Studies in History and Philosophy of Science 40 (2009) 267–280

Contents lists available at ScienceDirect

Studies in History and Philosophy of Science

journal homepage: www.elsevier.com/locate/shpsa

Gravity and Newton’s Substance Counting Problem

Hylarie Kochiras
Department of Philosophy, University at Buffalo, Buffalo, NY 14260, USA

a r t i c l e i n f o a b s t r a c t

Article history: A striking feature of Newton’s thought is the very broad reach of his empiricism, potentially extending
Received 1 July 2008 even to immaterial substances, including God, minds, and should one exist, a non-perceiving immaterial
Received in revised form 28 February 2009 medium. Yet Newton is also drawn to certain metaphysical principles—most notably the principle that
matter cannot act where it is not—and this second, rationalist feature of his thought is most pronounced
in his struggle to discover ‘gravity’s cause’. The causal problem remains vexing, for he neither invokes pri-
Keywords: mary causation, nor accepts action at a distance by locating active powers in matter. To the extent that he
Action at a distance
is drawn to metaphysical principles, then, the causal problem is that of discovering some non-perceiving
Active principles
Gravity
immaterial medium. Yet Newton’s thought has a third striking feature, one with roots in the other two:
Immaterial substance he allows that substances of different kinds might simultaneously occupy the very same region of space. I
Individuation elicit the implications of these three features. For Newton to insist upon all three would transform the
Isaac Newton causal question about gravity into an insoluble problem about apportioning active powers. More
seriously, it would undermine his means of individuating substances, provoking what I call ‘Newton’s
Substance Counting Problem’.
Ó 2009 Elsevier Ltd. All rights reserved.

When citing this paper, please use the full journal title Studies in History and Philosophy of Science

1. Introduction
For many seventeenth-century thinkers, the Scholastic
commonplace that no two bodies can be in the same place at the
same time translates neatly into the general principle that no two
things can be in the same place at the same time; this is so both
for the Hobbesian, whose ontology admits only matter, and for
the Cartesian, who, admitting spirits as well as matter, takes spirits
to lack spatial extension. It is an arresting feature of Newton’s
thought that no such translation can be made. Newton takes spatial
extension to be the condition of any substance’s existence, and
admitting the traditional objects of metaphysics into his ontol-
ogy—the immaterial spirits, God, and minds—he takes them to be
extended in space.1 If there are additional spirits—some non-perceiv-
ing immaterial substances, for instance—they too must be extended.
Two things can be in the same place at the same time, then, if those
two things are God and any material body, or a mind and a human
body. And on empirical grounds, Newton allows the general possibil-
ity that any immaterial substance might co-occupy place with a sub-
stance of another kind; the phenomena do not rule this out.
A second arresting feature of Newton’s thought is the broad
scope he assigns to empiricism—so broad that it extends to the tra-
ditional objects of metaphysics. ‘To treat of God from phenomena is
certainly part of natural philosophy’,2 he writes in the General Scho-
lium. And quite generally, the traditional objects of metaphysics per-
tain to physics, as long as they are investigated through phenomena.
What is taught in metaphysics, if it is derived from divine
revelation, is religion; if it is derived from phenomena through
the five external senses, it pertains to physics [‘ad Physicam
pertinet’]; if it is derived from knowledge of the internal actions
of our mind through the sense of reflection, it is only philosophy
about the human mind and its ideas as internal phenomena
likewise pertain to physics.3

E-mail address: kochiras@buffalo.edu

1
Newton argues in the General Scholium (Newton, 1999, p. 941), that God is substantially omnipresent, an issue that I address in more detail below. In De gravitatione, he
indicates that minds too are extended, being ‘diffused through space’ (2004, p. 26).
2
Newton (1999), pp. 942–943.
3
Unpublished Preface to the Principia, Cambridge University Library (ULC), MS Add. 3968, fol. 109 (in Cohen, 1999, p. 54).

0039-3681/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.shpsa.2009.07.003
268 H. Kochiras / Studies in History and Philosophy of Science 40 (2009) 267–280

So while Newton’s ontology admits immaterial substances, he does
not reserve such substances for rationalist consideration, but places
them in the domain of natural philosophy, potentially within reach
of empirical methods.
Yet despite his empiricism, Newton appears strongly inclined
toward certain metaphysical principles—that is, toward certain
rationalist or non-empirical principles. (Such principles are com-
monly referred to as metaphysical,4 and I shall employ that termi-
nology, though it is to some extent confounded by Newton’s own
use of the term ‘metaphysical’, specifically by his view that the tradi-
tional objects of metaphysics may be amenable to empirical meth-
ods.) The Principia implies the possibility that matter acts distantly,
such that the sun and planets attract one another across empty space
without any intervening medium to convey the effect. Yet Newton
considers such a possibility absurd, as he writes to colleague and theo-
logian Richard Bentley several years after the Principia’s publication:
It is inconceivable that inanimate brute matter should, without
the mediation of something else, which is not material, operate
upon and affect other matter without mutual contact, as it must
be, if gravitation in the sense of Epicurus, be essential and
inherent in it. And this is one reason why I desired you would
not ascribe innate gravity to me. That gravity should be innate,
inherent, and essential to matter, so that one body may act
upon another at a distance through a vacuum without the
mediation of anything else, by and through which their action
and force may be conveyed from one to another, is to me so
great an absurdity, that I believe no man who has in philoso-
phical matters a competent faculty of thinking can ever fall into
it. Gravity must be caused by an agent acting constantly accord-
ing to certain laws; but whether this agent be material or
immaterial, I have left to the consideration of my readers.5
Here Newton appears to affirm the Scholastic metaphysical princi-
ple: matter cannot act where it is not. This apparent inclination toward
such metaphysical principles is a third notable feature of Newton’s
thought.
The question of how exactly gravitational effects are produced
is a vexing problem for Newton, and one in which these three
elements of his thought converge. Or rather collide, for, as I shall
argue, it would not be possible for Newton to resolve the causal
problem about gravity as he conceives it while insisting upon all
three elements. To do so would transform the question about
gravity’s causal means of operation into an insoluble problem
about apportioning active principles or powers. More seriously, it
would undermine his means of individuating substances, provok-
ing what I shall call ‘Newton’s Substance Counting Problem’.
To establish this conclusion, I begin in the first section by
examining the nature and scope of Newton’s empiricism, as it re-
lates to his concept of matter and to substance generally. Newton’s
rationalist side emerges in the three subsequent sections, which
consider potential resolutions to the causal problem about gravity.
As Newton’s stance toward the possibility of action at a distance
has been both a question of longstanding interest and the focus
of several recent, influential works, my arguments in these sections
will be situated in some familiar terrain. Yet part of my purpose
will be to elucidate the conflict generated by the three aforemen-
tioned strains of Newton’s thought, and then to show how that
conflict produces the problem about individuation that I have
identified. That problem is the focus of the paper’s final section.
By way of preliminaries, I note that the term ‘distance forces’ is
intended to be neutral with respect to causal questions. It refers to
forces that operate between spatially separated material bodies,
and which therefore appear to involve action at a distance, but
the term itself implies no answer to the question of what gravity’s
full causal story might be. Also, my discussion considers Newton’s
speculations as well as those propositions that he asserts. In his
mature work, he draws that distinction sharply, confining hypo-
theses—whose proper role is only to furnish experiments6—to
unpublished manuscripts or to the queries of the Opticks. Finally,
my analysis focuses upon Newton’s internal reasoning, which is to
say the arguments and concepts that he employs. So while I do men-
tion certain concepts that Newton lacks, such as energy, and the role
that that absence plays in his reasoning, I consider Newton’s causal
problem about gravity in terms of his own concepts.
2. Substance and empiricism
In an unpublished Preface to the Principia, Newton writes:
To dispute about the objects of ideas except insofar as they are
phenomena is dreaming. In all philosophy we must begin from
phenomena and admit no principles of things, no causes, no expla-
nations, except those which are established through phenomena.7
Notorious among dreamers are, of course, Descartes and Leibniz. The
a priori or rationalist constraints that they impose upon their phys-
ical theories preclude any revision, even if countervailing evidence
should come to light. Descartes’s laws of nature are not revisable,
nor is his claim that the quantity of motion in the universe remains
fixed, for these are all derived from God’s immutable nature. Simi-
larly, Leibniz’s metaphysical principle of sufficient reason tells him
that atomism must be false; and the principle that bodies can act
upon one another only when contiguous grounds his claim that a
material vortex exists, for the gravitational effects in observable
bodies must be produced by contact. Because such metaphysical
principles are sacrosanct, any claims correctly derived from them in-
herit that exalted status. Newton’s empirical approach, by contrast,
denies non-mathematical propositions a sacrosanct status, and
instead is fundamentally a program of revisability.8 Rule 4 of the
Principia makes room for limited revisions, by allowing that further
investigation may reveal the need to make the propositions ‘more

4
See, for example, DiSalle (2002); Janiak (2008), pp. 76, 173.
5
Newton’s fourth letter to Bentley, 25 February 1692/1693 (Newton, 2004, pp. 102–103). This passage is of course a focal point in the debate about whether Newton ever
accepted action at a distance and, in connection with that debate, one might wonder how to reconcile the quoted passage’s opening remark (which suggests that without mutual
contact, matter could act only with the mediation of something immaterial) with its final remark (in which Newton indicates that he has left his readers to decide whether
gravity’s agent is material or immaterial). Accordingly, John Henry (correspondence of 10 December 2007) raised the following objection to my interpretation:

The standard reading of the passage in the letter to Bentley makes Newton seem at best distracted and at worst an idiot for saying that the mediating agent of gravity is
immaterial and then a couple of sentences later [saying] he doesn’t know whether the agent is material or immaterial. This is a problem for the standard reading . . . but
it isn’t a problem for me because I say that he first claims that an immaterial God gives matter the power to attract other matter, and that a couple of sentences later he
is no longer talking about God but is talking about the power that God gave to the matter—that power, or the way it operates, might be either material or immaterial.

Yet interpreting Newton as denying action at a distance by no means forces me to the view that he penned the first and last sentences in distraction. For, as I read them, Newton
states his own view in the first sentence, while in the last he describes what he did in the Principia, which was to refrain from stating his own view.
6
The remark appears in Newton’s letter to Oldenburg, for Pardies, 10 June, 1672 (in Newton, 1959–1971, Vol. 1, p. 169). Newton’s thoughts on hypotheses were by no means
static, and have been the object of an extensive literature. See in particular Cohen (1966), pp. 163–184.
7
Unpublished Preface to the Principia, ULC MS Add. 3968, fol. 109 (in Cohen, 1999, p. 54).
8
For an extensive discussion of Newton’s empiricism, conceived largely in terms of revisability, see Stein (2002), pp. 256–307; DiSalle (2006).
 H. Kochiras / Studies in History and Philosophy of Science 40 (2009) 267–280 269

