Professional Documents
Culture Documents
Does Ontology Exist?
Does Ontology Exist?
Archive
University of Zurich
Main Library
Strickhofstrasse 39
CH-8057 Zurich
www.zora.uzh.ch
Year: 2002
DOI: https://doi.org/10.1017/S0031819102000268
Wittgenstein, Carnap and Ryle.2 There are reasons for thinking that
ontology does not exist, not, that is, if it is conceived as a distinc-
tively philosophical investigation into the components of reality. Of
course there are ontologists, people who attend conferences on the
mereology of artefacts or write articles about the causal role of
tropes. But their endeavours are best understood as contributions to
a certain kind of conceptual analysis. Instead of establishing
whether, for example, tropes really exist, they in effect elucidate the
concept of a property and its relation to the concept of a bearer of
a property, or the role that property concepts play in causal expla-
nations.
Analytic ontologists tend to deny the existence of even the most
mundane and familiar things. Eliminative materialists, for example,
deny that there are minds. In a similar vein, Quine, who prides him-
self on a robust sense of reality, is nevertheless driven to the con-
tention that the only things that strictly speaking exist are pure sets,
that is, sets which do not even contain any individuals (TT 17–8).3
Such extraordinary claims lend at least prima facie support to the
idea that philosophy cannot contribute directly to the investigation
of reality by other disciplines. Philosophers are good at arguing,
2
Such an approach has recently been defended by S. Yablo, ‘Does
Ontology Rest on a Mistake?’, Proceedings of the Aristotelian Society
Supplementary LXXII (1998). But unlike Yablo, I do not take my cue
exclusively from Carnap. And although I defend a version of the exter-
nal/internal distinction, this defence differs from his. He contends that
internal claims like ‘There is a city of Chicago’ are part of a make believe
game. This contention is at odds with Carnap, because it treats internal
rather than external statements as non-factual. It is also implausible,
because, as Yablo himself writes (259), statements do not come any more
literal than that. In any event, the external/internal distinction is less
important to my deflationary case than considerations about quantifica-
tion, the concepts of existence and of an object, and the nature of logical
paraphrase.
3
References to works by Quine will be abbreviated as follows: ML—
Mathematical Logic (New York: Harper, 1952; 1. edn. 1940); FLPV—
From a Logical Point of View (Cambridge, Mass.: Harvard University
Press, 1980; 1. edn. 1953); WO—Word and Object (Cambridge, Mass.:
MIT Press, 1960); WP—Ways of Paradox and Other Essays (Cambridge,
Mass.: Harvard University Press, 1976; 1. edn. 1966); OR—Ontological
Relativity and Other Essays (New York: Columbia University Press, 1969);
PL—Philosophy of Logic (Englewood Cliffs, N.J.: Prentice-Hall, 1970);
TT—Theories and Things (Cambridge, Mass.: Harvard University Press,
1981); PT—Pursuit of Truth (Cambridge, Mass.: Harvard University
Press, 1992; 1. edn. 1990).
236
There are four main ideas that have contributed to the rise of ana-
lytic ontology. In rough chronological order, these are: Quine’s
naturalistic conception of ontology, Strawson’s revival of what he
calls ‘descriptive metaphysics’, the essentialist metaphysics derived
from Kripke’s and Putnam’s realist semantics, and the Austro-
Australian ‘truth-maker principle’.
In this section I shall discuss the last three. Of course, I cannot
do them full justice in such a short space. But I want to do indicate
how one might resist the conclusion that they furnish philosophical
insights into the constitution and essence of reality. My qualms are
ultimately fuelled by a Kantian worry about ontology.4 Kant main-
tained that there could be no a priori truths about reality, because
experience is our only way of finding out about reality. Even if there
are synthetic a priori truths, these are not de re. Instead of
describing mind-independent essences of objects, they articulate
‘necessary preconditions for the experience of objects’, that is, the
essential features of the way we experience them. The ontological
search for essences is thus transformed into a second-order
reflection on our conceptual scheme.
