Fundamental Reality

Jonathan Schaffer and L. A. Paul

I—JONATHAN SCHAFFER

THE ACTION OF THE WHOLE

I discuss an argument for the monistic idea that the cosmos is the one and
only fundamental thing, drawing on the idea that the cosmos is the one
and only thing that evolves by the fundamental laws.

It is possible to say a great deal about the world as a whole. We can

point to global structuring principles, universal processes of world ev-
olution, general symmetries, a common ontological basis of reality, a
single origin of the universe, and various universally conserved quanti-
ties … The world [has] a highly integrated and coherent structure.
——Ellis 2001, p. 249

What are the fundamental blocks from which reality is built? The
pluralist views the cosmos as pieced together from many tiny parts,
while the monist holds that the entire scene is painted onto one vast
unbroken whole. I provide an argument for monism—the argument
from nomic integrity—which draws on the idea that the fundamen-
tal laws of nature govern the temporal evolution of the cosmos as a
whole, applying at most approximately and derivatively to any
merely partial subsystem:
(1) Leibnizian Substance: Something is a substance if and only
if it evolves by the fundamental laws.
(2) Russellian Laws: The cosmos is the one and only thing that
evolves by the fundamental laws.
(3) Spinozan Monism: The cosmos is the one and only sub-
stance (from 1 and 2).
To fix another image: the cosmos ticks like a single clockwork. To
cast a slogan: reality acts as one.
Plan: In §§i–iii I discuss claims (1)–(3) in turn. In §iv I consider
a variant argument using weakened counterparts of (1) and (2), but
adding an Aristotelian principle about the mereology of substances.

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
68 I — JONATHAN SCHAFFER

Leibnizian Substance—premiss (1) of the argument from nomic

integrity—connects the notions of substance and lawhood. By ‘sub-
stance’ I mean a fundamental and integrated thing, a block from
which reality is built, a tile of the cosmic mosaic:
Substance Something is a substance if and only if it is a funda-
mental and integrated thing.1
A thing is fundamental if and only if it depends on nothing further,
and a thing is integrated if and only if it is not an arbitrary gerry-
mander but displays natural unity. Obviously there is much more to
be said about these notions, but I trust they are clear enough to
work with.
Leibnizian Substance connects this notion of substance—that of
a fundamental and integrated thing—to the notion of a fundamen-
tal law. By ‘fundamental law’ I mean something familiar from phys-
ics: Newton’s F = ma and Schrödinger’s equation are historical
candidates. I trust that this is also a clear enough notion to work
with.
The crucial component of Leibnizian Substance is the way it con-
nects substance to lawhood. Consider a candidate fundamental law
such as F = ma. This law does not say what systems it concerns, but
only how certain properties evolve through time. Indeed, one can
apply this equation to arbitrary systems—collections of particles or
billiard balls or rocket ships—with varying degrees of accuracy. So
how can one tell, given such a fundamental law, which things are
fundamental? How can one tell how the cosmic mosaic is tiled?
Leibnizian Substance provides an answer, which is essentially that
something is fundamental if and only if this fundamental equation
applies to it with full accuracy. More precisely, the answer is that a
substance must evolve by the fundamental laws, by which I mean
that a substance must be such that plugging its state at any given
time into the fundamental laws correctly predicts its actual behav-
1
This conception picks up on certain aspects of Aristotle’s conception of substance. For
Aristotle, substances are fundamental, and without them ‘it would be impossible for any of
the other things to exist’ (1984a, p. 5; cf. Spinoza 1994, p. 85). Aristotle likewise views sub-
stances as integrated, being ‘that which is compounded out of something so that the whole
is one—not like a heap, but like a syllable …’ (1984c, p. 1644). Though Aristotle’s concep-
tion also incorporates connections to further notions such as predication, on which the
present conception stays neutral.

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
 FUNDAMENTAL REALITY 69

iour. The tile of the mosaic are the loci of the dynamics.2
It is probably impossible to say anything more precise about what
it takes to evolve by the fundamental laws without substantive as-
sumptions about time and laws. For simplicity and definiteness, I
begin from a picture on which reality is built from fundamental con-
crete things enduring through time, with fundamental laws describ-
ing deterministic evolutions. The idea of evolving by the
fundamental laws can then be unpacked, beginning from the notion
of the state of a thing at a time:
State The state of a thing at a time is its fundamental intrinsic
character at the time.3
For instance, a given particle might have a certain mass, charge and
velocity at a given time. The state needs to be fundamental, since it
is going to serve as an input to the fundamental laws, and needs to
be intrinsic (or at least neighbourhood local; Butterfield 2006), or
else the connection would be trivialized.
From the notion of the state of a thing at a time, one can charac-
terize the prediction for that state, which takes in the state and the
fundamental laws as input, and outputs a temporal evolution,
which is a trajectory through the space of states across times:
Prediction The prediction for a state is the temporal evolution
that the fundamental laws output on the basis of
the state.
To illustrate, for a Newtonian one-body system at rest at t0, the pre-
diction is inertial ‘eternal rest’:
2
This idea is inspired by Leibniz, who—in his mature metaphysics—connects the notion of
substance to the idea of an internal lawful dynamics. Thus Leibniz opens his Principles of
Nature and Grace by defining a substance as ‘a being which is capable of action’ (1998b,
p. 258), and explains to Arnauld, ‘Each of these substances contains in its nature a law of
the continuation of the series of its own operations …’ (1989, p. 360).
3
In a quantum setting, at least under some interpretations, one might not even be able to
assign (pure) quantum states to things short of the cosmos, given pervasive entanglements.
Thus Everett—denying that the state of a local subsystem has any ‘independent existence’
(1957, p. 455)—maintains, ‘Subsystems do not possess states that are independent of the
states of the remainder of the system … It is meaningless to ask the absolute state of a
subsystem—one can only ask the state relative to a given state of the remainder of the sys-
tem’ (1957, p. 143). Perhaps this provides a direct ‘static’ argument for monism in an entan-
gled quantum universe (Schaffer 2010a, §2.1), without even considering the dynamics. But
the argument in the main text is intended to be independent of any such considerations, and
independent of any interpretative controversies over quantum mechanics. So in the main
text I allow the pluralist that one can at least assign states to the local subsystems she would
consider fundamental.