exact’, or show that they are ‘liable to exceptions’.9 And Newton’s
preface to the reader allows for more dramatic revisions; he writes
there not only that his work may shed light upon this mode of philos-
ophizing, but also that it may lead to ‘some truer one’.10
Newton’s empiricism is well evident in his concept of body or
matter. The fundamentals of his concept are developed in De
gravitatione, an early manuscript in which he positions himself
dramatically against Descartes—Newton will ‘venture to dispose
of his fictions’.11 Among these fictions are the Cartesian identifica-
tion of matter with extension, and the notion that we can know
the essences of substances, by reason. Mere extension does not pro-
duce sensations in us whereas, Newton holds, a thing must have that
power if it is to qualify as a body.12 Indeed, it is from perceived prop-
erties that his account of body begins. We call something a body if it
can produce sensations in us, if it is mobile, and if it is impenetrable,
having resistance and being reflected in accordance with certain
laws.13 These properties reappear in Rule 3 of the 1713 Principia,14
and there Newton emphasizes, contra Descartes, that all of them,
even extension, are known by sense rather than reason.
As Newton constructs his account of body, he is able to avoid
the Aristotelian Scholastic notion of prime matter, that epistemi-
cally inaccessible substrate said to persist through qualitative
changes. Newton’s own account eliminates this notion by associat-
ing perceived properties with determined regions of space:
It is not necessary that we suppose some unintelligible substance
to exist in which as subject there may be an inherent substantial
form . . . Extension takes the place of the substantial subject in
which the form of the body is conserved by the divine will.15
To be precise, however, we cannot say that these determined
regions of space associated with perceived properties are bodies
themselves, for De gravitatione gives an account of the things that
we call bodies, not an account of what bodies essentially are.16 For
while Newton allows that matter has some ‘essential and metaphys-
ical constitution’, he does not claim to know that essential nature.17
This early view of matter persists, and, decades later, Newton writes
in the General Scholium:
We certainly do not know what is the substance of any thing.
We see only the shapes and colors of bodies, we hear only their
sounds, we touch only their external surfaces . . . But there is no
direct sense and there are no indirect reflected actions by which
we know innermost substances.18
Since we have access only to the properties we perceive, and lack
knowledge of innermost substances, we would not be able to
distinguish between two entities that differed in their essential or
metaphysical constitution while sharing all perceptible proper-
ties.19 The view that Newton articulates with respect to matter is
one that he extends to God and to substances generally, and he
continues the above-quoted passages from the General Scholium
as follows: ‘Much less do we have an idea of the substance of God.
We know him only by his properties and attributes and by the wisest
and best construction of things and their final causes’.20 An explicit
statement of this same stance toward substances generally appears
in the unpublished Draft conclusion to the General Scholium:
We do not know the substances of things. We have no idea of
them . . . From the properties we infer that the things them-
selves exist and we call them substances: but we do not have
any more idea of substances than a blind man has of colors.21
It is in this same draft that Newton allows, on empirical grounds,
that substances of different kinds might penetrate one another,
which is to say that they might co-occupy regions of space:
That bodies do not penetrate each other we gather from the
phenomena alone; that substances of different kinds do not
penetrate each other does not at all appear from the phenomena.

9
Newton (1999), p. 796.
10
Newton, his author’s preface to the reader, Principia (1999), p. 383.
11
De gravitatione (Newton, 2004, p. 14). Although the fundamentals of Newton’s concept of body are developed in this early manuscript, he does not yet have the concept of
mass, as distinguished from weight. Two events lead him to draw the distinction: the discovery of Jean Richer’s scientific expedition, that weight is not invariant across terrestrial
locations but instead varies with latitude; and Newton’s conclusion in studying the 1680 comet, that the comet was affected by planets as well as the sun, and in proportion to
their quantity of matter. On both of these events, see Cohen (1999), pp. 19–20.
12
Descartes of course agrees that bodies have the power to excite sensations; this is part of his argument for the existence of bodies, in Meditation VI. (There, Descartes writes:
‘Now there is in me a certain passive faculty for sensing, that is, a faculty for receiving and knowing the ideas of sensible things. But I could make no use of it unless a certain active
faculty for producing or bringing about those ideas were either in me or in something else’. CSM, Vol. 2, p. 55.) Newton’s point against Descartes is that mere extension cannot
excite sensations.
13
In De gravitatione, Newton writes:

We can define bodies as determined quantities of extension which omnipresent God endows with certain conditions. These conditions are: (1) that they be mobile . . . (2) that two
of this kind cannot coincide anywhere, that is, that they may be impenetrable, and hence that oppositions obstruct their mutual motions and they are reflected in accord with
certain laws; (3) that they can excite various perceptions of the senses . . . in created minds and conversely be moved by them. (Newton, 2004, pp. 28–29)

14
Since De gravitatione’s discussion of bodies antedates Newton’s theory of gravity and thus charges of action at a distance, it does not contain Rule 3’s caveat about gravity.
According to that caveat, by ‘inherent force’, Newton intends only the force of inertia, since it is immutable. He is ‘by no means affirming that gravity is essential to bodies’, as
gravity diminishes with distance (Newton, 1999, p. 796).
15
De gravitatione (Newton, 2004, p. 29).
16
Similarly, the Principia defines quantity of matter (mass), and Rule 3 states the properties of matter, without defining matter itself.
17
De gravitatione (Newton, 2004, p. 27). On this point, see Stein (2002), pp. 290–291; as Stein reads Newton, ‘We cannot know with certainty the ultimate constitution of things’.
18
Newton (1999), p. 942.
19
This view is further suggested by the rest of the passage from De gravitatione:

Although it scarcely seems credible that God could create beings similar to bodies which display all their actions and exhibit all their phenomena, and yet would not be
bodies in essential and metaphysical constitution, as I have no clear and distinct perception of this matter I should not dare to affirm the contrary, and hence I am
reluctant to say positively what the nature of bodies is, but I would rather describe a certain kind of being similar in every way to bodies. (Newton, 2004, p. 27)

The view suggested here is one that Newton may never have abandoned, for it reappears in the Query 31 remark that God might vary the laws of nature by varying the forces
and properties of matter. One might object, however, by arguing that to vary the forces and properties of matter would be to create a substance that is not matter. Also, some
commentators would object to my suggestion that Newton’s skeptical stance toward matter’s essence extends past De gravitatione; a number of commentators interpret the
Principia as asserting the essential qualities of matter. (And here I intend a strong sense of ‘essential’, such that to be an entity of a given kind, the entity must possess the
essential property. I am not referring to the Scholastic sense of ‘essential’ that I take Newton to mean in Rule 3, which is, in McMullin’s terms, ‘intensity-invariant’.) For a dis-
cussion of matter’s essence in that stronger sense, see Janiak (2008), pp. 118–129.
20
Newton (1999), p. 942.
21
Draft conclusion for the General Scholium, ULC MS Add. 3965, fols. 361–362 (Newton, 1962, p. 360). Newton left this Draft conclusion unpublished, replacing it with his far
more compressed General Scholium.
270 H. Kochiras / Studies in History and Philosophy of Science 40 (2009) 267–280

And we ought not rashly to assert that which cannot be inferred
from the phenomena.22
We find in Newton, then, a strongly empirical approach to
matter and to substance generally. The approach permits an
account of body (or the things we classify as bodies) without
recourse to an epistemically inaccessible substrate, since proper-
ties are associated with the regions of space in which they are
perceived. Empiricism reaches potentially to all substances, includ-
ing the traditional objects of metaphysics, God and minds. It
grounds the possibility that immaterial substances might share
place with matter, as the phenomena do not exclude the possibility
that two things could be in the same place at the same time.23
Yet Newton’s empirical methods and their results notoriously
point toward matter acting distantly. The gravitational force holds
between spatially separated bodies, and the Principia’s second book
rules out any dense material medium; any medium with sufficient
resistance to move the planets by impact would cause deviations
from Kepler’s idealized elliptical orbits and, contrary to observa-
tions, would eventually bring the planets to a stop.24 As Newton
writes to Leibniz, ‘The heavens are to be stripped as far as may be
of all matter, lest the motions of planets and comets be hindered
or rendered irregular’.25 While it is possible that some other sort
of medium—perhaps something immaterial—exists to convey the
effects, Newton has failed to discover any such medium. And if such
a medium is nothing more than a bare possibility (one I shall discuss
in a subsequent section), then that by itself need not prevent Newton
from concluding that matter acts distantly. For according to Rule 4, a
proposition grounded in induction should be regarded as true or
very nearly true unless and until further evidence indicates a need
for revision; the proposition should not be ‘nullified by hypotheses’,
such as bare possibilities unsupported by induction. The application
of Rule 4 to Newton’s findings, then, at least opens the way toward
concluding that matter acts distantly. But while Newton treats the
gravitational force as causally efficacious—it is the force ‘by which
celestial bodies are kept in their orbits’,26 it ‘really exists’, and it ‘suf-
fices for all the motions of the heavenly bodies and of our sea’ (&ad
corporum caelestium & mari nostri omnes sufficiat)27—he refrains from
drawing the conclusion that Kant would later draw, instead writing
the General Scholium’s most famous lines: ‘I have not yet assigned a
cause to gravity . . . I have not as yet been able to deduce from
phenomena the reason for these properties of gravity, and I do not
feign hypotheses’.28 It is the causal problem about gravity that brings
Newton’s rationalist inclinations to the fore.
3. Gravity and the hypothesis of primary causation
In trying to discover the causal means by which gravitational
effects are produced, Newton has a full slate of options, for again
he draws no sharp distinction between physics and metaphysics.
For Leibniz, the two are divided; though he allows bodies and forces
at the phenomenal level of description needed for physics, he ulti-
mately restricts his ontology to percipient monads, and restricts
causal efficacy to God. Since Newton allows immaterial spirits into
natural philosophy’s domain, no substance is precluded, prior to
investigation, from being a candidate in gravity’s causal story.
One candidate for the cause of gravitational effects, then, is God,
that immaterial being who is substantially present throughout all
of space.29 If Newton attributed gravitational effects to primary

22
Ibid.
23
The view is not unique to Newton. More, who was influential upon Newton (Westfall, 1980, pp. 56 n. 39, 89, 97, 304, 348–349; McMullin, 1978, pp. 43–44), also took a
spatially extended God to share place with matter. As he writes to Descartes:

There is a quite obvious distinction between divine and corporeal nature, since divine nature can penetrate corporeal nature, but corporeal nature cannot penetrate
itself . . . These few considerations suffice to demonstrate that it would have been much safer to have defined matter as tangible substance, rather than as an extended
thing. (More to Descartes, AT V 239–40, in Garber, 1992, p. 145)

The view appears also in Locke: ‘We have the Ideas but of three sorts of Substances; 1. God. 2. Finite Intelligences. 3. Bodies . . . These three sorts of Substances, as we term them,
do not exclude one another out of the same place’ (Essay, II.xxvii.2; Locke, 1975, p. 329). For a contemporary perspective, see Wiggins (1968).
24
If Descartes’s material fluid were dense enough to push the planets, it would produce enough resistance to make the planets deviate from Kepler’s idealized elliptical paths,
just as terrestrial projectiles do; indeed, it would eventually bring the planets to a stop. Yet the planets do not deviate from those paths, and they are not brought to a stop, so such
a dense medium must not exist. If, on the other hand, the material fluid were too rare to produce deviations from the elliptical paths, it would not be capable of pushing the
planets. Thus the existence of a medium having sufficient resistance to push the planets is contradicted by observations, while a medium rare enough to conform to the
observations could not perform the function Descartes assigns to it. This is not to suggest that Newton and the Cartesians share much conceptual ground. Descartes lacked
the concept of mass, indicating in the Principles of philosophy (II.4; CSM, Vol. 1, p. 224) that a body without resistance is yet a body, and he supposed that the fluid in the vortex
could push the planets without the impacts that presume void space.
25
Newton to Leibniz, 1693 (in Newton, 2004, pp. 108–109).
26
Principia, Book III, Proposition 5, Scholium (Newton, 1999, p. 806).
27
I have employed Janiak’s translation, rather than that by Cohen and Whitman, in accordance with his complaint that their translation introduces the term ‘explain’, which has
no correlate in Newton’s original text, and which undercuts Newton’s presentation of the force as causally efficacious. (As Cohen and Whitman translate the passage, gravity ‘is
sufficient to explain all the motions of the heavenly bodies’; Newton, 1999, p. 943. See Janiak’s discussion, 2007, p. 129; 2008, p. 55.) I concur that while Newton cannot provide
the causal story, he nonetheless regards gravity as causally efficacious; this seems evident from the passages cited. The question of how Newton could take the force to be causally
efficacious without knowing its physical means of operation is tackled by Janiak in the aforementioned works. To summarize some central points, Janiak argues that Newton takes
himself to have identified a real force, as opposed to a mere calculating device, because he has shown that (i) a wide range of phenomena that previously seemed disparate in fact
have the same cause, (ii) mass and distance are the only salient variables in the causal chain, and (iii) gravity is not a mechanical cause in the commonly accepted sense of
operating by surface action. Newton has identified the force’s physical species—which Janiak, like Stein, takes to be given by the force’s law—but has not found the force’s physical
basis. But while Newton’s physical account is on that ground incomplete, he has nonetheless discovered how to measure the force—which is a physical quantity—by measuring
other physical quantities: mass and distance. In support of his claim that Newton can consider gravity a real, causally efficacious, physical quantity without knowing its physical
basis, Janiak points to a similar situation in Newton’s work in optics. Even without being able to say what light is—that is, whether it is a particle or a wave and thus whether it is a
substance or the property of some medium—we can say that it exists, because we can measure it; we can measure its speed, for instance.
28
Newton (1999), p. 943. It is worth noting that Newton is concerned only with the specific means by which the gravitational force is causally efficacious, not with the notion of
causation itself. Like Kepler, but unlike some of his successors, most notably Hume, Newton does not take causation to be an epistemologically troubling concept; significantly, he
sees no need to define it in the Principia.
29
The prevailing view was that God is virtually omnipresent—it is only his virtus or active power that is everywhere present. As Descartes, one notable representative of that
view, writes to More:

In God, in angels, and in our mind I understand no extension of substance but at most the extension of power, so that an angel can exercise its power now in a greater,
now in a lesser part of a corporeal substance. For if there were no body, I should think that there would also be no space with which an angel or God would be
coextensive. (AT V 342 [K249], in Garber, 1992, p. 146. See also Descartes to More, 5 February 1649, AT V 270; CSMK, p. 361.)

