The Fundamental Question of Metaphysics and

the Question of Fundamentality in Metaphysics

Brandon C. Look

Why is there something rather than nothing? This question is often considered the
fundamental question of metaphysics. This paper will concern a related question, one that is
arguably even more fundamental: Is nothingness possible? For it is only if nothingness is
possible that the fundamental question of metaphysics really develops any force. There are
two ways to justify the possibility of nothingness. One can hold that “nothing is simpler and
easier than something” – that nothingness is, as it were, the default state of the universe that
only an act of creation can overcome. Or one can argue, as Thomas Baldwin and others have
done, that it is possible to “subtract” concrete objects from worlds until one has arrived at a
world without concrete objects – an “empty world.” It will be argued in this paper that the
premises of the Subtraction Argument rest on tendentious and question-begging
assumptions about ontological dependence and the grounding relation. In other words,
questions of fundamentality in metaphysics reveal the fundamental question of metaphysics
to be ill-formed and arguments purporting to show the possibility of nothingness invalid.
Against the view of metaphysical nihilism, this paper argues for metaphysical aliquidism –
the view that there must be something.

1. Introduction

Why is there something rather than nothing? This question, Leibniz tells us in the Principles
of Nature and Grace, is the very first question that we ought to ask ourselves: “Assuming this
principle [the Principle of Sufficient Reason], the first question we have the right to ask will
be, why is there something rather than nothing?” 1 Indeed, Heidegger has called it “the
fundamental question of metaphysics [die Grundfrage der Metaphysik]”; he writes, “Die dem
Range nach erste, weil weiteste, tiefste und ursprünglichste Frage [ist]: »Warum ist
überhaupt Seiendes und nicht vielmehr Nichts?«” (1983, 3). Wittgenstein, too, seems to be
moved to awe: “Nicht wie die Welt ist, ist das Mystische, sondern dass sie ist” (1922, 186;
Tractatus 6.44). As fundamental as this question may be, it is not the question to be
addressed in this paper – at least not directly. Rather, this paper will concern a related
question, one that is arguably even more fundamental: Is the fundamental question of
metaphysics a well-formed and meaningful question? One might think, for example, that it is
a kind of complex question. The famous example, of course, is “Have you stopped beating
your wife?” Or, one might just recognize that Why?-questions are often problematic. The
fundamental question of metaphysics assumes that nothingness is possible, but is it? That is,
is nothingness possible? Or, given that there is obviously something, might there be (or might
there have been) nothing? For it is only if nothingness is possible that the fundamental
question really develops any force.

1“Ce principe posé, la premiere question qu’on a droit de faire, sera, Pourquoy il y a plustôt quelque
chose que rien?” (Leibniz 1965, hereafter ‘G’, VI 602)
 THE FUNDAMENTAL QUESTION 97