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
70 I — JONATHAN SCHAFFER

t +∞

Prediction for a Newtonian

t0 one-body system at rest at t0 :
eternal rest.

t −∞

Finally, one can compare the prediction for the thing at the time to
its behaviour, which is how it is over time, its actual temporal evolu-
tion:
Behaviour The behaviour of a thing is the temporal evolution
it actually displays.
The relevant comparison is just whether the thing always behaves as
predicted. Think of Prediction as a physics homework problem, and
Behaviour as a lab observation checking the answer:
Evolving Something evolves by the fundamental laws if and
only if, for any given time, its prediction at that
time matches its behaviour.
After all, it is a minimal condition on something evolving by certain
laws, that those laws hold for it.4 A Newtonian one-body cosmos al-
ways behaves as predicted, and so the world/particle in that scenar-
io satisfies Evolving:

t +∞

Behaviour for a Newtonian

t0 one-body system at rest at t0 :
eternal rest (as predicted).

t −∞

4
Match between prediction and behaviour is a minimal condition. In some scenarios
(where there are multiple levels of match) one might want more, such as the requirement
that the laws generate the behaviour. I return to this in §iv.

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
 FUNDAMENTAL REALITY 71

Evolving is fitted to an endurantist-deterministic setting, but the

idea readily generalizes. First, one can generalize away from evolu-
tions in time to evolutions in space-time (as called for in a relativis-
tic setting), simply by requiring that the thing always behaves as
predicted on any foliation.5 Indeed one can generalize to evolution
in any ‘arena’ (Paul 2012): there is no requirement on the number of
dimensions, the presence of metrical structure, etc. Secondly, one
can work with perdurance instead of endurance, by replacing talk
of the thing with talk of a sequence of stages united by temporal
counterpart relations. The relevant claim would then be that cos-
mic-width stages are the one and only stages linked by the funda-
mental laws.6 Thirdly, one can work with indeterminism (as called
for under some interpretations of a quantum setting), by including
propensities as an aspect of prediction and behaviour. The compari-
son of prediction to behaviour would then include a comparison of
predicted to displayed propensities.7 I leave open just how far
Evolving can be generalized, and limit discussion to the endurantist-
determinist setting.
So much for what Leibnizian Substance says. I consider it a plau-
sible thing to say, for three main reasons. First, Leibnizian Sub-
stance draws on the idea of substances as fundamental, plausibly
connecting fundamental things to fundamental laws (and thereby
providing an empirical handle on the substances). Just as it is widely
thought that the fundamental laws govern the fundamental proper-
ties (Lewis 1983, pp. 367–8), so it seems equally plausible to think
that the fundamental laws govern the distribution of fundamental
properties over fundamental things. More precisely: it is standard to
regiment laws via the schema ‘∀x(Fx t Gx)’. The fundamental
5
More precisely, in a relativistic setting, the state of a thing is to be understood as its fun-
damental intrinsic character along some time-like hypersurface. Prediction is then relativ-
ized to a full foliation of the space-time into time-like hypersurfaces. Given the invariance of
the dynamics under different foliations, prediction at all times under any arbitrary foliation
may be compared with behaviour within the manifold.
6
Complication: In a perdurantist setting, there would remain cross-temporal questions
about whether the stages are more or less fundamental than their fusions (‘worms’). The
argument from nomic integrity falls silent on this question, so given perdurantism it would
leave open whether there are many vast fundamental things (each cosmic stage) or one (the
whole serpent).
7
If one opts for a more reductive view of chances, then one must compare the actual behav-
iour to a predicted probability distribution over behaviours. This difficulty is compounded
if the probability distribution allots some probability to all possible behaviours, and com-
pounded further if the distribution allots the same measure zero probability to all possible
behaviours. I do not know how to generalize Evolving to such a setting.

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
72 I — JONATHAN SCHAFFER

things should then be the things quantified over in the fundamental

laws. Or perhaps better, one should think of laws as regimented by
differential equations describing ‘how the physical state of a system
or particle evolves through time’ (Maudlin 2007, p. 11). The funda-
mental things should then be the things with fundamental evolu-
tions.
Secondly, Leibnizian Substance draws on the idea of substances
as integrated, and can be understood as providing a nomic test for
natural unity. To evolve by the fundamental laws is to act in an inte-
grated way, forming an internally comprehensible and self-con-
tained system (this is what makes something count as a tile in the
mosaic). The natural unity of a thing is displayed in its dynamics. In
a slogan: to be one is to act as one.8
Thirdly, Leibnizian Substance provides a plausible explanation
for why certain conceptions of substance seem inadequate, as seen
in Leibniz’s critique of Malebranche’s occasionalism. Malebranche
claims that created minds are substances, and also that created
minds lack an internal capacity for action but act only via God.9
Leibniz sees a tension: how can created minds be substances if they
require the constant external support of a deity?10 Leibnizian Sub-
stance earns credence by providing a plausible explanation for the
8
These first two points about fundamentality and integration reveal how Leibniz’s concep-
tion of substance represents a culmination of the classical Aristotelian conception, as
Rutherford explains:
Rather than simply fixing what is to be called ‘substance’, [Leibniz] is working from a
complex, historically rooted conception of what it is to be a substantial being, and is
subsequently arguing that these conditions can only be satisfied if substance is con-
ceived as being by nature a principle of activity. We may see him as claiming that sub-
stance could not fulfil its prescribed metaphysical roles—as an ultimate explanatory
principle, as a being that is dependent for its existence on no other created being, as a
being that persists through change, and as a true unity—unless it is also conceived as a
principle of activity: a source productive of changes in its states or modifications, which
nevertheless persists as itself through those changes. (Rutherford 1993, p. 138)
9
For Malebranche, it is not merely created minds that lack internal power, but bodies too,
and indeed everything beside God: ‘[N]o power can convey it to where God does not convey
it, nor fix nor stop it where God does not stop it, unless God accommodates the efficacy of
His action to the inefficacious action of His creatures’ (Malebranche 1997, pp. 115–16).
10
Indeed, in Nature Itself, Leibniz warns that occasionalism leads to the Spinozan heresy:
[T]he doctrine of occasional causes which some defend can lead to dangerous conse-
quences … [T]his doctrine seems, with Spinoza, to make God into the very nature itself
of things, and to reduce created things to mere modifications of a single divine sub-
stance. For that which does not act, which has no active force, which is robbed of any
distinguishing characteristic, and finally of all reason and ground of permanence, can in
no way be a substance. (Leibniz 1998a, p. 221)