Against the prevailing view, Newton will argue in the General Scholium that it is not only God’s virtus or active power that is omnipresent, but God himself. I shall have more to
say about Newton’s position subsequently.
 H. Kochiras / Studies in History and Philosophy of Science 40 (2009) 267–280 271

causation—that is, to God’s direct action, rather than to the secon-
dary causes that the deity institutes in the natural (created)
order—he would avoid the possibility that matter acts distantly.
For God is ‘everywhere present to the things themselves’,30 and thus
at the site of all accelerating bodies. It is of course uncontroversial
that Newton’s God sometimes acts directly in the world; he acts di-
rectly to reform the planetary orbits, for instance, which the mutual
actions of planets and comets render irregular over time.31
Some commentators take God’s role to be greater, however,
arguing that Newton is simply being coy when, in the earlier-
quoted remarks to Bentley, he leaves his readers to determine
whether the agent producing gravitational effects is material or
immaterial. According to these commentators, he privately attri-
butes all gravitational phenomena directly to God. The view has
a number of variants. Hawes limits the range of phenomena for
which Newton accepted primary causation, arguing that Newton
attributed gravitational effects directly to God, but allowed action
at a distance for the electric force.32 McGuire limits the time during
which Newton accepted primary causation; he argues that Newton
held the view during the post-Principia period, but after the 1706
Optice pushed distance forces into the realm of secondary causation,
believing that gravity and the other, speculative distance forces had
causes within the natural order.33 Westfall, by contrast, sees no such
limitations. He holds that, privately, Newton believed that all of the
phenomena associated with distance forces— including both electri-
cal and gravitational attractions —were in reality the effects of God’s
direct and immediate action. Thus Westfall moves Newton closer to
Malebranche, for on his interpretation, the bulk of the story of causal
efficacy is a story of God’s primary causation; apart from the ‘passive
principles’ of the vis inertiae and the three laws of motion arising
from it, which Newton believes to explain very little motion,34
secondary causes are rejected.
Newton’s unpublished manuscripts certainly contain some sug-
gestive passages. In one passage from the 1690s, Newton writes
that spatially separated bodies cannot attract one another except
by the mediation of an ‘active principle’ to propagate the force;
and he then approvingly ascribes to the ancients the belief that this
active principle is the infinite spirit, God.35 Also suggestive is an
unpublished corollary, drafted in connection with the Principia’s
Proposition 6, a proposition asserting that all bodies gravitate
toward each of the planets: ‘Corol. 9. There exists an infinite and
omnipresent spirit in which matter is moved according to mathe-
matical laws’.36 Yet despite his approbation for the ancients’ belief
that God is the active principle, Newton does not quite assert it him-
self. And a remark in the General Scholium indicates that matter
moves in the infinite spirit, not that it is moved by that spirit: ‘In
him all things are contained and move, but he does not act on them
nor they on him’37 (In ipso continentur & moventur universa, sed sine
mutua passione. Deus nihil patitur ex corporum motibus: illa nullam
sentiunt resistentiam ex omnipraesentia dei).
Meanwhile, Newton repeatedly treats the gravitational force as
something independent of God: that is, as something requiring
only divine concurrence. More specifically, he treats the force as
something that God employs as a tool, and in some ways must
even oppose. Query 31, for instance, presents distance forces,
including gravity, as God’s tools. The active principles that Newton
associates with distance forces are ‘general laws of nature, by
which the things themselves are formed’; and, he continues, refer-
ring to his speculative interparticle force of cohesion, God com-
poses bodies ‘by the help of these principles’.38 Thus in Query 31,
distance forces, or at least the active principles associated with them,
are secondary causes, which God employs to achieve his ends.
Even during the 1690s, the post-Principia period, Newton treats
the gravitational force as distinct from God in that God must
oppose it. Once the notion of an attractive force between the sun
and planets was accepted, the following question arose. Why does
the universe not collapse—why are the fixed stars39 and the sun not
pulled together by the gravitational force? When Bentley poses the
question, Newton replies that God has placed the stars at immense
distances in infinite space,40 a position he reiterates in the General

30
Opticks, Query 31 (Newton, 1952, p. 403).
31
Ibid., p. 402, but Query 23 in the version of the essay upon which Leibniz was commenting says: ‘Some inconsiderable Irregularities . . . may have risen from the mutual
Actions of Comets and Planets upon one another, and . . . will be apt to increase, till this System wants a Reformation’. Newton’s suggestion here was the probable provocation to
Leibniz’s charge, in his 1715 Letter 1, that Newton’s God is like an imperfect watchmaker. See Alexander (1956), p. 11 n. a.
32
Hawes, 1968b, p. 205.
33
See McGuire (1968), pp. 207–208.
34
Opticks, Query 31 (Newton, 1952, p. 399).
35
See ULC MS Add. 3965.6, fol. 269 (in Westfall, 1971, pp. 397–398):

For two planets separated from each other by a great expanse of void do not mutually attract each other by any force of gravity or act on each other in any way except by
the mediation of some active principle that stands between them by means of which force is propagated from one to the other. [According to the opinion of the ancients,
this medium was not corporeal since they held that all bodies by their very natures were heavy and that atoms themselves fall through empty space toward the earth by
the eternal force of their nature without being pushed by other bodies.] Therefore the ancients who grasped the mystical philosophy more correctly taught that a certain
infinite spirit pervades all space, and contains and vivifies the entire world; and this supreme spirit was their numen; according to the poet cited by the Apostle: In him
we live and move and have our being. Hence the omnipresent God is recognized, and by the Jews is called ‘place’. To the mystical philosophers, however, Pan was that
supreme numen . . . By this symbol, the philosophers taught that matter is moved in that infinite spirit and by it is driven, not at random, but harmonically, or according
to the harmonic proportions as I have just explained.

McGuire dates this passage to ‘a manuscript probably related to the ‘‘classical” scholia of the abortive 1690’s edition of the Principia’. See also McGuire’s discussion (1968), p.
196.
36
ULC MS Add. 3965.6, fol. 266v (in Westfall, 1971, p. 509).
37
Newton (1999), p. 941.
38
Newton (1952), p. 401.
39
The stars were thought to lack the planets’ transverse motions, and thus were considered fixed. See, for example, Phenomenon 1 at the beginning of Principia, Book III: ‘The
circumjovial planets [or satellites of Jupiter], by radii drawn to the center of Jupiter, describe areas proportional to the times, and their periodic times—the fixed stars being at
rest—are as the 3/2 powers of their distances from that center’ (Newton, 1999, p. 797).
40
In his first letter to Bentley, 10 December 1692, Newton writes that collapse would occur only in a finite space, not in an infinite one:

If the matter of our sun and planets, and all the matter of the universe, were evenly scattered throughout all the heavens, and every particle had an innate gravity towards
azll the rest, and the whole space, throughout which this matter was scattered, was but finite; the matter on the outside of this space would by its gravity tend towards all
the matter on the inside, and by consequence fall down into the middle of the whole space, and there compose one great spherical mass. But if the matter was evenly
dispersed throughout an infinite space, it would never convene into one mass, but some of it would convene into one mass and some into another, so as to make an infinite
number of great masses, scattered at great distances from one to another throughout all that infinite space. But how the matter should divide itself into two sorts, and that
part of it, which is fit to compose a shining body, should fall down into one mass and make a sun, and the rest, which is fit to compose an opaque body, should coalesce, not
into one great body, like the shining matter, but into many little ones . . . I do not think explicable by mere natural causes, but am forced to ascribe it to the counsel and
contrivance of a voluntary agent. (Newton, 2004, pp. 94–95)
272 H. Kochiras / Studies in History and Philosophy of Science 40 (2009) 267–280

Scholium and in Query 28.41 And according to David Gregory’s
Memoranda recording his conversations with Newton, God works
directly to prevent the sun and stars from rushing together.42 If we
can accept the veracity of the Memoranda, Newton is again implying
a distinction between the gravitational force, which could cause the
system to collapse in the absence of divine action, and the divine
action that does in fact prevent it. If there were no physical force
of attraction among the bodies, and instead only God’s action, no
such distinction would need to be drawn. In short, if God were
moving the celestial bodies directly, he would not need to prevent
collapse; he would simply need to refrain from pushing the stars into
the sun. Even during the 1690s, then (the period to which McGuire’s
chronology dates Newton’s strongest attraction to the primary cau-
sation hypothesis), Newton takes the gravitational force to operate
independently of God. In other words, he expects the gravitational
force to operate by secondary causes. This raises the question of
what those secondary causes might be, and to what extent Newton’s
efforts to answer that question are motivated by metaphysical
principles.
4. Gravity and the hypothesis of active matter
One explanation of gravity in terms of secondary causes is of
course the possibility implied by the Principia: material particles
and thus aggregate bodies possess powers of gravitational attrac-
tion. While Newton does not assert that explanation, he at least
appears in some of his speculative writings to attribute active pow-
ers of attraction and repulsions directly to the particles of matter.
Most dramatic is the opening of Query 31
Have not the small Particles of Bodies certain Powers, Virtues, or
Forces, by which they act at a distance, not only upon the Rays
of Light for reflecting, refracting, and inflecting them, but also
upon one another for producing a great Part of the Phenomena
of Nature?43
To accept that material particles possess such powers would be
to allow them a power traditionally assigned only to perceiving
spirits, that of initiating ‘new motions’, as opposed to transferring
existing motion.44 For Newton classifies gravitational effects, along
with a wide range of other phenomena, as new motions—motions
that are introduced into the world anew and that thereby change
the total scalar quantity of motion in the world.45 As is well known,
Newton lacks a concept of energy, and in a much remarked-upon
passage from Query 31, he contests two related Cartesian claims:
that a body can lose motion only by transferring that motion to other
bodies, and that the universe’s total scalar quantity of motion
remains constant.46 Observing that motion is ‘always upon the
decay’, he speculates that the universe must therefore contain some
generative active powers—some ‘active principles’ that replenish the
scalar quantity of motion by generating new motions. Active powers
are of course conceptually distinct from distance forces. There is
nothing incoherent, for instance, about attributing certain chemical
reactions to active principles, while also supposing that the effect