2. The Possibility of Nothingness

There are two ways to justify the possibility of nothingness. One can hold, as Leibniz himself
did, that “nothing is simpler and easier than something” (G VI 602) – that is, that
nothingness is the natural or default state and only an act of will, an act of free creation, can
overcome this. Put differently, for Leibniz, given the Principle of Sufficient Reason, being
requires a ground, or a reason, or a justification; nothingness does not require metaphysical
grounding. Yet, this is less an argument for the possibility of nothingness than it is an article
of faith. And, indeed, as I suggested above, the possibility of nothingness needs to be shown
given that there is something. And this is what Thomas Baldwin (1996), Gonzalo Rodriguez-
Pereyra (1997; 2000; 2002; 2004) and others have tried to do, to show that it is possible to
“subtract” concrete objects from worlds until one has arrived at a world without concrete
objects – an “empty world.” This “Subtraction Argument” for “metaphysical nihilism” has
occasioned a fair bit of discussion in the philosophical literature in the last decade, and it is
worth considering seriously. David Armstrong (2004, 89–91) has written approvingly of the
Subtraction Argument and has given an additional argument appealing to truthmakers for the
possibility of an empty world. Jonathan Lowe (1998; 2002, 252–55), Alexander Paseau
(2002; 2006), Ross Cameron (2006) and others, on the other hand, have countered that the
Subtraction Argument fails and that metaphysical nihilism is not possible. Moreover, David
Lewis (1986, 73–74) takes his modal realism – or rather his definition of a world – to block
the possibility of a world without any concrete objects, and David Armstrong, in earlier work
(1989, 24–25, 63–64), likewise rejects the possibility of an empty world.
The Subtraction Argument as advanced by Baldwin and Rodriquez-Pereyra is as follows. We
start with the following premises:
(1) There might be world with a finite domain of ‘concrete’ objects.
(2) These concrete objects are, each of them, things that might not exist.
(3) The non-existence of any of these things does not necessitate the existence of any
other such thing.
Starting from some possible world w1, accessible from our actual world, we pick out a
concrete object x1, remove it completely, and go on to the next world w2, which is qualitatively
identical with w1, except that it lacks x1. By the premises above, our subtraction of a particular
concrete object, xn, does not entail the existence of any “new” objects in other possible worlds.
Now, we are to continue this process to wmin, a world with one concrete object. And, finally,
according to the Subtraction Argument, we are to imagine removing the one concrete object
of wmin, whereby we arrive at wnil – a world with no concrete objects. Thus, if accessibility is
transitive, an empty world is possible, and metaphysical nihilism is tenable.
This argument looks good on its surface. But are the premises above reproach? Most of the
attention in the literature has been directed at premise (3), and I will turn to that in a
moment. But before getting there let’s take a moment to focus on premise (1) – a seemingly
innocent premise: “There might be world with a finite domain of ‘concrete’ objects.”
Certainly, we all know what is meant by such a claim. But is it beyond reproach? It seems to
me that one could easily argue against it on three related grounds. One might believe in an
actual infinity in the world of concrete objects, as Leibniz did, at least with respect to the
material objects of the phenomenal realm, that any particular part of matter is actually
divided into or composed of an infinity of other creatures: “I believe that there is no part of
matter that is not – I shall not say divisible – but actually divided, and consequently the least
98 LOOK

article ought to be considered as if it were a world full of an infinity of different creatures.” 2 I

doubt there are many advocates of the Leibnizian position, but there are some. 3 I suspect
there might be advocates of a related view: that there is no fundamental level to reality. Chairs
are made up of atoms, which are made up of protons, neutrons and electrons, which in turn
are made up of subatomic particles, and so on. An added twist, and one that seems to have
more going for it scientifically, is to say that there is a fundamental level, but that it is
different states of a quantum field. Perhaps this entails “vague objects”; perhaps it simply
means that the idea of singular, discrete concrete objects is in itself rather question-begging.
Of course, advocates of the Subtraction Argument are not going to let this stand in the way of
a nifty argument. And perhaps they should not. Again, it is obvious enough what is meant by
concrete object. And Rodriguez-Pereyra makes this explicit in a later piece, saying that “a
concrete* object is an object that is concrete, memberless, and a maximal occupant of a
connected region [in space].” (Rodriguez-Pereyra 2002, 73) In other words, it is stipulated
that the concrete objects of the Subtraction Argument are discrete, atomic, and fundamental.
As I mentioned, the majority of philosophers who have objected to the Subtraction Argument
have been concerned with premise (3): “the non-existence of any of the concrete objects does
not necessitate the existence of any other concrete object.” Are there reasons to believe that
this is true? Even Baldwin concedes that “[i]t is not .. easy to think of a direct argument for
this premise.” (1996, 235) On its surface, it would seem that the premise says simply that
concrete objects or substances are independent from each other; that is, it would seem that
this premise simply appeals to the independence-criterion of substancehood, which has a
long tradition in philosophy. According to Aristotle, “Some things can exist apart and some
cannot, and it is the former that are substances.” 4 Or as Descartes puts it in the Principles of
Philosophy, “By substance we can understand nothing other than a thing which exists in such
a way that it depends on no other thing for its existence.” 5 Of course, Descartes quickly adds
to this definition of substance that, strictly speaking, only God is truly independent but you
know what I mean. And Spinoza, of course, knew exactly what Descartes meant, going from
this independence criterion of substance to monism. But that’s another story.
The notion of metaphysical or ontological dependence is at issue in another sense in premise
(3), which comes out in Jonathan Lowe’s criticism of the premise. According to Lowe,
premise (3) leads to metaphysical nihilism, a view he rejects; therefore, premise (3) must
itself be rejected. It’s as simple as that. The positive argument for his contrary position is also
formally quite simple, though metaphysically complex:
(4) Some abstract objects, like natural numbers, exist necessarily.
(5) Abstract objects depend for their existence upon there being concrete entities.
Therefore,
(6) It is necessary that there are concrete entities. 6
Now, this argument takes us into very deep metaphysical waters. I agree with Lowe, but I
would like to find independent reasons to reject the Subtraction Argument.