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
 FUNDAMENTAL REALITY 73

inadequacy of Malebranchian occasionalism.

Of course, one might reject Leibnizian Substance for various rea-
sons,11 but perhaps the most interesting response vis-à-vis the debate
over monism is to allow for substances that are merely among a plu-
rality that co-evolve by the fundamental laws:
Leibnizian Substance, Plural Some things are substances if and
only if they co-evolve by the fundamental laws.
Leibnizian Substance, Plural can claim to preserve a link between
fundamental things and fundamental laws, while allowing for a plu-
rality of things (such as particles) to be substances, so long as they
co-evolve as predicted. This may be the pluralist’s most natural
move.
But Leibnizian Substance, Plural does not fit the idea of substanc-
es as integrated. From the fact that some things co-evolve by the
fundamental laws, one can only infer that this plurality acts as one.
One cannot infer that any individual in this plurality has the natural
unity of a substance (a tile in the mosaic). Indeed, it seems to me
that a plausible way to decide whether a given composite system is
most fundamentally a single composite substance or a mere heap of
various elements is to see whether the plurality acts as one. So if
some things co-evolve by the fundamental laws, it seems to me that
one ought to conclude, not that each of these things is an individual
substance, but rather that their system is a composite substance, for
that is what forms a natural unit with respect to the dynamics.
Relatedly, without any test for individual integrity, Leibnizian
Substance, Plural would count arbitrary heaps as substances, as
long as they collectively compose the cosmos. Thus consider the plu-
rality whose two members are the sum of all left feet, and its mereo-
logical complement (the cosmos minus the sum of all left feet). They
collectively evolve by the fundamental laws just because they to-
gether compose the cosmos. But obviously they are each mere heaps
without natural unity.
And relatedly, Leibnizian Substance, Plural would ruin Leibniz’s
plausible critique of Malebranche. After all, Malebranche does
think that the plurality whose members are the created minds and
11
It may be worth noting that Leibnizian Substance is separable from Leibniz’s theism with
its pre-established harmony, and separable from Leibniz’s monadology with its pluralist ide-
alism. (Though the fact that Leibniz combined Leibnizian Substance with pluralism helps
show that Leibnizian Substance does not presuppose monism.)

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
74 I — JONATHAN SCHAFFER

the deity co-evolves by the fundamental laws (the volitions of the

created minds help inform the actions of the deity). What is missing
is the means to see that the minds themselves lack any internal ca-
pacity for action. To see that, one must consider each mind individ-
ually and see how it can act.

II

Whereas Leibnizian Substance connects the notions of substance

and lawhood, Russellian Laws—premiss (2) of the argument from
nomic integrity—makes an empirical claim about laws, bringing in
the cosmos. By ‘the cosmos’ I mean the whole material universe, the
total system, the sum of all concrete things:
Cosmos The cosmos is the whole material universe.
The existence of such a thing claims intuitive and empirical support.
Intuitively, the cosmos is hardly a strange fusion undreamt of by the
folk, but is an entity for which natural language supplies a singular
term.12 Empirically, the cosmos is an entity posited in physics, and
indeed the subject of cosmology, which Hawley and Holcombe
characterize as ‘the study of the formation, structure, and evolution
of the universe as a whole’ (2005, p. 5). Only the most radical views
of mereological composition, contravening both intuition and sci-
ence, could refuse the cosmos.
(I continue to work with an ‘endurantist’ picture on which things
can exhibit numerical identity across time and through change. So I
am treating the cosmos as a single enduring thing exhibiting numer-
ical identity across each moment of time and through all changes.)
Russellian Laws claims that the cosmos has a unique relation to
the fundamental laws, being the one and only thing that evolves by
them. Plugging in Evolving, the claim is that the cosmos is the one
and only thing that always behaves as predicted. As Maudlin writes,
‘[T]he fundamental laws of nature appear to be laws of temporal ev-
olution: they specify how the state of the universe will, or might,
evolve from a given initial state’ (2007, p. 172). The fundamental
laws govern the temporal evolution of the whole system, and apply

12
‘Cosmos’ derives from the Greek ‘κόσμος’ for order, and served as the title of a 1980s
public television series featuring Carl Sagan.