41
See the General Scholium (Newton, 1999, p. 940), and Query 28. In Query 28, Newton infers the existence of a designer from the fact that the stars do not aggregate together.
He does not tell us whether collapse is prevented through direct or indirect action, but again the gravitational force is implied to be something distinct from divine action, in that
some of its effects may need to be countered by divine action: ‘What hinders the fix’d Stars from falling upon one another? . . . Does it not appear from phenomena that there is a
Being incorporeal, intelligent[?]’ (Newton, 1952, p. 369).
42
Gregory attributes to Newton the claim ‘that a continual miracle is needed to prevent the Sun and the fixed stars from rushing together through gravity’. 446 Memoranda of
David Gregory, 5, 6, 7 May 1694 (Newton, 1959–1971, Vol. 3, pp. 334 (original text), 336 (translation)). Interestingly, the notion that there could be a continual miracle is at odds
with the notion of a miracle found in Clarke’s letters; suggesting that all activity derives from God, Clarke classifies unusual events as miracles, and regular events as natural.
43
Opticks, Query 31 (Newton, 1952, pp. 375–376). There are other texts in which, I would argue, an initially compelling interpretation of action at a distance becomes difficult to
sustain upon further reflection. In one draft of the General Scholium, for instance, Newton initially appears to be attributing powers of electrical attraction directly to particles,
writing ‘The particles of very many bodies seem to be endowed with an electric force and to act upon each other at small distances even without friction’ (ULC MS Add. 3965, fols.
357–358, in Newton, 1962, pp. 353–354). On the basis of such passages, Hawes argues that while Newton rejected distant action for gravity, he accepted it for the electric force.
Indeed, it initially appears that if an emitted spirit conveys the force with friction, yet the force acts over shorter distances without friction, then without friction the particles may
be acting upon one another without any intervening medium. However, examination of related texts reveals this not to be the case. Newton holds that there is a spirit that
abounds even without friction, and that friction simply extends its range. So there is a medium, both with and without friction, by which the electrical attraction is conveyed. The
relevant passage is contained in the Draft conclusion to the Principia (ca. 1704–1712):

By these experiments it is fully enough clear that glass at small distances always abounds in electric force, even without friction, and therefore abounds in an electric
spirit which is diffused through its whole body and always surrounds the body with a small atmosphere, but never goes out far into the air unless stirred up by friction.
(In Cohen, 1999, p. 289)

Another intriguing passage appears in the posthumously published A treatise of the system of the world (which Newton originally intended to serve as Book III of the Principia, but
which he replaced, in order to avoid disputes, with a technical version). There, Newton discusses the gravitational force in causal terms, writing that the ‘cause of the action’ is
‘twofold’—it is the ‘disposition of each body’. This language is suggestive of action at a distance, for one might think the disposition in each body just is the power to attract other
matter from a distance. Yet there are also other possibilities; the disposition in each body could be a capacity to be affected by some sort of immaterial medium that would
convey the force without any distant action. Without overreaching, it does not seem possible to interpret the Treatise passage as defending any particular causal conclusion
about gravity, since its main purpose is not to analyze the gravitational force causally, but to establish its relational character. The bulk of Newton’s attention is devoted to
showing that gravity is not a monadic property, and that the attraction between the sun and Jupiter consists not in two actions but in ‘one intermediate action’. (For an opposing
view of the Treatise passage, see Schliesser, 2009. See also Stein, 2002, including his translation of the passage, which I have employed here.)
44
The question of how one body could transfer motion to another was itself a puzzle. More tried to dispel the mystery with a metaphor, suggesting that one body rouses the
other, as if from sleep, and Locke remarked upon the puzzle in his Essay; experience makes the communication of motion by impulse familiar to us, yet the causes and manner of
production are obscure (Essay, II.xxiii.28; Locke, 1975). Leibniz explained the communication of motion in terms of the active vis viva.
45
There are two main cases that Newton considers in Query 31 for reaching his (speculative) conclusion against the Cartesian view, and it is interesting that, in both cases, he
disregards explanations that could have preserved the Cartesian claim that the universe’s quantity of motion remains constant. The first case is a thought experiment, a system of
two globes connected by a rod that rotates as its center of gravity moves in a right line. As Newton analyzes the case, the quantity of motion is greater when the rod is aligned with
the right line motion than it is when perpendicular to it, which shows that motion can be ‘got and lost’. (There is, incidentally, an irony in Newton’s analysis of this case, since it
implies that motion can be generated without force—a charge that he leveled against Descartes’s doctrine of relative motion in De gravitatione.) The only way to reproduce
Newton’s conclusion that the sum of the motions differs for the two orientations is by taking the numerical sum. (I thank Lon Becker for bringing this to my attention.) If Newton
had taken the vector sum, the case would have provided him no reason to think that the universe’s total quantity of motion is variable. In considering elastic collisions, he again
concludes that motion is genuinely lost. While the pre- and post-collision momenta are equal, in accordance with Corollary 3, the world is nonetheless very different after the
collision. In the case of absolutely hard bodies, for instance, their states of absolute motion prior to the collision are succeeded by states of absolute rest (the motion of the earth
being disregarded). Yet, as in the two-globe case, there is an available hypothesis that favors the Cartesian view. Instead of concluding that motion is genuinely lost in collisions
and from friction, Newton could have supposed that motion lost at the macro level is simply transferred to the micro level, as heat. This supposition had precedents, for instance
in Boyle. Yet Newton concludes that motion is genuinely lost, invoking the need for active principles. (As for heat, he associates it with forces that appear to act at a distance, and
concomitantly with active principles.)
46
Opticks, Query 31 (Newton, 1952, pp. 397–402).
 H. Kochiras / Studies in History and Philosophy of Science 40 (2009) 267–280 273

depends upon a contiguity of parts that results once the chemicals
are combined. Nonetheless, Newton does associate active principles
with distance forces.
To accept that material particles possess such active powers,
then, would also be to accept that bodies act upon one another
across distances without any intervening medium to convey the ef-
fect—even across the vast distances separating the sun and Jupiter.
For while Query 31 speculates that short-range forces of attraction
and repulsion may explain such phenomena as the cohesion of
atoms into aggregate bodies, it is the gravitational force that serves
as the model for such short-range forces. Nothing is known of
those speculated short-range forces, whereas Newton has discov-
ered the gravitational force’s mathematical proportions. Since
gravity, being the best-known force, is the model for any additional
forces speculated to act between spatially separated bodies, an
acceptance that those speculated short-range forces involve dis-
tant action without a medium would presume a prior acceptance
of that for gravity.47 Newton is of course unwavering in his belief
that gravity is not essential to matter. During their epistolary
exchange some five years after the Principia’s first publication, he
takes Bentley to task for imputing to him the contrary view,48 and
in the Principia’s 1713 edition he concludes Rule 3 by explicitly stat-
ing that while gravity is universal to matter, it is not inherent or
essential.49
Might he have allowed, however, that matter possesses active
powers of attracting (and repulsing) in some weaker sense?50
However difficult to comprehend, the notion that matter can initiate
new motion, and moreover can do so from afar, was consistent with
his belief in an omnipotent God; and the view that God had endowed
matter with such powers, though unusual, was not unheard of.
Newton was acquainted with this view of an elder contemporary,
Matthew Hale: while matter does not by its nature contain a princi-
ple of self-motion, it has that power inessentially, as a power ‘super-
added’ to it by God.51 Moreover, both Bentley and Locke came to
believe that matter possesses superadded powers to attract spatially
distant matter, without any intervening medium, and did so on the
basis of Newton’s gravitational theory. After an epistolary exchange
with Newton, Bentley concluded, ‘Gravitation . . . is a constant En-
ergy infused into Matter by the Author of all things’.52 And Locke,
who initially insisted that motion can be transmitted only by im-
pulse, eventually was convinced by ‘Mr. Newton’s incomparable
book’53 that bodies can attract one another, even when separated

47
Here I concur with McMullin (1978), p. 51. For an opposing view, see Cohen (1999), pp. 61–62. Cohen argues that Newton’s speculations about short-range interparticle forces
are grounded not in his results about the gravitational force, but in his attempts to understand matter via his alchemical and chemical investigations. Because Cohen argues for
this inverted evidentiary relation between gravity and the speculated short-range forces, he poses the following question: ‘Does the reasonableness of such short-range forces
provide a warrant for belief in the existence of long-range forces acting over huge distances?’. On the view that I have defended, this question cannot be posed.
48
In his second letter to Bentley, Newton writes ‘You sometimes speak of gravity as essential and inherent to matter. Pray do not ascribe that notion to me; for the cause of
gravity is what I do not pretend to know, and therefore would take more time to consider of it’ (Newton, 2004, p. 100).
49
Rule 3 closes with this caveat: ‘I am by no means affirming that gravity is essential to bodies. By inherent force I mean only the force of inertia. This is immutable. Gravity is
diminished as bodies recede from the earth’ (Newton, 1999, pp. 795–796). Newton appears to intend a weaker sense of ‘essential’ here than that intended in De gravitatione,
where he denied knowing the ‘essential and metaphysical’ nature of bodies. For as noted in Section 2, that early view from De gravitatione is reiterated in the General Scholium,
where Newton writes that we do not know innermost substances. Since the General Scholium and Rule 3 make their first appearances simultaneously in the Principia’s second
edition, and since a strong sense of ‘essential’ in Rule 3 would conflict with the General Scholium remark, a weaker sense seems indicated. That weaker sense appears to be one
inherited from the Scholastics: essential properties are those whose intensity cannot be increased or decreased. This sense is consistent with the General Scholium remark and,
concomitantly, with the possibility that Newton sketches in Query 31: namely, that God could perhaps ‘vary the laws of nature’ by creating particles of ‘different densities and
forces’. According to the weaker sense of ‘essential’, the vis inertiae is essential in that it is intensity-invariant, but perhaps not essential in the strong sense that matter could not
exist without it. I mean this qualifying term ‘perhaps’ in two ways. First, I mean it to refer to Newton’s position that we do not know essences—we do not know innermost
substances. Second, I mean it to raise a question about how firmly Newton can adhere to that position, especially with respect to matter’s qualities of spatial extension (which
characterizes all substances), and the vis inertiae; for in a letter to the editor of the Memoirs of literature (discussed below, at the end of Section 5), Newton speculates about a
substance that is immaterial in virtue of lacking the vis inertiae, and this remark suggests that the vis inertiae is essential to matter in the strong sense. A different question raised
by Rule 3 is whether all of the properties that Newton attributes to matter are indeed intensity-invariant. One might argue that hardness, for instance, is not. On this point, see
McMullin (1978), §1.4, especially pp. 22–26.
50
The question of whether Newton allowed action at a distance is of course the focus of debate. The extreme positions are occupied by John Henry (1994) and Andrew Janiak
(2007, 2008) Henry argues that Newton fully accepted action at a distance by accepting active powers as inessential powers of matter, superadded by God. In support of this
interpretation, he takes Newton’s reference to Epicurus, in the above-quoted letter to Bentley, to be highly significant; he reads Newton as rejecting only the Epicurean notion that
gravity is essential to matter. Accordingly, Henry reads Newton’s remark that gravitation requires ‘the mediation of something else which is not material’ to refer to God’s
mediation between the way that matter is essentially, and the way that God actually created it, by superadding active powers. For Henry, the opening of Query 31 is then a
straightforward assertion that material particles do possess active powers that enable them to act distantly. Newton’s qualifying remarks there and in the General Scholium, to
the effect that he does not know gravity’s cause, should not be understood to mean that he denied active powers to matter or denied action at a distance. He rather means to say
that he simply does not know which sort of matter bears these active powers—ordinary matter, or the matter of the aether. In further support of his position, Henry points to
Bentley and Locke as Newton’s spokesmen, for both accepted gravity as a superadded active power of matter. (While I cannot fully engage with Henry’s argument here, I have
argued that the thesis of superaddition cannot easily be squared with Newton’s empiricism, and, when Newton speaks of mediation, contextual remarks indicate that he has
spatial mediation in mind. As for Newton’s disavowal of knowing gravity’s cause, one difficulty I see in Henry’s explanation is that the notion of a material aether is quite
problematic; as noted subsequently, it threatens regress.)
At the other extreme, Janiak argues that Newton never allowed the possibility of distant action by any substance, and proposes the following as a likely explanation. Newton
will not submit his conception of action to empirical considerations that could open the way to revision of his conception of action, because that conception is fundamentally
bound up with his conception of God and concomitantly his conceptions of space and time. To allow that material bodies can act distantly would imply the same possibility for
divine action; yet ‘if God acts non-locally, a principal reason for characterizing God as spatiotemporally ubiquitous is removed’ (2008, p. 174; see also p. 173). Here Janiak refers to
Newton’s reasoning in the General Scholium to the effect that since every power must belong to some substance, and since God’s power is omnipresent, God himself must be
omnipresent (cf. Stein’s quite different interpretation, 2002, p. 270). In my view, Janiak’s position is too strong. While Newton was clearly drawn to a belief in local causation, I see
the aether of Query 21 as an indication that he considered abandoning it; and while Janiak is correct in noting that some further medium might be responsible for the aether’s
action, the possibilities are quite problematic. I have more to say about this subsequently.
51
For Matthew Hale, the claim that matter has superadded active powers is clearly ontological. He writes:

