand last chaptersof the fourthbook ; finally,the

Nicomachus and Ptolemy

A.D. 124) translated Nicomachus' arithmetical treatise into Latin,

quaintance with Ptolemy's works is furtherexplained by his probable

Book V and theOriginalScope of De institutionemusica

Having completed the translationof Nicomachus' Eisagoge in Book

Potestenimalia quoqueessedivisio,in qua nonunustantummodo

more exact translation in the De institutionearithmeticabegins after

Tout homme court

Cet homme-cicourt et cet homme-l court, etc.

ainsi que les propositions:

ainsi que les formules:

Le parallle est certes assez frappant pour suggrer une analogie

Cependant cette thse a t vivement conteste par Matthews [2].

Aucune chimren'est bleue

et comment le pourrait-on en effet,puisqu'on ne peut en dsigner

Aucune chose blanche n'est colore

tant vraie, implique