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
 FUNDAMENTAL REALITY 75

at most approximately and derivatively to any merely partial sub-

system.13
The content of Russellian Laws can be further elucidated by con-
sidering ‘monadistic’ scenarios with local subsystems—windowless
bubbles—that behave as predicted. In such scenarios, the temporal
evolution of the world state may be understood piecemeal, via the
independent evolutions of its monads. The core content of Russelli-
an Laws is that our world is not like this. One cannot correctly
specify independent evolutions of distinct subsystems first, and then
patch together the dynamics of the whole. We can only specify evo-
lutions in the context of the whole system. The evolutions of subsys-
tems are thus to be understood as derivative abstractions from the
fundamental evolution of the whole system.
So much for what Russellian Laws says. I consider it a plausible
thing to say, for three main reasons. First, local subsystems are al-
ways liable to outside disruption. For instance, if someone strikes a
match, whether and how the match will light depends on external
factors like the wind. The cosmos—being the one system for which
nothing is ‘outside’—is the one system immune to disruption. This
means that the cosmos is the one and only thing that evolves by
strict laws. Local subsystems at best have a ceteris paribus (more
precisely, ceteris absentibus) evolution. Given that the fundamental
laws are strict rather than ceteris paribus, Russellian Laws follows.

13
This idea is inspired by Russell, who—rejecting the crude ‘philosopher’s view’ that there
are fundamental laws linking one little event to another (e.g. linking the striking of the
match to its lighting)—writes:
The case where one event A is said to ‘cause’ another event B, which philosophers take as
fundamental, is really only the most simplified instance of a practically isolated system. It
may happen that, as a result of general scientific laws, whenever A occurs throughout a
certain period, it is followed by B; in that case, A and B form a system which is practically
isolated throughout that period. It is, however, to be regarded as a piece of good fortune if
this occurs; it will always be due to special circumstances, and would not have been true if
the rest of the universe had been different … (Russell 1992, p. 204)
And so Russell concludes that the fundamental laws—being exceptionless principles which
do not merely hold ‘as a piece of good fortune’—cannot link local states. The ‘rest of the
universe’ must be included. And, moreover, these fundamental laws will not divide global
states into ‘A’ versus ‘the rest,’ they will simply characterize the temporal evolution of the
system as a whole. As Loewer explains:
Russell (1913) observed that the fundamental laws—he was thinking of the differential
equations of classical mechanics, but the same holds for quantum mechanics—specify
how the whole state of an isolated system evolves (or the chances of possible evolutions)
but don’t specify which parts of the state at one time are causally connected to which
parts of the state at other times. (Loewer 2007, p. 296; cf. Maudlin 2007, p. 178)

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
76 I — JONATHAN SCHAFFER

A second reason for finding Russellian Laws plausible is that lo-

cal subsystems are in fact constantly disrupted. There are no closed
systems in nature short of the whole cosmos. So any lawful predic-
tion of the behaviour of any local subsystem is not just potentially
falsifiable, but in fact false.14 Open systems generally deviate from
their internally predicted course as they are buffeted about by exter-
nal forces (save for the miraculous circumstance in which the im-
pinging forces perfectly counterbalance). For instance, in a
Newtonian setting—ignoring mathematical curiosities like objects
entering asymptotically ‘from infinity’ (Earman 1986)—the state of
the cosmos at any given time determines its state at all other times.
But this is not true for any local subsystem, due to external forces,
including the instantaneous gravitational ‘tug’ of every massive
body throughout the cosmos.15
To see this point in action, again consider a Newtonian two-parti-
cle system, with an initial state t0 at which both particles are at rest.
First consider each particle individually at t0. Since each particle is
at rest, the lawfully predicted behaviour for any one particle, if only
considered alone, is just what is predicted for a one-body system,
namely, eternal rest:

t+∞
Behaviour for each particle in
a two-body Newtonian system
at rest at t0 : eternal rest.

t0

Such a prediction obviously will not match the actual behaviour of

the system. Assuming that both particles have non-zero mass, and
considering only gravity, there will be mutual attraction:

14
In a partially related vein, Cartwright (1983) argues that the laws of physics ‘lie,’ and are
at best idealizations. Russellian Laws can be understood as the claim that Cartwright is
almost completely right. There is just one system, namely the cosmos, about which the laws
speak truly.
15
In a relativistic setting these influence are limited to hypersurfaces of back light-cones,
but—even in inflationary ‘multiverse’ models—there are still no completely closed subsys-
tems in nature.

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
 FUNDAMENTAL REALITY 77

t+∞
Behaviour for each particle in
a two-body Newtonian system
at rest at t0 : mutual attraction.

t0

It is only the lawfully predicted behaviour for the whole system—in

this case the two-body system—which matches actual behaviour.
Only the whole system acts as one.
A third reason supporting Russellian Laws comes from conserva-
tion laws, which only apply to the whole. No subsystem need be
conservative so long as the remainder ‘compensates’. (Alternatively,
one might think of conservation laws as only applying to closed sys-
tems; then the point is again that there are no closed systems in
nature short of the whole cosmos.) Indeed, by Noether’s theorem,
conservation laws can be understood as space-time symmetries,
where a symmetry is an invariance under certain global transforma-
tions. Such laws are global by construction. In this vein, Bigelow, El-
lis and Lierse conclude:

[T]he most fundamental laws of nature, e.g. the conservation laws, the
principles of relativity, and the symmetry principles, … neither ascribe
properties to things within the world, nor describe correlations be-
tween things in the world. It is natural to construe them, rather, as
characterizing … the world as a whole … (Bigelow, Ellis and Lierse
1992, p. 384)

Of course one might challenge Russellian Laws for various reasons.

For instance, one might maintain that the world in fact contains all
sorts of practically isolated subsystems, and that this suffices for
Evolving (or some emended version thereof). It is true that the world
in fact contains all sorts of practically isolated systems, but false that
this suffices for Evolving (or any plausible emendation thereof).
Practically isolated systems remain systems for which prediction does
not match behaviour. Thus they remain systems that do not exhibit
the right connections with the fundamental laws to be fundamental,
and do not act in a sufficiently unified way to be integrated. What
does it matter how many decimal points deep the mismatch lies?