2 “Ainsi je crois qu’il n’y a aucune partie de la matiere qui ne soit, je ne dis pas divisible, mais
actuellement divisée, et par consequent, la moindre particelle doit estre considerée comme un monde
plein d’une infinité de creatures differentes.” (Letter to Foucher, 1693; G I 416).
3 Consider the discipline of continuum mechanics. “The continuum theory regards matter as indefinitely

divisible.” (Lai, Rubin, and Krempl 2009, 1)

4 Metaphysics, XII.5: 1070b36-1071a1 from Aristotle (1984).

5 “Per substantiam nihil aliud intelligere possumus, quam rem quae ita existit, ut nulla alia re indigeat

ad existendum.” (Descartes 1996, VIII 24/Descartes 1985, I 210, translation altered).

6 This is Rodriguez-Pereyra’s version (2000, 335) of what is found in (Lowe 1998, 252–55).
 THE FUNDAMENTAL QUESTION 99

Alexander Paseau (2002) presents a more direct challenge to premise (3), arguing that on
either of two interpretations of (3) the argument is ultimately invalid. Labelling the finitely
many concrete objects of (1) and (2) x1 to xn, Paseau claims that (3) could mean that “the non-
existence of any given one of the finitely many objects .. does not necessitate the existence of
any other given one of these xi.” Or (3) could mean something stronger: “the non-existence of
any one of these xi does not necessitate that there is even one of the xi.” (2002, 74) But Paseau
asks us to consider a model, containing three entities: x1, x2, and x3. We, therefore, have the
following seven sets or worlds: {x1}, {x2}, { x3}, { x1, x2},{ x1, x3}, { x2, x3}, { x1, x2, x3}. It should
be easy to see that on this model, (1), (2), and either reading of (3) are all true, and yet there is
no empty set (“null world”). The argument is, therefore, invalid.
Rodriguez-Pereyra appears to concede that, on the above interpretations of (3), the
Subtraction Argument is invalid. But he claims that there is another reading of (3), which, is
more perspicuous and gets at the original intention. This time it is more explicitly expressed
in terms of possible worlds:
(3*) The non-existence of any of the xi that exist in w1 does not necessitate the existence
of any other concrete* object, whether or not these exist in w1. That is: for all worlds
w and for all the concreta* xi in w1, if xi exists in w then if there is a world w* where
xi does not exist, then there is a world w** where the only existing concreta* are
those of w except xi (i.e. w** is such that for every concrete* object y, y exists in w**
if and only if y ≠ xi and y exists in w). (2002, 172)
In other words, if xi is a contingent concrete object in w, there is some world w** that
contains everything in w except for xi. Thus, the Subtraction Argument can proceed as before:
we subtract contingent concrete entities until we arrive at some world wmin, which has only
one object, and there is nothing preventing us from subtracting that object, so that we have
wnil – the null world.
I shall not concern myself with the question of whether or not the possible-worlds
interpretation is equivalent to the original premise (3). Rather I wish to consider the amended
argument as it stands. The scenario the nihilist imagines is clear enough. Suppose we have the
following concrete objects: x1, x2, and x3; we then have the following sets or possible worlds:
{x1, x2, x3}, {x1, x2}, {x1, x3}, {x2, x3}, {x1}, { x2}, {x3}. Our largest world is wmax = {x1, x2, x3}. But
there are also singleton worlds, including w = {x1}. Since there is a w* that does not have x1 as
a member (e.g., {x2, x3}), x1 is contingent. Now, according to premise (3), there must also be a
world, w** = {Ø}, for the non-existence of x1 does not necessitate the existence of any other
concrete objects.
Even with this improvement to premise (3), I believe that there is still an obvious problem.
The Subtraction Argument may be acceptable in talking about the subtraction from wmax to
any of the worlds of two or one element. But, surely (!), the “subtraction” of concrete objects
from two-object worlds to singleton worlds differs greatly from the subtraction of lone
concrete objects from singleton worlds to a null-world. Indeed, it would seem that this is
deeply question-begging, for the null-world is presupposed in the very argument that is
intended to prove its possibility. Moreover, I think that the clause “there is a world w** where
the only existing concreta* are those of w except xi” is deceptive, for, when there are no
concreta in w other than xi, the world-mates of xi are described (and treated) as both existent
and non-existent. Further, we ought to be suspicious of the existential quantifier “there is a
world w**..,” absent any kind of definition of a world. Indeed, it would seem that subtraction
to an empty world is only possible if one has already objected to a pretty straightforward
notion of a world as a set of objects that are spatio-temporally or causally connected to each
other. As I said earlier, Lewis, for example, thinks that his definition of a world blocks the
very possibility of a null world at the outset, and I am inclined to agree with him.
100 LOOK