Strawson’s descriptive metaphysics respects this stricture, and is
metaphysics only in a Kantian sense. It investigates not de re
essences, but the conceptual framework of our thought and expe-
rience. Strawson’s aim is ‘to describe the actual structure of our
thought about the world’, ‘the most general features of our con-
ceptual structure’. To be sure, he talks about relations of ‘onto-
logical priority’ between different types of objects. In fact, how-
ever, what he has in mind are connections and dependencies
between various parts of our conceptual framework. A type of
thing X is ontologically prior to a type of thing Y not because it
underlies or explains the latter’s existence. Rather, objects of type
X are ontologically prior to objects of type Y if and only if the
identification and re-identification of objects of type X is presup-
posed by the identification and re-identification of objects of type
Y, but not vice versa.5
The ontological revival inspired by Kripke and Putnam also pays
homage to Kant. It grants that we cannot establish the real essences
of things through a priori Wesensschau, while insisting that we can
Critique of Pure Reason A 247/B 303.
4
Quinton, The Nature of Things (London: Routledge & Kegan Paul, 1973),
246–7.
238
If these reflections are along the right lines, among the current
contenders, only Quine remains in the running. His work was the
first and most influential force behind the rise of analytic ontology.
Furthermore, in two respects it is the most auspicious. Analytic
ontologists tend to disparage epistemological and methodological
questions as derivative. As a result, they are often oblivious to the
challenge of explaining how philosophy can gain insight into reality
and essences. By contrast, Quine assigns pride of place to the ques-
tion of how we acquire our knowledge about the world, and he has
made sophisticated contributions to the methodological questions
of what ontology is and how it should proceed.
Quine also avoids the Kantian objection ab initio, since he
ridicules the idea that philosophers are capable of getting essences
into the hair-crosses of their intellectual periscopes, whether by a
priori or by empirical means. He also denies that a priori philo-
sophical reflection can establish what kinds of things there are. In
short, Quine is ‘no champion of traditional metaphysics’.
Nevertheless, he finds a use for ‘the crusty old word’ ‘ontology’,
which he regards as nuclear to its traditional employment (WP
203–4). Like traditional ontology, Quine’s naturalistic ontology
seeks to establish what kinds of things there are. But it does not pur-
sue this aspiration directly or in isolation. Instead, it attempts to
help science in drawing up an inventory of the world.
4. Ontological Evasions
Platonism holds that abstract objects like numbers, properties and
propositions inhabit a super-natural world beyond space, time and
causation, a world to which we—or at any rate suitably trained logi-
cians and mathematicians—have access by a kind of ‘intellectual
intuition’. Nominalists protest that this hinterworld is a myth, and
that the signs which apparently refer to its denizens are flatus vocis
that can be avoided in respectable discourse.
Ontological deflationists try to undermine this debate. For them,
the way to avoid Platonism does not lie in purging our theories of
reference to and quantification over abstract objects. Like Quine,
they view the prospects of this enterprise as dim. Unlike Quine,
they regard the attempt to avoid intensional abstract terms as equally
futile. At the same time, the deflationists do not accept that such
failure would lumber us with mysterious entities. Once we clarify
what it is for abstract singular terms to refer, they claim, we shall
realize either that they do not refer to genuine objects or existents at
all, or that the existence of abstract objects is at any rate a perfectly
intelligible and low-key affair.
How convincing are these pleas of ontological innocence? I shall
discuss this question by considering three different versions of
deflationism, which I shall call existential, objectual and linguistic.18
Existential deflationism is based on Ryle’s claim that the term
‘exists’ is ambiguous.19 In the sturdy spatio-temporal sense in which
material objects exist, abstract objects do not, but they were never
meant to by our pre-theoretical statements about numbers, proper-
ties or propositions. The dispute between Platonism and nominal-
ism would then be based on a failure to keep apart these different
18
A position that I shall ignore here is contextual deflationism, the view
that abstract singular terms are harmless because of Frege’s context-prin-
ciple. See Dummett, op. cit., ch. 14, and, for a critical discussion, B. Hale,
Abstract Objects (Oxford: Blackwell, 1987), ch. 7.
19
The Concept of Mind (Harmondsworth: Penguin, 1980; 1. edn. 1949),
23–4.
249
20
WO 241–2, §27; FLPV 3, 131; Rundle, op. cit., §3.
250
31
Entity and Identity (Oxford: Clarendon, 1997), 58.
257
6. Conclusion
33
I am grateful to the Alexander von Humboldt Foundation for a
research grant. For comments and advice I should like to thank David
Bakhurst, Ansgar Beckermann, Peter Hacker, Anthony O’Hear, John
Hyman, Wolfgang Künne, Christian Nimtz, Jay Rosenberg and Eike von
Savigny, as well as audiences at Bielefeld, Konstanz, Vercelli and Kingston
(Ontario).
260