Matter it self simply considered as such, though it be susceptive of Motion (as we daily see) is not the immediate principle of Motion in those subjects that seem to be
self-moving . . . And this entity I call Vis or Virtue activa, superadded to Matter, and giving immediately those motions to it . . . without which, Matter would be stupid,
dull, unactive, and always at rest in its self unless accidentally moved ab extrinseco. (Hale, 1677, pp. 3–5)

Henry argues that Hale’s view, along with Charleton’s, was influential upon Newton; see Henry (1986). Research by Westfall (1962) and by Henry (1986) reveals that Newton
was acquainted with doctrines of superaddition that antedated Locke’s expression of it.
52
Bentley, ‘Eighth Boyle lecture’, preached 5 December 1692 and published 1693 (Newton, 1978, p. 363). The exchange between the two was occasioned by Bentley’s desire to
understand the theological implications of Newton’s theory of gravity, as he prepared some lectures for publication. Prior to its publication, Newton read a draft of Bentley’s
Seventh Boyle lecture (contained in his one extant letter from the exchange), but that draft does not express Bentley’s view about gravity so forthrightly as the quoted passage
from the Eighth lecture.
53
Locke to Stillingfleet (in Locke, 1824, Vol. 4, p. 467).
274 H. Kochiras / Studies in History and Philosophy of Science 40 (2009) 267–280

by a million miles.54 Even one of Newton’s own remarks might be
construed as hinting that matter might possess superadded
active powers. While draft material for the 1706 Optice includes this
seemingly unequivocal remark, ‘Matter is a passive principle &
cannot move itself’,55 it also includes the following: ‘Bodies (alone
considered as long, broad & thick . . .) are passive’.56 One might
interpret the italicized phrase in terms of Hale’s position: while
bodies alone considered are passive, perhaps they have superadded
active powers.
Yet despite this apparent hint toward his compatriots’ views,
both conceptual and textual reasons weigh against the suggestion
that Newton takes matter to possess inessential, superadded active
powers. The suggestion is conceptually problematic because this
ontological thesis of superaddition does not square with Newton’s
empiricism.57 It presumes that we can know the essence of matter,
recognizing certain qualities as essential (in the strong sense that
matter could not exist without it) and others as superadded. Yet as
we have seen, Newton does not attempt to say what matter is ‘essen-
tially and metaphysically’, and he repeats in a number of texts,
including the General Scholium, that we know only external proper-
ties, not innermost substances. It is therefore difficult to see what
grounds, empirical or otherwise, Newton could marshal for pairing
his claim that gravity is universally realized with a claim that gravity
is inessential in the strong sense that matter could exist without it.
The thesis of superaddition therefore should not be attributed to
Newton.
Textual evidence also indicates that while Newton attributes
forces to matter—certainly the gravitational force, and specula-
tively the short-range attractive and repulsive forces he thinks
may explain phenomena such as cohesion and fermentations
(chemical phenomena)—he does not construe those forces as inter-
nal powers of matter to act distantly, without any medium. He
does not take the problem about the means by which those forces
operate to be solved. He denies knowing gravity’s cause,58 both in
the above-quoted passage from the General Scholium and in a letter
responding to Leibniz’s charges.59 Similarly, the dramatic opening of
Query 31, which seems to attribute active powers directly to
material particles, is soon followed by a disclaimer: ‘How these
Attractions may be perform’d, I do not here consider. What I call
Attraction may be perform’d by impulse, or by some other means
unknown to me’.60
These disclaimers should not be interpreted as a ruse designed
to placate critics, most notably Leibniz, who charged Newton with
allowing action at a distance. If Newton is sincere in denying
knowledge of gravity’s cause, then he must also be sincere in deny-
ing knowledge of the causes of Query 31’s speculative short-range
forces of attraction and repulsion. For in formulating hypotheses
about unsolved problems, such as the nature of the force producing
cohesion, we can draw only upon what is known about gravity,
which is to say its mathematical proportions and its efficacy
between spatially distant bodies, not its physical means of produc-
ing those effects.61
And Newton is sincere in denying knowledge of gravity’s cause.
As he writes in the earlier-quoted letter to Bentley, to suppose ‘that
one body may act upon another at a distance through a vacuum
without the mediation of anything else, by and through which
their action or force may be conveyed from one to another’ is
absurd. Instead of attributing attractive powers directly to material

54
Ibid.:

The gravitation of matter towards matter, by ways inconceivable to me, is not only a demonstration that God can, if he pleases, put into bodies powers and ways of
operation, above what can be derived from our idea of body, or can be explained by what we know of matter, but also an unquestionable and every where visible
instance, that he has done so.

55
ULC MS Add. 3970, fol. 619r, written in English, cited in McGuire (1968), p. 171, and identified by him as draft variants of the 1706 Optice. As the passage continues, Newton
points to life and will as active principles, in contrast to matter:

These are passive laws & to affirm that there are no other is to speak against experience. For we find in orselves a power of moving our bodies by or thought. Life & Will
(thinking) are active Principles by wch we move our bodies, & thence arise other laws of motion unknown to us.

56
ULC MS Add. 3970, fols. 255r–256r, draft variants of the 1706 Optice, in ibid., pp. 170–171. As indicated elsewhere in this section, the active powers that Newton hesitates to
attribute directly to matter alone are those of generating the new motions that alter the universe’s total scalar quantity of motion. It is that power that Newton suggests should be
denied to matter when he writes that matter is passive, and not the powers of resistance and of transferring motion that he associates with the laws of motion. This is evident as
the passage continues: ‘By their vis inertiae they continue in their state of moving or resting & receive motion proportional to ye force impressing it & resist as much as they are
resisted; but they cannot move themselves’. The passage closes with a vitalist suggestion: ‘We cannot say that all nature is not alive’.
57
While Locke’s thesis of superaddition may well be epistemic, the version relevant to Newton’s question about gravity’s causal means of operation is, like Hale’s version,
ontological. (For discussions of the epistemic reading of Locke’s thesis of superaddition, see Downing, 2007; Stuart, 1998.)
58
Henry, who argues that Newton accepted distant action by matter, interprets Newton’s expression of uncertainty as meaning that he does not know whether it is ordinary
matter that acts distantly, or the matter of the aether (see Henry, 1994, p. 133). Concomitantly, Henry understands Newton’s talk of mediation in the letter to Bentley as referring
to God’s mediation between matter as it is essentially, and matter as God actually created it, with superadded active powers. I disagree with this, and take Newton to be referring
to spatial mediation, as I have argued. I also disagree with Henry’s explanation of Newton’s expression of uncertainty about gravity’s cause. One difficulty with it is that the notion
of a material aether is problematic, for reasons I discuss subsequently.
59
Newton to the editor of the Memoirs of literature, unpublished, written ca. May 1712 (Newton, 2004) , pp. 116–117:

Certainly God could create planets that should move round of themselves without any other cause than gravity that should prevent their removing through the tangent.
For gravity without a miracle can keep the planets in. And to understand this without knowing the cause of gravity, is as good a progress in philosophy as to
understand . . . the frame of the bones and muscles and their connection in the body of an animal and how the bones are moved by the contracting or dilating of the
muscles without knowing how the muscles are contracted or dilated by the power of the mind, is [in] the philosophy of animal motion.

60
Opticks, Query 31 (Newton, 1952, p. 376).
61
We might see Newton’s speculations about the existence of short-range-distance forces as a preliminary step in an investigation that would, should the forces exist, first
attempt to determine their mathematical proportions, and then attempt to determine their physical means of operation. The mathematical problem is to be tackled before the
physical one, for, as Newton indicates in the Principia, we must first investigate the ‘quantities of forces and their proportions that follow from any conditions that may be
supposed’. It is only after this that we can ‘come down to physics’ and begin to determine, among other things, the ‘physical species’ and ‘physical causes’ of the forces (Principia,
Scholium to Book I, Section 11; Newton, 1999, p. 588). Whereas for gravity he has solved the mathematical problem, leaving only the question of its physical means of operation,
for phenomena such as cohesion the mathematical as well as the physical problems remain outstanding. He cannot say, for instance, whether the attractions and repulsions vary
with the inverse cube (as he finds to be approximately the case for the magnetic force), nor can he say how many forces are at work. (It might be only a single, dual-natured force
that is attractive at some distances but repulsive at others; for he speculates in Query 31 that, as numbers pass from negative to positive, so a force might pass from repulsive to
attractive (Newton, 1952, p. 395). On the question of whether Newton’s remarks genuinely suggest the possibility of such a dual-natured force, see Boas (1952), p. 516; Newton
(1962), pp. 210–211; Hawes (1968a), 127–128.)
 H. Kochiras / Studies in History and Philosophy of Science 40 (2009) 267–280 275

bodies, Newton expects something to mediate between the bodies,
and to mediate spatially. For in this context, to mediate is to con-
vey the force or action from one point in space to another—from
the center of the sun, for instance, to the center of Jupiter.62 There
must be some entity that occupies the space between the bodies,
conveying their gravitational effects upon one another, and to think
otherwise is an absurdity.
Although Newton will go on to write in the General Scholium
that he does not feign hypotheses, he here appears to declare alle-
giance to a metaphysical principle, the Scholastic maxim that mat-
ter cannot act where it is not.63 This is not to say that he asserts any
metaphysical principles. Nothing of the sort appears in the Principia,
for instance, and his thoughts are instead confined to speculative
texts, such as the letter to Bentley. Even in his speculations, he does
not always insist upon the metaphysical principles to which he is
drawn, as we shall see.
Yet he clearly is drawn to certain metaphysical principles. In
addition to the principle that matter cannot act distantly, there
are two more general principles from which that narrower principle
might derive. One is the principle that matter is passive; that is,
material particles alone can act only to the extent codified in the
three laws of motion arising from the vis inertiae, which Newton de-
scribes in Query 31 as ‘passive laws’.64 According to this principle,
material particles can resist and can transfer motion to one another,
but they cannot by themselves generate the new motions that alter
the universe’s scalar quantity of motion. A second fundamental prin-
ciple is that nothing can act distantly; that is, all causation is local,
occurring only by contact. This principle is quite general, applying
to both material and immaterial entities. What constitutes local cau-
sation, however, depends upon the kinds of entities involved in the
interaction. Material bodies cannot interpenetrate,65 so in cases
involving material bodies alone, a causal interaction is local if it
occurs by contact between the bodies’ surfaces. For interactions
involving immaterial entities as well as material ones, however, local
causation involves the co-occupation of place. Divine causation, for
instance, can be local in this sense; God is substantially present
throughout all of infinite space, and while material bodies are
stopped at one another’s surfaces, the immaterial deity can act locally
by penetrating to the very centers of material particles. Indeed, one
reason that God was an intriguing candidate for gravity’s cause is that
the ability to penetrate matter is the condition that Newton sets in
the General Scholium. The force, which acts in proportion to the
quantity of solid matter rather than in proportion to surface area,
‘arises from some cause that penetrates as far as the centers of the sun
and planets, without any diminution of its power to act’.66
But does Newton suppose only that divine causation can be
local, or that it must be? His argument in the General Scholium
suggests the latter, echoing an argument by Henry More. Contest-
ing the prevailing view that only God’s virtus or active power is
omnipresent, and advancing instead his view that God himself is
spatially extended, More wrote to Descartes:
How could he have impressed motion on matter . . . unless he,
as it were, immediately touches the matter of the universe, or
at least, did so once? This could never have happened unless
he were everywhere and occupied every single place. Therefore,
God is extended in his own way and spread out.67
Newton reasons similarly in the General Scholium: ‘He is omnipres-
ent not only virtually but also substantially; for action requires
substance [lit. for active power [virtus] cannot subsist without
substance]’.68 Why does Newton think that, in order for God’s power
to be omnipresent, the substance, God himself, must be present
everywhere in space? The answer, it appears, is that Newton
believes that a substance must be present where it acts.69
Now as argued earlier, Newton does not attribute gravitational
effects to primary causation, but instead seeks some secondary
cause. If his very concept of God is grounded in a principle of local
causation, he will be extremely reluctant to allow powers of