Although premise (3) seems to capture the apparently innocent notion of the independence
criterion of substance, we need to notice how truly odd the wording of this premise really is:
the non-existence of any concrete object does not necessitate the existence of any other
concrete object – in another possible world. How many dubious metaphysical ideas are at
work in this premise? We have here (a) the causal powers of absences or absences grounding
other states of affairs and, perhaps, (b) the causal powers or grounding of a state of affairs in
one world on another world. If premise (3) is taken to be metaphorical – just because you
take away an object, that doesn’t mean that another one has to pop up in its place – then
how seriously do we have to take the argument itself?

3. Reason and Ground

I have referred to Leibniz several times thus far, and let me do so again by returning to one of
his “great principles of all our reasoning” – the Principle of Sufficient Reason (PSR). The PSR
seems to be making something of a comeback among metaphysicians these days, and there is
a perfectly good reason that this rationalist workhorse is doing so. It is simply an ancestor to
the truth-maker principle endorsed by so many contemporary philosophers. Leibniz, of
course, employs the Principle of Sufficient Reason throughout his metaphysics, most
famously, in giving his cosmological argument for the existence of God. When we consider the
Subtraction Argument, however, we should realize that premise (3) violates the Principle of
Sufficient Reason in an odd way. The PSR, as Leibniz usually states it, is this:
PSR* (Nihil est sine ratione)
There is nothing for which there is not a reason why it is (and why it is so and not
otherwise).
Of course, the PSR, with its double-negation, can also be reformulated thus (a reformulation
that Leibniz himself also uses):
PSR** (Omne ens habet rationem)
For every being, there is a reason (or ground) for its being (and being so and not
otherwise).
While we have seen that there is a relatively standard and innocent independence criterion of
substancehood at work in premise (3), at the same time it would seem to violate a normal
reading of the Principle of Sufficient Reason as PSR**. Every finite contingent being must be
grounded in some other being. When we get to the Subtraction Argument’s wmin with only
one finite, contingent, concrete object, we can ask, supposing PSR**, where or what is its
ground? It should be noted that the Subtraction Argument’s “progression” from a wmax to a
wmin resembles Leibniz’s move from contingent, finite causes back to an original, contingent
being. For Leibniz, there’s no problem in identifying its ground: given PSR**, the ground is
ultramundane, God. But I do not believe any of the advocates of metaphysical nihilism would
follow Leibniz here. Rather, I believe they would simply reject the Principle of Sufficient
Reason (PSR**); that is, advocates of metaphysical nihilism require that PSR** not hold in
the case of premise (3).
But should the Principle of Sufficient Reason be so quickly jettisoned? When we focus on the
truth-maker relation, we can just as easily see that there is no metaphysical or ontological
ground, no reason, no truth-maker for any fact concerning a world of no concrete objects. 7
And if metaphysical nihilists reject the PSR**, then their argument is in trouble for a different

7 I realize there are ways of explaining absences in truth-maker lingo. But it is not entirely clear to me

that there can be a truth-maker for a proposition about a world of no concrete objects. At the very least,
it seems to me that some work needs to be done here.
 THE FUNDAMENTAL QUESTION 101