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
78 I — JONATHAN SCHAFFER

What is relevant is whether there is a mismatch, not its magnitude.16

Secondly, one might challenge Russellian Laws on grounds that
local subsystems do possess an internal capacity for action. For in-
stance, one might think that local subsystems—unlike Malebranchi-
an minds—at least have a counterfactual capacity for action, since
they would evolve by the fundamental laws if alone. But substances
need to be actual unities, and the fact that something might poten-
tially act as one if alone does not entail that it actually acts as one
when accompanied. Moreover, Malebranchian minds may well pos-
sess such a counterfactual capacity for action. For the Malebranchi-
an, individual minds without a deity are impossible, and so the
associated counterfactuals may be vacuously true. So Leibniz’s plau-
sible critique of Malebranche would fall.
Or instead, one might think that local subsystems—unlike occa-
sionalist minds—at least generate component forces which match
components of the behaviour exhibited (they merely fail to match the
overall resultant behaviour due to further component forces). But—
granting arguendo the existence of component forces—there is clear-
ly no matching component behaviour. As Cartwright aptly notes,
‘When a body has moved along a path due north-east, it has travelled
neither due north nor due east’ (1983, p. 60). Moreover, occasionalist
minds are essential (though non-vectorial) ‘components’ of a mind-
deity system in which the mind freely chooses a course of action, and
only then is the deity guided to impart a behaviour to the body. So
again, Leibniz’s plausible critique of Malebranche would fall.
What emerges is that local subsystems are like Malebranchian
minds in a relevant sense, namely that both lack an internal lawful
evolution, instead requiring external support. With Malebranchian
minds, the external support required is the grace of the deity. In the
case of local subsystems, the external support required is the context
of the cosmos. Thus neither sort of thing exhibits the right connec-
tion to the fundamental laws to be fundamental. And neither sort of
thing can count as integrated, for neither sort of thing acts as one.

16
Practically isolated local subsystems are crucial to the epistemology of science, allowing
global laws to be subjected to approximate local tests. Thus Chew explains the possibility of
scientific knowledge: ‘All phenomena ultimately are interconnected, so an attempt to under-
stand only a part necessarily leads to some error, but the error is often sufficiently small for
the partial approach to be meaningful’ (1968, p. 763).

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
 FUNDAMENTAL REALITY 79

III

From Leibnizian Substance and Russellian Laws, Spinozan

Monism—the conclusion of the argument from nomic integrity—
follows. (The historically informed reader might note the irony of
combining ideas from Leibniz and Russell—two of the greatest op-
ponents of monism—to reach the Spinozan view.) The cosmos ticks
like a single clockwork. It displays the connection with the funda-
mental laws that befits a fundamental thing, and an internal dynam-
ics that befits an integrated unity. Nothing short of the cosmos—no
mere cog of the whole clockwork, no arbitrary fragment of the con-
tinuous mosaic—displays this connection with the fundamental
laws, or such an internal dynamics.17
The inference from global structure to the unity of the whole is
indeed drawn explicitly by Ellis, albeit without connection to the
classical monism debate. Thus Ellis speaks of the world as a ‘a high-
ly integrated and coherent structure’ (2001, p. 249), and adds:
[T]he fact that these global properties and this structure exist at all im-
plies that the world is a unified whole. If the world consisted of unre-
lated kinds of things that were just thrown together somehow, these
properties and this structure would be inexplicable. (Ellis 2001, p. 251)
The argument from nomic integrity merely connects this orthodox
claim in the philosophy of physics, to a classical but neglected view-
point in metaphysics.
Of course, one might take the conclusion of Spinozan Monism to
suggest a reductio of one of the premisses of the argument. I would
only ask why Spinozan Monism should be rejected (especially if it

17
This idea is inspired by Spinoza, for whom ‘in Nature there is only one substance …’
(1994, pp. 93–4), and who invokes something like nomic integrity in a letter to Oldenburg:
[A]ll bodies are surrounded by others, and are determined by one another to existing
and producing an effect in a certain and determinate way, the same ratio of motion to
rest always being preserved in all of them at once, that is, in the whole universe. From
this it follows that every body, in so far as it exists modified in a certain way, must be
considered as a part of the whole universe, must agree with the whole to which it be-
longs, and must cohere with the remaining bodies. (Spinoza 1994, pp. 83–4)
Indeed (as I have learned from Katja Vogt) this sort of argument may be traced back to the
Stoics. As Alexander reports:
[The Stoics] say that since the world is a unity which includes all existing things in itself
and is governed by a living, rational, intelligent nature, the government of existing
things which it possesses is an everlasting one proceeding in sequence and ordering …
In this way all things are bound together. (Long and Sedley 1987, p. 337)

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
80 I — JONATHAN SCHAFFER

follows from plausible premisses). Monism may have been scorned

in the twentieth century, but for the preceding millennia it was con-
sidered a profound idea, worthy of the allegiance of many of our
greatest philosophers. Perhaps the monistic vision was scorned for
good reason. I only ask to hear it.
The closest I can find to an argument against Spinozan Monism
comes from the idea that it conflicts with Moore’s platitudes such as
‘Here is one hand, and here is another’ (Moore 1993, p. 166), and
with Russell’s ‘common-sense belief that there are many separate
things’ (Russell 1985, p. 36). But there is no conflict. Spinozan Mon-
ism does not say that exactly one entity exists, and it does not even
say that exactly one thing exists. It only says that exactly one
substance—one fundamental and integrated thing—exists. It is thus
compatible with the existence of any number of dependent things,
which is surely what Moore’s hands are, and what Russell’s common-
sense belief concerns. To speak of the cosmos as a single clockwork is
not to deny that it has parts, but rather to characterize the parts as de-
pendent cogs. To say that the whole is prior to its parts—as per the
classic monistic dictum—is to presuppose that the whole has parts,
and to assert a claim about the dependencies between these things.18
Of course, if one operates with a background schedule of catego-
ries on which all entities are either substances or modes, then Spino-
zan Monism would entail that Moore’s hands can at most be
modes.19 But the problem here is not Spinozan Monism, but the
background schedule of categories, which leaves no space for de-
pendent things like hands. The pluralist—were she to operate with
this schedule—would face the same problem. For such a pluralist,