62
This spatial interpretation of the term ‘mediation’ is the prevailing interpretation. (See for example McMullin, 1978, p. 59.) Cf. Henry’s very different interpretation
(1994, p. 130), mentioned earlier in n. 58.
63
For a discussion of the principle’s history, see Hesse (1955).
64
Opticks, Query 31 (Newton, 1952, p. 401). In saying Newton is drawn to a principle that matter is passive, I do not intend anything inconsistent with the vis inertiae and the
laws of motion. (For an extensive discussion of the senses in which Newton takes matter to be active, see McMullin, 1978.) My terminology follows that found in Query 31 and in
related draft passages, such as the following:

Matter is a passive principle & cannot move itself. It continues in its state of moving or resting unless disturbed. It receives motion proportional to the force impressing
it, and resists as much as it is resisted. These are passive laws & to affirm that there are no other is to speak against experience. (ULC MS Add. 3970, fol. 619r, in McGuire,
1968, p. 171)

While I do not take Newton to insist upon the principle that matter is passive (see Section 5’s discussion of the aether hypothesis), I do take him to be drawn to that principle,
primarily because he refrains from locating active principles in matter. In Query 31 (1952), p. 401, the particles of matter ‘have’ the passive principle of the vis inertiae, but they
‘are moved by’ active principles; this language suggests that active principles do not belong to matter, but are instead external to it. A similar uncertainty about the location of
active principles is evident in draft material for the 1706 Optice. In one passage, Newton writes ‘How it [the active principle] stands related to matter is difficult to explain’ (ULC
MS Add. 3970, fol. 620r, in McGuire, 1968, p. 171).
65
See De gravitatione; Rule 3 of the Principia; Query 31 of the Opticks.
66
Newton (1999), p. 943; my emphasis. In a draft for the General Scholium, he makes the same point:

[Gravity] proceeds from some cause that penetrates to the very centers of the Sun and Planets without any diminution of its virtue, and which acts not on the surfaces of
particles alone, but on all matter to the very centre since its action is proportional to the quantity of matter in all bodies. It proceeds from a cause by which the single
particles of bodies act at immense distances with a virtue decreasing in the duplicate ratio of the distances reciprocally. (Newton, 1962, p. 353; my emphasis)

67
More to Descartes, AT V 238–239 (in Garber, 1992, p. 144; his translation). More is careful to note that God is extended ‘in his own way’, so as to avoid suggesting that God is
divisible. In De gravitatione, Newton’s discussion of a spatially extended deity and spatially extended minds includes similar qualifications.
68
Principia, General Scholium (Newton, 1999, p. 941). Clarke represents Newton’s view accurately; then, in his Second reply to Leibniz (§4 in Alexander, 1956, p. 21):

A living substance can only there perceive, where it is present either to the things themselves, (as the omnipresent God is to the whole universe;) or to the images of
things . . . Nothing can any more act, or be acted upon, where it is not present; than it can be, where it is not.

69
Although Newton’s argument in the General Scholium resembles More’s, which is quite explicit about a prohibition against distant action, one also gets the sense that quite
apart from questions about distant action, Newton finds the notion of unextended substance to be unintelligible. How it would be possible for something unextended to be
unified with an extended thing he wonders in De gravitatione’s critique of Cartesian minds:

If we say with Descartes . . . that mind has no extension at all, and so is not substantially present in any extension, that is, exists nowhere . . . this seems the same as if we
were to say that it does not exist, or at least renders its union with body thoroughly unintelligible and impossible. (Newton, 2004, p. 26)
276 H. Kochiras / Studies in History and Philosophy of Science 40 (2009) 267–280

distant action to any created substance. Given his penchant for lo-
cal causation and secondary causes, he has reason to seek some
created (and non-perceiving)70 medium that fulfills the General
Scholium’s condition.
5. Gravity and the hypothesis of a medium
There is a marked contrast between the kind of medium needed
to preserve his metaphysical principles and the kind of medium that
actually figures in Newton’s speculations. Since, as Book II of the
Principia demonstrates, any dense material medium would bring
the planets to a standstill, preserving the principle that all causation
is local would require a medium that was both continuous and
immaterial. Yet while Newton does sometimes appear to associate
those characteristics with the electric spirit,71 that speculated med-
ium is never a candidate for gravity’s cause.72 The medium that New-
ton instead proposes most prominently as a hypothesis for
investigation is the aether of Query 21. According to Query 21’s
hypothesis for gravity, the planets are propelled inward toward the
sun by the pressure of a very rare aether whose density increases with
distance from the sun. The pressure is created by a tremendous repul-
sive force, which operates among the aether’s particles. Not only is
this medium particulate, it most likely is material as well. The aether
that figures elsewhere in the Opticks, for instance in Query 28, is
material, and the particulate nature itself suggests materiality, in part
because it is modeled upon a material substance. Much like its ances-
tor, the material aether of the 1679 Boyle letter, this aether acts in a
way resembling the spring of the air.73
Assuming that Newton intended the hypothesis of Query 21 for
serious investigation,74 this spring or repulsive action makes it
notoriously puzzling. Reductionist goals provide one motivation—
Newton might have hoped to reduce the number of forces by loading
activity into the aether—but the hypothesis has also been inter-
preted as an attempt to avoid unmediated causal interaction be-
tween the celestial bodies. The hypothesis itself seems to involve
distant action, however, as the pressure upon the celestial bodies re-
sults from the force that the spatially separated aethereal particles
exert upon one another. One possible explanation is that Newton
finds distant action inconceivable only if it acts over long ranges,
and that he is willing to accept it if it occurs over relatively short
ranges.75 Yet this suggestion is conceptually problematic. For what
could make the notion of action at a distance over a short range less
unintelligible than that over the very long distances between the
celestial bodies, if not that when considering the former, it is easier
to imagine that there might after all be some medium to bridge the
gap? If action at a distance is objectionable because it is unintelligi-
ble, the range should be irrelevant.
There is another possible explanation of Query 21, one that
reads Newton as keeping faith with the principle of local causation.
He might be relying upon the possibility of some further medium,

70
I add this parenthetical note, distinguishing non-perceiving substances from inanimate ones, because in keeping with the vitalist strain of influence upon his thinking,
Newton does at some points speculate that ‘all nature’ might be ‘alive’. (See n. 56.)
71
It is possible that Newton at least sometimes conceives of the electric spirit as a genuinely continuous, immaterial substance. In his Draft conclusion to the Principia (in Cohen,
1999, p. 287), he refers to it as being ‘continuous’. And in that same text, there is a suggestion that it may be immaterial. Comparing the electric and gravitational forces to
magnetism he writes that the magnetic force is communicated by contact, whereas the other forces are not. Since the magnetic force was widely held in Newton’s time to be
communicated by material effluvia emanating from the magnetic body, the suggestion here is that the electric and gravitational forces are not communicated by material contact.
This leaves open the possibility that they might be communicated by some immaterial substance, and so one might reasonably conclude that, at least in this text, he conceives of
the electric spirit that communicates the electric force as an immaterial spirit. Yet we can at best take this to be one possibility; we cannot conclude that the electric spirit is either
immaterial or continuous. For one thing, Newton’s use of the term ‘spirit’ is not confined to immaterial entities; he also uses the term to refer to substances that lack detectable
resistance—and this could include very rare material substances. (See for example ULC MS Add. 396, fol. 437v, in McMullin, 1978, p. 100.) Additionally, in most contexts at least, it
is clear that when Newton speaks of a particulate substance, he means a material one. It is notable, then, that despite his above-mentioned description of the electric spirit as
‘continuous’, he elsewhere describes the spirit in particulate terms, writing in Query 22 for instance that it is ‘rare and subtle’. And even the text that describes the spirit as
continuous can easily be reconciled with the claim that the spirit is in fact particulate. For in that context, he speculates that the electric spirit might effect heat transfer and, in
referring to the spirit as continuous, he might simply mean that the (particulate) spirit surrounding one body meets that surrounding another body. (See Draft conclusion to the
Principia, in Cohen, 1999, p. 287.)
72
In a number of texts, Newton speculatively attributes a wide variety of phenomena to the electric spirit, but gravitational effects do not appear on the list. See for example the
General Scholium’s abbreviated discussion, the much lengthier Draft conclusion to the Principia (in Cohen, 1999, pp. 287–292), and draft material for the Opticks (ULC MS Add.
3970, fol. 235v, in McGuire, 1968, p. 176). See also the discussion in McGuire, ibid., p. 175, and Hawes (1971), p. 102. Moreover, the electric spirit does not meet the condition
Newton sets upon gravity’s cause. While it does pervade gross bodies, it does not fill the heavens, reaching to the very centers of the sun and planets. It pervades some bodies, but
its reach is limited. It surrounds glass bodies, for instance, but it does not reach far from them, even with friction (Draft conclusion to the Principia, in Cohen, 1999, p. 289). Cohen
takes Newton to be initially optimistic about explaining gravity via the electric spirit, and then to abandon those hopes by the time of the Principia’s third edition (ibid., p. 25).
73
The much earlier aether of the Boyle letter is clearly material; it is ‘much like air in all respects, but far more subtle’ (Newton to Boyle, 28 February 1678/1679, in Newton,
2004, p. 1). In Query 28, Newton implies that the aether is material:

It’s necessary to empty the heavens of all matter, except perhaps some very thin vapours, steams, or effluvia [discharges], arising from the atmospheres of the earth,
planets, and comets, and from such an exceedingly rare aethereal medium as we described above. (Newton, 1952, p. 368)

74
Some commentators deny that Newton ever intended the aether hypothesis for serious investigation. See for example McGuire (1968), p. 187, and Hawes (1968b), pp. 209–
210; Hawes argues that Newton simply wished to end the disputes over Leibniz’s charges. McMullin (1978), p. 151, cites Laudan as a dissenting voice, but adds that Laudan’s
discussion of the Opticks aether focuses upon optical and chemical phenomena, not upon the Query 21 hypothesis about gravitation. The aether is certainly problematic, not least
because the density gradient does not account for universal gravitation. By universal gravitation, accelerations should occur not only between, say, a pair of bodies lying along the
density gradient, such as the sun and the earth, but also between a pair of bodies lying perpendicular to the gradient. If these bodies lying perpendicular to the gradient gravitate
toward one another, it is not in virtue of Query 21’s density gradient, for that impels the bodies only inward, toward the area of lesser aethereal density, not toward one another.
Despite such problems, it is difficult simply to dismiss the aether hypothesis. For one thing, it would be uncharacteristic for Newton, who held so much back and who so carefully
drafted what he did publish, to publish something that he did not intend to be taken seriously. For another, the aether would do nothing to silence Newton’s critics, for, as many
commentators have noted, it at least appears to operate by distant action. John Henry (1986), pp. 347–348, finds a model for Newton’s aether hypothesis in Hooke’s Micrographia.
There, Hooke asks:

whether the Phaenomena of gravity might not by this means be explained, by supposing the Globe of Earth, Water, and Air to be included with a fluid, heterogeneous to
all and each of them, so subtil as not only to be every where interspersed through the Air (or rather the air through it), but to pervade the bodies of Glass, and even the
closest Metals, by which means it may endeavor to detrude all earthly bodies as far from it as it can; and partly thereby, and partly by other of its properties, may move
them towards the Center of the Earth. (Hooke, Micrographia, quoted in Henry, 1986, pp. 347–348)