– though related – reason. The Principle of Sufficient Reason, or the notion of metaphysical
or ontological grounding, underlies the principles that lead us to say that there is some
fundamental level of reality. They are the principles that seem to prevent us from saying, “It’s
turtles all the way down..” Therefore, denying the Principle of Sufficient Reason or the idea
that there must be an ontological ground for any being undermines the idea that allowed us to
admit the finite, concrete objects of premise (1). In other words, if the metaphysical nihilist
accepts some kind of grounding thesis to motivate his intuitions about finite, concrete objects
in premise (1), then he should likewise admit of a grounding thesis in premise (3). If he denies
a grounding thesis in premise (3), then he cannot have the finite, concrete objects that he
wants for premise (1).
A conciliatory (or purposefully ambiguous) conclusion would be that the Subtraction
Argument proves nothing. But I should like to be more blunt: in my view, the Subtraction
Argument fails. Its explicit and implicit premises rest on tendentious and question-begging
assumptions about ontological dependence and the grounding relation. 8 For example, the
crucial premise (“the non-existence of any of the concreta does not necessitate the existence
of any other concrete object”) assumes that there is no ontological dependence of one
concrete thing on another. The argument assumes that abstract objects are not grounded in
concrete objects – which might be taken as a reason to reject the argument prima facie. And
the argument assumes that a world is not grounded in or dependent upon concrete objects.
While the Leibnizian intuition that nothingness is more natural than something and that
therefore there must always be a reason or ground for the existence of something, the
Subtraction Argument assumes that its empty worlds, constituted presumably by abstracta,
can exist and can exist so simply by fiat. Or rather, it is assumed that there is no reason or
ground for the worlds’ So-sein – which may be as metaphysically presumptuous as anything
Leibniz dreamed of. Thus, questions of fundamentality in metaphysics reveal the
fundamental question of metaphysics to be ill-formed and arguments purporting to show the
possibility of nothingness invalid.

4. The Fundamental Question Forsaken

A lot is riding on the outcome of this dispute. If we can show that metaphysical nihilism is not
possible (that is, that it is not possible that there be nothing), it seems that we can undercut
the motivation of the fundamental question of metaphysics. For it is only when nothingness is
possible that we need to address the question why there is something rather than nothing.
Naturally, if the fundamental question loses its force, so too does any form of the
cosmological argument for the existence of God. If, on the other hand, it is impossible that
there be nothing, something must exist. Therefore, either we need give no explanation of the
fact that something exists – it’s a brute fact – or we say, as Spinoza and Russell did, that the
world is, as it were, a causa sui. Of course, one could demand why things are this way and not
some other way, but that is a different question, answerable in large part (perhaps
completely) by natural science. Thus, if metaphysical nihilism is impossible and if we
recognize that the Leibnizian prejudice that nothingness is simpler and easier than being is
simply a prejudice, then the fundamental question of metaphysics can no longer be sensibly
asked.
In this paper, I have not shown that metaphysical nihilism is impossible, only that one of the
contemporary arguments for it is wanting. But the Subtraction Argument’s weakness should
also alert us to some problematic aspects of any argument purporting to show the possibility
of nothingness, and my analysis should also lead to a scepticism with respect to the
fundamental question of metaphysics itself. For example, it is unlikely that an argument for

8 On these matters, see Correia (2008), Fine (2010), and Rosen (2010).
102 LOOK

metaphysical nihilism can be produced that does not depend upon a controversial or
ambiguous notion of a “world.” Moreover, the fundamental question of metaphysics derives
much of its force from the historically and culturally contingent thought that “God could have
chosen not to create anything” – a thought that should hardly move a steely-eyed atheist. The
fundamental question of metaphysics also implicitly appeals to the thesis that conceivability
entails possibility – for while I can (perhaps) think of a world prior to creation or not having
any “stuff” in it, this thought differs from a proof of its metaphysical possibility. Finally, the
fundamental question also derives force from the completely unanalyzed claim that “nothing
is simpler and easier than something.” There is, then, a presumption towards nothingness
that underlies the fundamental question of metaphysics that I reject.
It seems rather that the burden of proof should really be upon those who advocate
metaphysical nihilism to show that it is possible that nothing exist. To the credit of advocates
of the Subtraction Argument, they have tried to provide such an argument. But, as I hope to
have shown, this attempt is a failure, and to date I have been unconvinced of any attempt to
prove the possibility of nothingness. Indeed, the contrary view, that there must be something,
should become the default position, a position I should like to call “metaphysical aliquidism”
– the view that there must be something. There are two versions of metaphysical aliquidism.
The first might be seen as a kind of monism; its thesis is simply that there is a necessarily
existing concrete object – the world. Such a view is attributable, of course, to Spinoza, Russell
and, more recently, Jonathan Schaffer (e.g. in Schaffer (2009; 2010)). The second version can
be expressed thus: for any world, necessarily, there is a concrete object in it. This view is
advocated by Lewis – though one need not be a modal realist to endorse it. Either version of
aliquidism, I believe, is more in line with our intuitions of ontological dependence. And if we
accept metaphysical aliquidism, the fundamental question of metaphysics ought no longer to
be a question for us. 9