18
Thus Proclus writes, ‘[S]ince the monad is everywhere prior to the plurality, all beings
must be attached to their particular monads. In the case of bodies, the whole that precedes
the parts is the whole that embraces all separate beings in the cosmos’ (Proclus 1979, p. 79).
Likewise Joachim speaks of the part as ‘a fragment torn from its context, in which alone it
has its being and significance’, explaining that the part ‘owes its nature to its place in the
whole system of bodies which together constitute the corporal universe’ (Joachim 1901,
p. 23). See Schaffer (2010a, 2010b) for further discussion of this view, labelled priority
monism. Spinozan Monism is also compatible with the more radical existence monism,
which denies existence to anything other than the whole (Horgan and Potr 2008). My point
is just that Spinozan Monism is compatible with priority monism.
19
Spinoza himself works with the categories substance and mode, though Spinoza’s official
definition of a mode is simply that of ‘the affections of the substance, or that which is in
another through which it is also conceived’ (1994, p. 85), which might be thought to cover
dependent things. If so, then my complaint is rather that his mode conflates dependent
property-bearers like hands with the properties they bear, when a categorical distinction is
called for.

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
 FUNDAMENTAL REALITY 81

Moore’s hands would equally count as neither substances (not fun-

damental) nor modes (not properties). A decent schedule of catego-
ries needs to leave room for dependent things which are neither
substances nor modes, and any schedule that achieves this will al-
low the friend of Spinozan Monism to hold on to Moore’s hands.20
Indeed I think that a decent schedule of categories should forgo
substance altogether. I take it that entities fall under their categories
essentially (or perhaps better: no contexts permit cross-categorical
counterparts). But fundamentality seems inessential. The monist can
grant that her cosmic substance may be embedded in a larger whole,
and then it would no longer be a substance. Likewise, the pluralist
who treats, say, a given electron as a substance can grant that it may
be divisible into smaller constituents, and then it would no longer
(by her lights) be a substance. And all sides can agree that a given
mind is a dependent entity at a physicalist world, but that it may be
a substance at a dualist world. So, given that things are substances
only contingently, substance cannot be a category.
There is a natural amendment to hand, which is to replace sub-
stance with thing, where a thing is a bearer rather than a property,
surface rather than paint. A substance is a thing that just happens to
have some special features, namely fundamentality and integration.
Replacing substance with thing makes room for dependent things,
and fits the observation that fundamentality seems contingent. The
categorical divide that nothing can cross is between bearer and
property, not between fundamental and dependent.21
But—whatever schedule of categories one ultimately prefers— the
underlying point is that all sides need to allow for dependent things, if
they would sustain Moorean platitudes. Moore’s hands offer no sup-
port for pluralism. If there is anything wrong with Spinozan Monism,

20
What is the category of Moore’s hands on an Aristotelian schedule? For Aristotle, a hand
is not a substance, since it depends on its host organism (indeed a hand is not even possibly
a substance, since it would not survive separation from its organism). But a hand is surely
not a quality or relation or member of any other Aristotelian category either.
21
There is also the option of rejecting the divide between bearer and property, and operat-
ing with a one-category ontology instead. In this vein Paul (2012) defends a property-only
ontology. Though on her approach there remains the question of whether the fundamental
properties are global or local, and nomic integrity could be used to argue that the most fun-
damental bundle is a bundle of global properties. Relatedly, Campbell—who advocates a
one-category trope ontology—claims, ‘All basic tropes are space-filling fields, each one of
them distributes some quantity, perhaps in varying intensities, across all of space-time’
(1990, p. 146), and thereby defends ‘Spinoza’s conclusion, that there is just one genuine
substance, the cosmos itself …’ (1990, p. 154).

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
82 I — JONATHAN SCHAFFER

it is at least not what Russell and Moore said was wrong. Given that
Spinozan Monism follows from two seemingly plausible premisses, I
ask: why not follow the argument where it leads?

IV

The argument from nomic integrity is not the only route from the
global character of the fundamental laws to a monistic conclusion. A
variant argument—the argument from cosmic substantiality—is
available, using weakened versions of both Leibnizian Substance
and Russellian Laws plus an Aristotelian principle about the mereol-
ogy of substances, to reach a restricted version of Spinozan Monism:
(4) Leibnizian Substance, Sufficiency: If something evolves by
the fundamental laws, then it is a substance.
(5) Russellian Laws, Positive: The cosmos evolves by the fun-
damental laws.
(6) Cosmic Substantiality: The cosmos is a substance (from 4
and 5).
(7) Aristotelian Principle: No proper part of a substance is a
substance.
(8) Spinozan Monism, Restricted: Among the cosmos and its
proper parts, the cosmos is the one and only substance
(from 6 and 7).
Leibnizian Substance, Sufficiency allows that there may be other
sorts of substances that lack the right connection with the funda-
mental laws for nomic integrity—it merely posits that such a con-
nection suffices for substantiality. And Russellian Laws, Positive
allows that there may be other systems that evolve by the funda-
mental laws—it merely says that the cosmos does. So even someone
who would reject Leibnizian Substance and/or Russellian Laws
might still accept these weakened counterparts. Together they entail
that the cosmos is a substance, as per Cosmic Substantiality.22

22
Cosmic Substantiality fits Aristotle’s claim in On the Heavens that ‘the whole or
totality’—by which he means ‘all body included within the extreme circumference’—forms
a substance (1984b, p. 462). This is the individual which, in hylomorphic terms, has the
sum of all matter as its matter and wholeness as its form. Thus, as Matthen observes, for
Aristotle, ‘the corporeal cosmos is a single substance with a motion proprietary to itself …’
(Matthen 2001, p. 171).