And for both Hooke and Newton the air served as a model, for by the experiment of placing a sheep’s bladder filled with air into a Torricellian vacuum, the air was known to
have a ‘spring’ by which the bladder would expand.
75
On Cohen’s view (1999, p. 61–62), it is reasonable to take Newton as accepting short-range action while hesitating to allow long-range action. As indicated in a previous note,
he takes Newton’s speculations about short-range forces to be grounded in alchemical and chemical experiments, rather than being grounded analogically in the gravitational
force.
 H. Kochiras / Studies in History and Philosophy of Science 40 (2009) 267–280 277

one that causes the aethereal particles’ repulsive action upon one
another, such that they do not act distantly upon one another.76
Yet this is problematic. Assuming that the Opticks aether is material,
the suggestion that its action might be produced by some further
medium leads to a dilemma. For by universal gravitation, any mate-
rial particles must gravitate. If one supposes the gravitation of the
aethereal particles to be due to some further material medium, one
that has escaped detection because of its rarity, then a regress
ensues; for however rare it may be, in virtue of being material this
further medium must itself gravitate, requiring yet another medium.
(Newton was of course well aware of the regress threat, and in the
first edition of the Principia used the consequence of a regress to
show that one cannot explain gravitational effects by invoking an
aether that either lacks gravity or has a lesser tendency to gravitate,
in proportion to its quantity of matter.77) If on the other hand one
supposes the gravitation of the aethereal particles to be due to some
further, immaterial substance, two problems ensue. First, the sugges-
tion implies a non-unified explanation of gravitational phenomena;
the gravitation of gross bodies (i.e., detectable matter) would be
due to a material aether, but the gravitation of that aether’s material
particles would be due to some further, immaterial substance.
Second, such a suggestion renders the aether itself superfluous; if
an immaterial substance could be supposed to cause the gravitation
of the aethereal particles, it could just as well be supposed to cause
the gravitation of gross bodies directly.
Given this dilemma, let us return to the possibility of an imma-
terial medium, one that would act upon gross bodies directly.
While Newton does not propose the electric spirit for gravity’s
cause, he does speculate more generally that the cause might lie
in some sort of non-perceiving medium that is immaterial. That
possibility appears, alongside some others, in an unpublished letter
responding to Leibniz’s charges.
If any man should say that bodies attract one another by a
power whose cause is unknown to us, or by a power seated in
the frame of nature by the will of God, or by a power seated in
a substance in which bodies move and float without resistance
and which has therefore no vis inertiae but acts by other laws than
those that are mechanical: I know not why he should be said to
introduce miracles and occult qualities and fictions into the
world.78
In the emphasized passage, Newton suggests a substance that is
clearly immaterial. Whereas a material aether would have to have
some resistance in virtue of being material, even if that resistance
had thus far escaped detection, the substance that figures in
Newton’s speculations here simply does not possess the vis inertiae.
Just as perceiving immaterial substances provide no resistance to
bodies (‘bodies feel no resistance from God’s omnipresence’,
Newton writes in the General Scholium79), the non-perceiving
substance suggested in the letter provides no resistance.80 It is an
immaterial version of the aether. Because it lacks the vis inertiae, this
medium acts by laws other than the three laws of motion arising
from that inherent force. And because it is immaterial, it could fulfill
the General Scholium’s condition of penetrating to the very centers
of material particles. To the extent that Newton expects matter to
act locally and allows that the methods of empiricism may extend
to immaterial substances, the problem about gravity appears to
be one of discovering such an immaterial substance. It appears
to be a problem of finding something other than matter to bear
active powers, to fill space, and thus to underpin physically the
attractive and repulsive forces operating between spatially separated
material bodies. Yet beneath this problem lies a more fundamental
one.
6. Newton’s Substance Counting Problem
The three strains of Newton’s thought considered here are his
attraction to metaphysical principles such as a principle of local
causation; the very broad reach of his empiricism, which extends
even to immaterial substances; and his suggestion that material
and immaterial substances can occupy the same place at the same
time. This last suggestion has its metaphysical side, in that
Newton’s conception of a spatially extended God, who does share
place with matter, is grounded in the metaphysical principle of
local causation. Yet it also has its empiricist side; the phenomena
do not exclude the possibility of additional immaterial substances,
non-perceiving ones that share place with matter. These elements
of Newton’s thinking combine to block a resolution to the causal
problem about gravity, and to produce a problem about individu-
ating substances.
As we have seen, Newton attributes gravitational effects to
secondary causation; although God meets the General Scholium’s
condition for gravity’s cause, Newton treats gravity as a tool that
the deity uses, and as a force that God must even oppose. Newton
refrains from taking the secondary cause to be an active power of
matter, since he is drawn to the metaphysical principles that
matter is passive and causation is local. His inclination toward

76
Janiak (2007), p. 144; (2008), p. 79, who takes Newton to deny action at a distance, puts stock in this possibility.
77
For the 1713 edition of the Principia, Newton writes but ultimately does not publish a corollary setting out the regress argument. In all editions of the Principia, Corollary 1 to
Proposition 6 of Book III states that the weights of bodies depend only upon the quantity of matter, not upon forms or textures. In the 1687 edition, Corollary 2 then states that
there cannot be a material aether, or any other matter, that either fails to gravitate or else gravitates less in virtue of its form (and here Newton presumably means the fineness of
its particles). Newton’s reasoning here depends upon a hypothesis that he will eliminate for successive editions of the Principia (i.e., Hypothesis III, which states: ‘Every body can
be transformed into body of any other kind, and can assume successively all intermediate degrees of quality’). In accordance with the elimination of that hypothesis, Newton
revises Corollary 2 for the 1713 edition, but the content of the Corollary remains much the same. Another revision—a further corollary that he writes but ultimately does not
include—states the regress argument about the aether. Any attempt to explain gravitational effects by postulating a rare material medium that lacks gravity or has a lesser
gravitational tendency will end in a regress:

If anyone should deny these Hypotheses and have recourse to a third hypothesis, namely, that one admit some matter with no gravity by which the gravity of the
perceptible matter may be explained; it is necessary for him to assert two kinds of solid particles which cannot be transmuted into one another: the one [kind] of
denser [particles] which are heavy (have gravity) in proportion to the quantity of matter, and out of which all matter with gravity and consequently the whole
perceptible world is compounded, and the other [kind] of less dense particles which have to be the cause of the gravity of the denser ones but themselves have no
gravity, lest their gravity might have to be explained by a third kind and that (again by a fourth) and so on to infinity. (ULC MS Add. 3965.6, fol. 267r; in McGuire, 1995,
p. 264)

78
Newton to the editor of the Memoirs of Literature, unpublished, written ca. May 1712 (2004, pp. 116–117; my emphasis).
79
Newton (1999), p. 942.
80
I take this immaterial substance, like the material aether of Query 21, to be non-perceiving, since there is no indication here that Newton means to refer to any perceiving
spirit, notably God, and, as argued in Section 3, Newton seeks secondary causes for gravity. For an opposing view, see Westfall, who interprets Newton’s references to such media
ultimately as references to God (see Westfall, 1971, p. 400 for his general conclusion; p. 396 for his commentary on Newton’s remarks to Bentley: ‘If not a material aether, then an
immaterial aether—the passage to Bentley contrived to suggest it twice. What could an immaterial aether be? To Newton, it was the infinite omnipotent God, who by His infinity
constitutes absolute space and by his omnipotence is actively present throughout it’).
278 H. Kochiras / Studies in History and Philosophy of Science 40 (2009) 267–280
these principles creates pressure to find an immaterial medium.
But while he places all substances, including immaterial ones,
potentially within the reach of empirical methods, it is difficult
to see how the desired immaterial medium could in fact lie within
reach of those methods.
The problem is easily seen by conceiving an experiment
designed to determine whether air plays any causal role in
gravitational effects, and then noting why similar efforts to test
for the role of an immaterial medium would be thwarted. To deter-
mine whether air is in any way necessary to gravitational effects,
one might suspend a body within a glass container that has been
evacuated of air, and then release it. Upon observing that the body
falls in the absence of air, one concludes that air is not a catalyst for
gravitational effects. No such determination could be made about
an immaterial medium, however. According to Newton’s specula-
tions, immaterial substances might share place with material ones,
in which case it would be impossible to isolate them by means of
material bodies. One could not, for instance, be assured of having
successfully evacuated a glass container of an immaterial medium,
since for all one knows, such a substance might move freely
through the glass. If it is not possible to remove immaterial sub-
stances or to determine whether or not they are present, it is not
possible to know whether gravitational effects would occur in their
absence. The problem naturally extends to phenomena other than
gravitational effects, such as the exothermic reactions produced
when certain chemicals are combined. Newton’s speculated active
principles might belong to the material chemicals themselves, but
they might equally well belong to a non-perceiving immaterial
substance sharing the chemicals’ location, or to God; there are no
empirical grounds for making a judgment. That is to say, there
are no empirical grounds in such cases for deciding between
primary and secondary causation, or for deciding between
different hypothesized secondary causes, should one hypothesis
involve an immaterial substance.
One might contest my claim that the problem arises only when
the possibility of an immaterial substance is allowed. For if a very
rare material aether pervades the interstices between a gross
body’s parts, there is a sense in which it shares space with the
body, though its particles would not co-occupy the very same
regions of space occupied by the body’s constituent particles. Yet
with a material aether, the problem is potentially eliminable.
Because its particles, being material, must possess the vis inertiae,
this aether might be detected, perhaps through condensation; and
because its particles, being material, could not share place with
the body’s constituent particles, this material aether could poten-
tially be extracted from the body, so that its properties could be
distinguished from those of the body. This is not the case for the
speculated immaterial substance. Because it has no vis inertiae
and because it might share the very same places occupied by
the body’s material particles, there is no potential means of con-
trolling its spatial location; there is no means of isolating it or
extracting it from the body, so as to determine how properties
should be apportioned. Thus together with the other strains of
Newton’s thought, it is the admission of immaterial substances
that renders the problem ineliminable. Having allowed on
empirical grounds the possibility that substances of different kinds
might share regions of space, Newton has recourse only to
metaphysical principles as a means of apportioning active powers
to substances. Yet he refrains from asserting the principles that
81
Essay, IV.iii.6 (Locke, 1975, pp. 540–541); Locke to Stillingfleet (in Locke, 1824, Vol. 4, p. 33). On knowledge of substances in Descartes, see Nelson (1997).
82
Some of Newton’s remarks suggest that he considers dispensing with the notion of substance altogether. He dispenses with the notion of prime matter, and even suggests in
De gravitatione that one attribute, the will by which God creates, should perhaps be conceived as ‘subsisting of itself, without any substantial subject’. See Stein (2002), pp. 281–
282, for a discussion of this passage, including its implications for the Christian doctrine of the Trinity. As we have seen, however, Newton more often retains the notion of
substance.
guide his search for a medium, and accordingly refrains from
positing the medium needed to preserve those principles. Thus
he has no means of locating active powers in an immaterial sub-
stance rather than in matter.
The difficulty created by the three strains in Newton’s thought
extends beyond that of apportioning properties, most notably
active powers, to substances whose existence has already been
established. It raises the question of how to infer substances from
properties in the first place. As we saw earlier, Newton’s episte-
mological stance toward substance is dramatically different from
that of Descartes. For Descartes, substances can be known—their
essences can be known, and not through the composition of
perceived properties but through the decomposition of confused
ideas, until the simple nature is grasped. A set of perceived
properties only occasions the intellect’s work, and that work is
not to infer a substance by conjoining the perceived properties
but rather to abstract away all that can be abstracted away. At
the conclusion of this a priori process, what famously remains is
the attribute of thought alone for the mind, and of extension alone
for body. As Descartes finds these two attributes to be repugnant
to one another, there is no question of extension belonging to an
immaterial mind, or of imagination, sensation, or will belonging to
a body. Things are quite otherwise for Locke, who denies that we
know things immediately, and accordingly allows not only a
superadded power of gravitational attraction to matter, but even
the possibility that some matter might be capable of thought.81
The difficulty for Newton is that, on the question of whether sub-
stances can be known directly, he stands with Locke rather than
with Descartes. Whereas Descartes reaches knowledge of a
substance by analysis, Newton composes perceived properties into
a group, and the claim that any substance exists as their bearer is
only an inference.
Since Newton takes substances to be inferred from perceived
properties, a question about identity conditions arises.82 How
many entities should be inferred from a set of properties? The
answer would be straightforward if only material bodies existed. It
initially appears straightforward in De gravitatione, since Newton’s
attention is confined to bodies alone when he conceives of them as
perceived properties associated with determined regions of space.
For if only material bodies existed, or if only material bodies were
spatially extended, then to determine the number of entities that
should be inferred from spatially distributed properties would just
be to determine the number of bodies, using a spatial criterion in
the familiar way. Upon perceiving properties such as hardness,
impenetrability, and the light reflectivity we see as color in one re-
gion of space, we say there is one billiard ball, or one planet; and
when we perceive those sorts of characteristics in two regions of
space, we say that there are two billiard balls, or two planets. But
this method of counting bodies works only because bodies, being
impenetrable by one another, cannot share the same space, and
further because we assume that it is not possible for any other
substances to make competing claims, as it were, for the properties
in a given space by sharing that space.
The method of counting via a spatial criterion does not general-
ize from counting bodies alone to counting entities generally, then,
and this is Newton’s Substance Counting Problem. Since his ontol-
ogy includes spatially extended immaterial substances as well as
matter, and since these different sorts of extended substances
might be able to occupy the same place at the same time, it is
 H. Kochiras / Studies in History and Philosophy of Science 40 (2009) 267–280 279