Brandon C. Look
University of Kentucky
look@uky.edu

References

Aristotle. 1984: The Complete Works of Aristotle. J. Barnes (ed.). 2 vols. Princeton: Princeton
University Press.
Armstrong, D. M. 1989: A Combinatorial Theory of Possibility. Cambridge: Cambridge
University Press.
— 2004. Truth and Truthmakers. Cambridge: Cambridge University Press.
Baldwin, T. 1996: ‘There Might Be Nothing’, Analysis 56, 231–38.
Cameron, R. 2006: ‘Much Ado About Nothing: A Study of Metaphysical Nihilism’, Erkenntnis
64, 193–222.
Correia, F. 2008: ‘Ontological Dependence’, Philosophy Compass 3, 1013–1032.
Descartes, R. 1985: The Philosophical Writings of Descartes. J. Cottingham, R. Stoothoff, and
D. Murdoch (eds.). 2 vols. Cambridge: Cambridge University Press.
Descartes, R. 1996: Oeuvres de Descartes. C. Adam and P. Tannery (eds.). 11 vols. Paris: J.
Vrin.
Fine, K. 2010: ‘Some Puzzles of Ground’, Notre Dame Journal of Formal Logic 51, 97–118.

9 My thanks to the audience at the GAP8 conference for helpful questions and suggestions.
 THE FUNDAMENTAL QUESTION 103

Heidegger, M. 1983: Einführung in die Metaphysik. Frankfurt a.M.: Vittorio Klostermann.

Lai, W. M., D. Rubin, and E. Krempl. 2009: Introduction to Continuum Mechanics. 4th ed.
Amsterdam: Elsevier.
Leibniz, G.W. 1965: Die Philosophischen Schriften. C. I. Gerhardt (ed.) 7 vols. Hildesheim:
Olms.
Lewis, D. 1986: On the Plurality of Worlds. Oxford: Basil Blackwell.
Lowe, E.J. 1998: The Possibility of Metaphysics: Substance, Identity, and Time. Oxford:
Oxford University Press.
— 2002: ‘Metaphysical Nihilism and the Subtraction Argument,’ Analysis 62, 62–73.
Paseau, A. 2002: ‘Why the Subtraction Argument Does Not Add Up’, Analysis 62, 73–75.
— 2006: ‘The Subtraction Argument(s)’, Dialectica 60, 145–56.
Rodriguez-Pereyra, G. 1997: ‘There Might Be Nothing: The Subtraction Argument Improved’,
Analysis 57, 159–66.
— 2000: ‘Lowe’s Argument Against Nihilism’, Analysis 60, 335–40.
— 2002: ‘Metaphysical Nihilism Defended: Reply to Lowe and Paseau’, Analysis 62, 172–80.
— 2004: ‘Modal Realism and Metaphysical Nihilism’, Mind 113, 683–704.
Rosen, G. 2010: ‘Metaphysical Dependence: Grounding and Relation’, in B. Hale and A.
Hoffmann (eds.): Modality: Metaphysics, Logic, and Epistemology, Oxford: Oxford
University Press, 109–35.
Schaffer, J. 2009: ‘On What Grounds What’, in D. Chalmers, D. Manley, and R. Wasserman
(eds.): Metametaphysics: New Essays on the Foundations of Ontology, Oxford: Oxford
University Press, 347–83.
— 2010: ‘Monism: The Priority of the Whole’, Philosophical Review 119, 31–76.
Wittgenstein, L. 1922: Tractatus-Logico Philosophicus. London: Routledge & Kegan Paul.