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
 FUNDAMENTAL REALITY 83

The argument from cosmic substantiality then proceeds through

Aristotelian Principle, which prohibits proper parthood relations be-
tween substances.23 This principle seems plausible for three reasons.
First, it befits the guiding claim that substances are not mere aggre-
grates but real unities. A substance is one. Something that has a sub-
stance as a proper part will not itself be one but must be more (a ‘one-
plus’). So Aristotle says, ‘a substance cannot consist of substances
present in it actually’ since anything consisting of two substances
must be ‘actually two’ and so ‘never actually one’ (1984c, p. 1640).
A second reason to support Aristotelian Principle is that it befits the
guiding claim that substances are fundamental entities, given the fur-
ther plausible thought that fundamental entities are open to free re-
combination (Schaffer 2010b). When entities are related as whole to
part, they are not open to free recombination, since these entities—
assuming that they retain their mereological relation—constrain each
other. To illustrate, the semicircles cannot be green if the circle is red.
A third reason to support Aristotelian Principle is that it befits the
Leibnizian conception of substances as the sites of fundamental
powers. Principles of action for both whole and part would yield an
implausible overdetermination of fundamental powers, yielding at
best redundancy (if the fundamental powers attributed to wholes
and parts happen to cohere), and at worst incoherence. To illustrate,
the semicircles cannot turn left if the circle turns right.
Of course, one might reject Aristotelian Principle for various rea-
sons, but perhaps the most pressing concern is that it seemingly con-
flicts with Leibnizian Substance (and equally with Leibnizian
Substance, Sufficiency). For what if both whole and proper part
evolve by the fundamental laws? What about a cosmos that contains
closed subsystems? For instance, consider a ‘two-monad’ world of
two non-interacting windowless bubbles, with purely inertial laws:
the whole system evolves by the fundamental laws, but so does each
monad. Leibnizian Substance would then rule that both whole and
proper part are substances, contradicting Aristotelian Principle.
But the conflict is only apparent. Leibnizian Substance and Aris-
totelian Principle are not modalized, and so say nothing about other
worlds (like the two-monad world). Of course Leibnizian Substance
and Aristotelian Principle are metaphysical claims, and so I think
23
Aristotelian Principle follows from my stronger tiling constraint on fundamental things
(Schaffer 2010a, pp. 38–42): the tiles that make the mosaic should cover the cosmos with-
out overlap.

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
84 I — JONATHAN SCHAFFER

they ought to follow from metaphysically necessary principles. But

that does not mean that the backing principles must be Leibnizian
Substance prefixed with a box plus Aristotelian Principle prefixed
with a box.
Indeed, there is a natural alternative metaphysical principle be-
hind Leibnizian Substance. Leibnizian Substance encodes the mini-
mal condition of match between prediction and behaviour. This
minimal condition is sufficiently discriminating for worlds like ours
in which prediction and behaviour match at exactly one mereologi-
cal level, but needs strengthening in worlds like the two-monad
world in which prediction and behaviour match at multiple mereo-
logical levels, to discriminate the level of match reflecting the work-
ings (‘governance’) of the fundamental laws from the level resulting
as a by-product:
Leibnizian Substance, Modalized It is metaphysically necessary
that something is a substance if and only if it has an evolution
governed by the fundamental laws.
Provided that governing is always restricted to a single mereological
level (to avoid overdetermination of fundamental powers), Leib-
nizian Substance, Modalized would not conflict with Aristotelian
Principle.24
There is also a natural alternative metaphysical basis for Aristote-
lian Principle. For overdetermination of fundamental powers—so
long as they cohere—yields a redundancy that seems not metaphys-
ically impossible but merely methodologically dispreferred:
Aristotelian Principle, Modalized It is metaphysically necessary
that if there is no redundancy then no proper part of a sub-
stance is a substance.
Aristotelian Principle, Modalized entails Aristotelian Principle at
worlds like ours that are (presumably) without redundancy, but
would not conflict with Leibnizian Substance.
In any case, whatever modal generalizations one ultimately pre-
fers, Leibnizian Substance and Aristotelian Principle can and should
stand together at our world. Both remain plausible claims for our
24
Leibnizian Substance, Modalized does not yet say which mereological level reflects the
workings of the fundamental laws in a two-monads world. That is a substantive matter. Still
the monist might argue that, given that the fundamental laws work at the level of the whole
in worlds like ours, it is plausible to extend this pattern.

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
 FUNDAMENTAL REALITY 85

world, and both can be seen as reflections of various compatible

metaphysical principles.
Cosmic Substantiality and Aristotelian Principle together entail
Spinozan Monism, Restricted, which is a restricted but still powerful
monism, ruling in the substantiality of the cosmos and ruling out the
substantiality of particles, organisms, and other major candidates for
the office. Indeed, as long as one gets to Cosmic Substantiality—
whether via Leibnizian Substance, Sufficiency plus Russellian Laws,
Positive, or in some other way—one can apply Aristotelian Principle
to infer Spinozan Monism, Restricted. The argument from cosmic
substantiality thus takes on a life of its own. The image of the cos-
mos as an integrated nomic system—one vast clockwork—merely
provides one reason to view the cosmos as a substance.25

Rutgers University
College Avenue Campus
1 Seminary Place
New Brunswick, nj 08901
usa

References

Aristotle 1984a: Categories. In Aristotle 1984d, vol. 1, pp. 3–24.

——1984b: On the Heavens. In Aristotle 1984d, vol. 1, pp. 447–511.
——1984c: Metaphysics. In Aristotle 1984d, vol. 2, pp. 1552–728.
——1984d: The Complete Works of Aristotle, ed. Jonathan Barnes. Prince-
ton, nj: Princeton University Press.
Bigelow, John, Brian Ellis, and Caroline Lierse 1992: ‘The World as One of
a Kind: Natural Necessity and Laws of Nature’. British Journal for the
Philosophy of Science, 43(3), pp. 371–88.
Butterfield, Jeremy 2006: ‘Against Pointillisme about Mechanics’. British
Journal for the Philosophy of Science, 57(4), pp. 709–53.
Campbell, Keith 1990: Abstract Particulars. Oxford: Basil Blackwell.