not clear how many substances should be inferred from the prop-
erties perceived in a given region of space. Descartes avoids the
problem,83 by pairing the commonplace that no two bodies can be
in the same place at the same time with the claim that only bodies
are spatially extended.84 Hobbes too avoids the problem; while he
agrees with Newton that all existents are spatially extended, he con-
cludes that all those existents are material bodies, impenetrable by
one another.85 For Newton, the problem about counting substances
persists so long as the three strains of his thought persist, threaten-
ing even to undermine his concept of body.
Acknowledgements
For comments and discussion, I would like to thank Alan Nelson
and Andrew Janiak. I also benefited from some objections posed by
John Henry, Tad Schmaltz, Koen Vermeir, Friedel Weinert, and an
anonymous referee of this journal. For discussions following pre-
sentations of earlier versions of this paper, I thank the audiences
at the American Philosophical Association’s Eastern Division meet-
ing, the History of Science Society’s meeting in Pittsburgh, the
Sixth European Congress of Analytic Philosophy meeting in Kra-
kow, the Princeton–Bucharest Seminar on the Foundations of Early
Modern Thought, the meeting in Vancouver of the Canadian Soci-
ety for History and Philosophy of Science, and the Royal Institute
of Philosophy Lecture Series at the University of Bradford. I further
thank Mary Domski for her commentary at the APA; Piotr Lesniak
for his assistance in Krakow; and, for a grant facilitating my pre-
sentation in Pittsburgh, the National Science Foundation.
References
Alexander, H. G. (Ed.). (1956). The Leibniz–Clarke correspondence. Manchester:
Manchester University Press.
Boas, M. (1952). The establishment of the mechanical philosophy. Osiris, 10,
412–541.
Cohen, I. B. (1966). Newton’s conception of hypotheses. Physis, rivista internazionale
di storia della scienza, 8, 163–184.
Cohen, I. B. (1999). A guide to Newton’s Principia (with contributions by M.
Nauenberg, & G. Smith). In I. Newton, The Principia: Mathematical principles of
natural philosophy (I. B. Cohen, & A. Whitman, Trans.) (pp. 1–370). Berkeley:
University of California Press.
Descartes, R. (1985). The philosophical writings of Descartes, Vols. 1 & 2 (J.
Cottingham, R. Stoothoff, & D. Murdoch, Trans.). Cambridge: Cambridge
University Press. (Abbreviated CSM)
Descartes, R. (1991). The philosophical writings of Descartes, Vol. 3 (J. Cottingham, R.
Stoothoff, D. Murdoch, & A. Kenny, Trans.). Cambridge: Cambridge University
Press. (Abbreviated CSMK)
DiSalle, R. (2002). Newton’s philosophical analysis of space and time. In I. B. Cohen,
& G. Smith (Eds.), The Cambridge companion to Newton (pp. 33–56). Cambridge:
Cambridge University Press.
DiSalle, R. (2006). Understanding space-time. Cambridge: Cambridge University
Press.
Downing, L. (2007). Locke’s ontology. In L. Newman (Ed.), The Cambridge companion
to Locke’s Essay concerning human understanding (pp. 352–380). Cambridge:
Cambridge University Press.
Garber, D. (1992). Descartes’ metaphysical physics. Chicago: University of Chicago
Press.
Garber, D., Henry, J., Joy, L. S., & Gabbey, A. (1998). New doctrines of body and its
powers, place and space. In D. Garber, & M. Ayers (Eds.), The Cambridge history of
seventeenth-century philosophy (2 vols.) (Vol. 1, pp. 553–623). Cambridge:
Cambridge University Press.
Hale, M. (1667). Observations touching the principles of natural motions; and especially
touching rarefactions and condensations: Together with a reply to certain remarks
touching the gravitation of fluids. London: Printed by W. Godbid for W.
Shrewsbury.
Hawes, J. (1968a). Newton and the ‘electrical attraction unexcited’. Annals of Science,
24(2), 121–130.
Hawes, J. (1968b). Newton’s revival of the aether hypothesis and the explanation of
gravitational attraction. Notes and Records of the Royal Society of London, 23(2),
200–212.
Hawes, J. (1971). Newton’s two electricities. Annals of Science, 27(1), 95–103.
Henry, J. (1986). Occult qualities and the experimental philosophy: Active
principles in pre-Newtonian matter theory. History of Science, 24, 335–381.
Henry, J. (1994). ‘Pray do not ascribe that notion to me’: God and Newton’s gravity.
In J. Force, & R. Popkin (Eds.), The books of nature and scripture: Recent essays on
natural philosophy, theology and biblical criticism in the Netherlands of Spinoza’s
time and the British Isles of Newton’s time (pp. 123–147). Dordrecht: Kluwer.
Hesse, M. (1955). Action at a distance in classical physics. Isis, 46(4), 337–353.
Hobbes, T. (1951). Leviathan (C. B. Macpherson, Ed.). New York: Penguin.
Hooke, R. (1665). Micrographia: Or some physiological descriptions of minute bodies
made by magnifying glasses, with observations and inquiries thereupon. London:
Printed by J. Martyn and J. Allestry.
Janiak, A. (2007). Newton and the reality of force. Journal of the History of Philosophy,
45, 127–147.
Janiak, A. (2008). Newton as philosopher. Cambridge: Cambridge University Press.
Locke, J. (1975). An essay concerning human understanding (P. Nidditch, Ed.). New
York: Oxford University Press.
Locke, J. (1824). The works of John Locke (12th ed.) (9 vols.). London: C. & J. Rivington.
McGuire, J. E. (1968). Force, active principles, and Newton’s invisible realm. Ambix,
15, 154–208.
McGuire, J. E. (1995). Transmutation and immutability: Newton’s doctrine of
physical qualities. In idem, Tradition and innovation: Newton’s metaphysics of
nature (pp. 262–286). Dordrecht: Kluwer Academic.
McMullin, E. (1978). Newton on matter and activity. Notre Dame, IN: University of
Notre Dame Press.
Nelson, A. (1997). Descartes’s ontology of thought. Topoi, 16, 163–178.
Newman, L. (2001). Unmasking Descartes’ case for the Bête machine doctrine.
Canadian Journal of Philosophy, 31, 389–426.
Newton, I. (1952). Opticks, or A treatise of the reflections, refractions, inflections and
colors of light. Based on the fourth edition of 1730. New York: Dover.
Newton, I. (1959–1971). Correspondence of Isaac Newton (H. W. Turnbull, J. F. Scott,
A. R. Hall, & L. Tilling, Eds.) (7 vols.). Cambridge: Cambridge University Press.
Newton, I. (1962). Unpublished scientific writings of Isaac Newton (A. R. Hall, & M. B.
Hall, Eds. & Trans.). Cambridge: Cambridge University Press.
Newton, I. (1978). Isaac Newton’s papers and letters on natural philosophy (I. B. Cohen,
Ed.) (2nd ed.). Cambridge, MA: Harvard University Press.
Newton, I. (1999). The Principia: Mathematical principles of natural philosophy (I. B.
Cohen, & A. Whitman, Trans.). Berkeley: University of California Press.
Newton, I. (2004). Newton: Philosophical writings (A. Janiak, Ed.). Cambridge:
Cambridge University Press.
Schliesser, E. (2009). Without God: Newton’s relational theory of attraction. In D.
Jalobeanu, & P. Anstey (Eds.), Vanishing matter and the laws of motion from
Descartes to Hume. (Under review)

83
A referee of this journal raises the following question:

Could even a Cartesian be saddled with some version of the Substance Counting Problem? How can we tell, just by looking at the behavior of a young child and that of an
adult chimpanzee, that the former’s behavior is the result of two distinct types of substance working in tandem, but the latter’s is not?

While the question as posed may suggest that the threat of a substance counting problem arises from Descartes’s bête machine doctrine, it actually arises from the problem of
other minds. Descartes can avoid any problem about counting substances, then, if he has a convincing answer to the problem of other minds. Taking spoken words to be signs of
thought (CSMK, p. 303), he attributes the thoughts expressed by those words to a mind while attributing the spatially extended face from which the words issue to a body. If we
grant Descartes his bête machine doctrine, then, he has analogical grounds for attributing a pre-linguistic child’s behavior to two substances while attributing the chimpanzee’s
behavior to one: only children develop from non-linguistic beings into linguistic ones. (And if we deny the bête machine doctrine, arguing instead that the chimpanzee manifests
some signs of thought, then we have returned to the problem of other minds.) For an analysis of Descartes’s bête machine doctrine, see Newman (2001).
84
Descartes takes impenetrability to be implied by the essence of extension. Responding to More’s claim that they are separable, Descartes writes:

We cannot even understand one part of an extended thing penetrating another part equal to it without understanding by that very fact half of that extension eliminated or
annihilated. But what is annihilated does not penetrate another thing. And thus, in my judgment, it is demonstrated that impenetrability pertains to the essence of extension.
(AT V 342, in Garber, 1992, p. 147; his translation)
See the discussion there and in Garber et al. (1998), p. 579.
85
See Hobbes (1951), pp. 428–429, 689–690.
280 H. Kochiras / Studies in History and Philosophy of Science 40 (2009) 267–280

Stein, H. (2002). Newton’s Metaphysics. In I. B. Cohen, & G. Smith (Eds.), The
Cambridge companion to Newton (pp. 256–307). Cambridge: Cambridge
University Press.
Stuart, M. (1998). Locke on superaddition and mechanism. British Journal for the
History of Philosophy, 6(3), 351–379.
Westfall, R. S. (1962). The foundations of Newton’s philosophy of nature. British
Journal for the History of Science, 1(2), 171–182.
Westfall, R. S. (1971). Force in Newton’s physics: The science of dynamics in the
seventeenth century. London: Macdonald & Co.; New York: American Elsevier
Publishing Company.
Westfall, R. S. (1980). Never at rest: A biography of Isaac Newton. New York:
Cambridge University Press.
Wiggins, D. (1968). On being in the same place at the same time. The Philosophical
Review, 77(1), 90–95.