The Evolution of Modern Metaphysics: Making Sense of Things (87 page)

One might think that it is quite immune to this complaint. After all, what greater conceptual revision could there be than to challenge our fundamental conviction that every thought is either true or not true? Furthermore, Dummett himself expressly says that, despite his great admiration for Wittgenstein, he sees no reason to share Wittgenstein’s ‘belief that philosophy, as such, must never criticise but only describe’ (
Logical Basis
, p. xi).

But let us not forget that Wittgenstein too was prepared to criticize, when he thought that he was dealing with failed attempts to make linguistic sense. Indeed, in his reflections on mathematics, his criticisms were not unlike Dummett’s.
⁶²
The significant issue – here yet again as in Wittgenstein – is not what metaphysical rationale there is for correcting the use to which
we put the concepts that we already have, but what metaphysical rationale there may be for having new concepts. What scope is there, to paraphrase the Novelty Question, for making sense of things in a way that is not only maximally general but also radically new? There is not quite the animus against such radical innovation in Dummett that there was in Wittgenstein; that is true. But for an enthusiastic reception of such innovation we must wait until
Part Three
.

1
See esp.
Frege
I,
Frege
II, and
Frege
III. There are also numerous articles, of which ‘Frege’, ‘Frege and Analysis’, and ‘Frege as a Realist’ stand out.

Note: throughout this chapter I use the following abbreviations for Dummett’s works: ‘Allard’ for Dummett (
2007d
); ‘Analytical Philosophy’ for Dummett (
1978m
); ‘Autobiography’ for Dummett (
2007a
); ‘Beards’ for Dummett (
2007h
); ‘Campbell’ for Dummett (
2007e
); ‘Deduction’ for Dummett (
1978j
); ‘Fitch’s Paradox’ for Dummett (
2009
); ‘Frege’ for Dummett (
1978c
);
Frege
I for Dummett (
1981a
);
Frege
II for Dummett (
1981b
);
Frege
III for Dummett (
1991a
); ‘Frege and Analysis’ for Dummett (
1991c
); ‘Frege and Kant’ for Dummett (
1991f
); ‘Frege and Wittgenstein’, for Dummett (
1991g
); ‘Frege as a Realist’ for Dummett (
1991d
); ‘Gödel’s Theorem’ for Dummett (
1978f
); ‘Indeterminacy’ for Dummett (
1978l
);
Intuitionism
for Dummett (
2000
); ‘Intuitionistic Logic’ for Dummett (
1978g
); ‘Knowledge of a Language’ for Dummett (
1993c
);
Logical Basis
for Dummet (1991b); ‘Mathematics’ for Dummett (
1993e
); ‘McDowell’ for Dummett (
2007f
); ‘McGuiness’ for Dummett (
2007b
);
Origins
for Dummett (
1993a
); ‘Pears’ for Dummett (
1994
); ‘Preface’ for Dummett (
1978a
); ‘Realism and Anti-Realism’ for Dummett (
1993f
); ‘Rumfitt’ for Dummett (
2007g
); ‘Sense and Reference’ for Dummett (
1978d
); ‘Theory of Meaning (II)’ for Dummett (
1993b
); ‘The Past’ for Dummett (
1978k
);
Thought and Reality
for Dummett (
2006
); ‘Truth’ for Dummett (
1978b
);
Truth and the Past
for Dummett (
2004
); ‘Verificationism’ for Dummett (
1992
); ‘Victor’s Error’ for Dummett (
2001a
); ‘Wang’s Paradox’ for Dummett (
1978h
); and ‘Wittgenstein’s Philosophy of Mathematics’ for Dummett (
1978e
).

2
See again the material cited in
Ch. 8
, §1.

3
Within a month of Frege’s death.

4
Cf.
Origins
, Chs 13 and 14 passim.

5
This way of putting Dummett’s thought is due to John McDowell: see McDowell (
1996
), p. 124.

6
In the opening sentences of
Ch. 1
of
Thought and Reality
, Dummett characterizes them as questions about what there is. This characterization looks somewhat different, but they are really equivalent. Within a couple of pages he explains that by ‘what there is’ he means not just ‘what kinds of
object
there are’ but ‘what kinds of
fact
obtain’ (pp. 2–3, emphasis in original).

7
Linguistic sense includes what I called in
Ch. 9
, §7, ‘propositional sense’. But it includes more besides; e.g. it includes sense of whatever kind attaches to linguistic items other than full declarative sentences.

8
This is but one illustration, among the many that we have witnessed in this enquiry, of the inhibiting and disconcerting effect that heightened self-consciousness can have.

9
This is the triad of questions to which Kant assigned such central importance: see
Ch. 5
, n. 44.

10
On Wittgenstein, see ‘Analytical Philosophy’, ‘Frege and Wittgenstein’,
Origins
, pp. 164–166, ‘Pears’, and ‘McGuiness’. On Quine, see ‘Sense and Reference’, pp. 134–140, and ‘Indeterminacy’.

11
Unlike
Quine
we shall need to take seriously linguistic sense itself, suitably construed. Among other things this will entail a recoil from the idea that what are confirmed or confuted by different possible courses of sense experience are only ever entire bodies of theory (though not a recoil from the idea that the confirmation or confutation of individual statements by different possible courses of sense experience is sometimes intelligible only in relation to other statements): see e.g. ‘Deduction’, pp. 304–305; ‘Indeterminacy’, p. 382;
Logical Basis
,
Ch. 10
; and
Origins
, pp. 190–191. It will also entail a rehabilitation of the distinction between analytic truths and synthetic truths: see e.g. ‘Indeterminacy’, pp. 414–415. Both these points bear on what we saw Quine argue in
Ch. 12
, §4.