25
Thanks to Einar Bøhn, Ross Cameron, Shamik Dasgupta, Michael della Rocca, Martin
Glazier, Shieva Kleinschmidt, Martin Lin, Mohan Matthen, George Musser, Laurie Paul,
Michael Rea, Kelly Trogdon, Katja Vogt and Peter Vranas, as well as to audiences at
Columbia University, University of Wisconsin-Madison, and the participants at Reality-
making: Exploring Fundamentals in Metaphysics in Nottingham, the participants at Mini-
Workshop on Metaphysics and the Philosophy of Mind in Oslo, the members of the Notre
Dame Metaphysics and Philosophy of Religion Reading Group, and the members of the
Corridor Reading Group.

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
86 I — JONATHAN SCHAFFER

Cartwright, Nancy 1983: How the Laws of Physics Lie. Oxford: Clarendon
Press.
Chew, Geoffrey 1968: ‘“Bootstrap”: A Scientific Idea?’ Science, 161,
pp. 762–5.
Earman, John 1986: A Primer on Determinism. Dordrecht: D. Reidel.
Ellis, Brian 2001: Scientific Essentialism. Cambridge: Cambridge University
Press.
Everett, Hugh 1957: ‘“Relative State” Formulation of Quantum Mechan-
ics’. Review of Modern Physics, 29(3), pp. 454–62.
Hawley, John, and Katherine Holcombe 2005: Foundations of Modern
Cosmology, 2nd edn. Oxford: Oxford University Press.
Horgan, Terry, and Matjaž Potr cš 2008: Austere Realism: Contextual Se-
mantics Meets Minimal Ontology. Cambridge, ma: mit Press.
Joachim, Harold 1901: A Study of the Ethics of Spinoza: Ethica Ordine Ge-
ometrico Demonstrata. Oxford: Clarendon Press.
Leibniz, G. W. 1989: Philosophical Papers and Letters, 2nd edn. Edited and
translated by Leroy Loemker. Dordrecht: Kluwer.
——1998a: ‘Nature Itself; or, The Inherent Force and Activity of Created
Things—Confirming and Illustrating the Author’s Dynamics’. In Leibniz
1998c, pp. 209–22. Oxford: Oxford University Press.
——1998b: ‘Principles of Nature and Grace’. In Leibniz 1998c, pp. 258–
66.
——1998c: Philosophical Texts. Edited and translated by R. S. Woolhouse
and Richard Francks. Oxford: Oxford University Press.
Lewis, David 1983: ‘New Work for a Theory of Universals’. Australasian
Journal of Philosophy, 61(4), pp. 343–77.
Loewer, Barry 2007: ‘Counterfactuals and the Second Law’. In Richard
Corry and Huw Price (eds.), Causation, Physics, and the Constitution of
Reality: Russell’s Republic Revisited, pp. 293–326. Oxford: Oxford Uni-
versity Press.
Long, A. A., and D. N. Sedley 1987: The Hellenistic Philosophers, vol. 1:
Translations of the Principal Sources, with Philosophical Commentary.
Cambridge: Cambridge University Press.
Malebranche, Nicolas 1997: Dialogues on Metaphysics and on Religion.
Edited by Nicholas Jolley and David Scott. Cambridge: Cambridge Uni-
versity Press.
Matthen, Mohan 2001: ‘The Holistic Presuppositions of Aristotle’s Cos-
mology’. Oxford Studies in Ancient Philosophy, 20, pp. 171–99.
Maudlin, Tim 2007: ‘A Modest Proposal Concerning Laws, Counterfactu-
als, and Explanations’. In his The Metaphysics Within Physics, pp. 5–49.
Oxford: Oxford University Press.
Moore, G. E. 1993: ‘Proof of an External World’. In Thomas Baldwin (ed.),
G. E. Moore: Selected Writings, pp. 147–70. London: Routledge.

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x
 FUNDAMENTAL REALITY 87

Paul, L. A. 2012: ‘Building the World from its Fundamental Constituents’.

Philosophical Studies, 158(2), pp. 221–56.
Proclus 1979: Commentary on Plato’s Parmenides. Translated by Glen
R. Morrow and John M. Dillon. Princeton, nj: Princeton University Press.
Russell, Bertrand 1985: ‘The Philosophy of Logical Atomism’. In David
Pears (ed.), The Philosophy of Logical Atomism, pp. 35–155. La Salle,
il: Open Court Press.
——1992: ‘On the Notion of Cause’. In John Slater (ed.), The Collected Pa-
pers of Bertrand Russell, Volume 6: Logical and Philosophical Papers
1909–1913, pp. 193–210. London: Routledge.
Rutherford, Donald 1993: ‘Natures, Laws, and Miracles: The Roots of
Leibniz’s Critique of Occasionalism’. In Steven Nadler (ed.), Causation
in Early Modern Philosophy: Cartesianism, Occasionalism, and Pre-
established Harmony, pp. 135–58. University Park, pa: Pennsylvania
State University Press.
Schaffer, Jonathan 2010a: ‘Monism: The Priority of the Whole’. Philosoph-
ical Review, 119(1), pp. 31–76.
——2010b: ‘The Internal Relatedness of All Things’. Mind, 119, pp. 341–
76.
Spinoza, Benedict de 1994: A Spinoza Reader: The Ethics and Other
Works. Edited and translated by Edwin Curley. Princeton, nj: Princeton
University Press.

©2013 The Aristotelian Society

Proceedings of the Aristotelian Society Supplementary Volume lxxxvii
doi: 10.1111/j.1467-8349.2013.00220.x