12
See ‘Verificationism’.

13
This connects with something on which Dummett has time and again insisted, namely that a satisfactory theory of linguistic sense must deliver a satisfactory theory of the grasp of it; a satisfactory theory, in other words, of linguistic understanding. Among countless references, see e.g.
Origins
, p. 11.

14
For Dummett’s opposition to Quine’s naturalism, see ‘McGuiness’, p. 51.

15
See the references in the discussion in
Ch. 8
, §6, esp. those in n. 56. See also
Thought and Reality
, pp. 9ff.

16
For a succinct account of what such a theory must
deliver
, see
Thought and Reality
, pp. 14–15.

17
Dummett has ‘propositions’ where I have inserted ‘thoughts’. But he shows elsewhere (e.g.
Thought and Reality
, p. 9) that he is equally comfortable with the identification of facts with true thoughts; and the reference to thoughts here makes the connections I wish to emphasize more graphic. For discussion of the relation between ‘thoughts’ and ‘propositions’, see
Thought and Reality
, pp. 4ff. and 29–30.

18
Dummett (
1991b
) – which I am abbreviating as
Logical Basis
.

19
Cf.
Frege
I, pp. 671–672;
Frege
II, pp. 66–67; and
Logical Basis
, pp. 10ff. Cf. also ‘Preface’, p. xl, where he writes, ‘The whole point of my approach to these problems [i.e. fundamental problems about the relations between realty and our capacity to know it] has been to show that the theory of meaning underlies metaphysics.’ But note that Dummett is thereby forced to conceive his own work, which is primarily concerned with what a systematic theory of linguistic sense must look like, less as a contribution to metaphysics than as a prolegomenon to metaphysics (cf.
Logical Basis
, pp. x–xi). At the end of his Introduction to
Logical Basis
he writes, ‘The layman wants the philosopher to give him a reason for believing, or for disbelieving, in God, in free will, or in immortality…. I am not proposing to answer [such questions]. I propose only to try to provide a base from which we might set out to seek for the answers’ (p. 19). Not that he is inclined to be apologetic about this. He goes on to remark, ‘Philosophical writing of the past, and of the present day as well, supplies answers to the great questions of metaphysics; and the answers usually satisfy no one but their authors…. I believe that we shall make faster progress only if we go at our task more slowly and methodically’ (ibid.). In this there is something at once curiously reminiscent of, and strikingly different from, the boldness that we saw in the prefaces to Kant’s first
Critique
and Wittgenstein’s
Tractatus
: see
Ch. 5
, §2, and
Ch. 9
, §2, respectively; and cf.
Ch. 9
, n. 21.

– For opposition to Dummett’s view that (all of) metaphysics has a logical base, see Blackburn (
1996
), pp. 76–79.

20
Dummett himself would insert the word ‘determinately’ before ‘either’: see
Frege
II, pp. 435–436. I remain unconvinced that this makes the difference he says it makes. But if you agree with Dummett, then take as read the insertion of ‘determinately’ in all the relevant contexts hereafter.

21
For discussion of some of the issues involved, see Russell (
1993
) and Strawson (
1993
). And see again
Ch. 8
, n. 41, for the controversy concerning what Frege would say.

22
It is part of Leibniz’ principle of contradiction, one of the two fundamental
a priori
principles that he recognized; see
Ch. 3
, §1.

23
By what standards? Conclusively? Beyond reasonable doubt? For any relevant practical purposes? It is an extremely important feature of Dummett’s line of thought that this matter should remain unresolved. As will become clear, his concern is not with any one clearly delineated scruple. It is with a family of scruples.

24
Cf. Wittgenstein (
1967a
), Pt I, §§133–143 and 242; and cf. Quine’s semantic empiricism (
Ch. 12
, §2).

25
See esp. ‘Truth’; ‘Intuitionistic Logic’; ‘The Past’;
Frege
II,
Ch. 20
; ‘Theory of Meaning (II)’; and ‘Realism and Anti-Realism’. The literature on this argument is vast. See e.g. McDowell (
1976
) and (2007a); McGinn (
1979
); Craig (
1982
); Wright (
1992
); Williamson (
1994b
), (
2007
), pp. 281–284; and Campbell (
2007
). For suspicion of the whole project, see P.F. Strawson (
1976
–1977), p. 21, where Strawson writes, ‘Few things are more implausible that [
sic
] the idea that we can be rapidly forced into a wholesale revision … of our metaphysics … by a dogmatic interpretation of the observation, in itself irreproachable, that our understanding of a language is manifested only in our use of it.’ (Cf. Diamond (
1993
), Lecture Two, §XI.)

26
Or if not a contribution to metaphysics, then at least part of a prolegomenon to metaphysics (see n. 19). For further discussion of the connection with metaphysics, see
Logical Basis
, pp. 325–327, and
Intuitionism
, p. 267.

27
He also, less frequently, considers the idea that the thought expressed by a declarative sentence is a matter of how things must be in order for us to recognize that the sentence is false. This idea is well explored in Rumfitt (
2007
), to which Dummett responds (enthusiastically) in ‘Rumfitt’.

28
In
Frege
I, pp. 683–684, Dummett suggests that it was ‘historically necessary’, if not ‘logically necessary’, for Frege to be immune to this alternative. This is because the revolution that Dummett takes Frege to have effected involved a retreat from various mind-centred approaches to philosophy that were dominant at the time, and his realism helped him to keep these at bay. For further discussion, in relation to Kant, Fichte, and Hegel, see
Frege
II, pp. 496–500.