The Evolution of Modern Metaphysics: Making Sense of Things (61 page)

22
1.1 mentions ‘things’ rather than ‘objects’, but 2.01 suggests that these terms are interchangeable.

23
We shall see the importance of this qualification in §8.

24
For two superb compendia, see McGuiness (
1990
),
Ch. 9
, and Sullivan (
2004
). For an excellent, much fuller account, see Morris (
2008
).

25
See 4.46ff. It is surprisingly common for commentators to say that it
is
to deny these things: see e.g. McGuiness (
1990
), p. 312, and Floyd (
2000
), p. 241. Admittedly, there is material in the 4.06s that may appear to confirm their view. But 4.064 states that ‘every proposition must … have a sense,’ and this suggests that the 4.06s are tacitly concerned only with those propositions that are not senseless.

26
The material in this paragraph draws especially on the 1s, the 2s, the early 3s, and the mid 4s. Question: does Wittgenstein retain anything like Frege’s sense/
Bedeutung
distinction? Answer: not except on an extremely generous interpretation of ‘anything like’. He holds that the only expressions that have senses are propositions, and that propositions do not stand to anything in a relation akin to that in which names stand to objects. Ironically, one of the places where Wittgenstein most clearly registers this departure from Frege is 3.3, where he also, at the same time, commits himself to a version of Frege’s context principle. (See further Carruthers (
1989
),
Ch. 3
.)

27
I am greatly indebted to Cora Diamond in this paragraph. See e.g. Diamond (
1991a
), pp. 30–31, and (
1991c
), pp. 106–107, the latter including quotations from Wittgenstein’s later work. (For dissent, see Hacker (
2000
), pp. 365ff.) Diamond traces many of Wittgenstein’s ideas back to Frege. The reference above to an object’s not being able to satisfy another object should certainly have rung Fregean bells; see §7 of the previous chapter. But we should pause to note an important corollary of this. Just as there is a problem, for Frege, about the
Bedeutungen
of predicates, so too there is a problem, for Wittgenstein – indeed there is a related problem for Frege – about predicates themselves. If a proposition is a fact, if a name is an object (and hence something that can itself be named), and if a predicate is what results when a name is removed from a proposition, then a predicate is not an object (it is not something that can be named). So just as there is something untoward about the predicate ‘… is a property’, so too there is something untoward about the predicate ‘… is a predicate’. See further §5.

28
It can be
known
. But to know that it is either raining or not raining is not to know anything about reality (4.461), and to accredit someone with the knowledge is not to say anything about reality (5.1362). (I cannot however resist reproducing an excellent joke in this connection from Brian McGuiness (
1990
). McGuiness himself quotes 4.461: ‘I know nothing about the weather when I know that it is either raining or not raining.’ He then adds a footnote in which he comments laconically, ‘[Wittgenstein] had been out of England for some time when he wrote this’ (
Ch. 9
, n. 21).)

29
For very helpful overviews, see Conant (
1991
), pp. 137ff., and Hacker (
2001a
), §5.

30
Thus a disjunction of two propositions
p
₁
and
p
₂
does not depend for its truth on the behaviour of some function of the sort described above, at work in the world. It depends solely on the truth of
p
₁
and
p
₂
. And ‘unless’ is just a sign that enables us to produce the disjunction of any two propositions. It does not designate anything in reality.

31
What about epistemic possibility? Cannot imperfections in our understanding render that which is logically impossible epistemically possible? Not on Wittgenstein’s view. Cf. §2: imperfect understanding, for Wittgenstein, is not understanding at all. In fact, in a quite trivial sense, anyone who understands anything knows that any given logical impossibility is false: see n. 28.

32
Cf. 6.123. And cf. Diamond (
1993
), Lecture One, §IV.

33
Cf. Sullivan (
2003
), esp. pp. 209–211, and (
2004
), p. 34. And cf. certain currents of thought in Wittgenstein’s later work. E.g. at one point in
Philosophical Grammar
, commenting on the infinitude of the sequence of cardinal numbers, Wittgenstein insists that we should not say, ‘There is no largest cardinal number’ – as though we were excluding some possibility – but should rather say, ‘The expression “last cardinal number” makes no sense’ (p. 465). And in the
Investigations
, having remarked that ‘
essence
is expressed by grammar’, he says of a puzzle that he is wrestling with there, ‘The great difficulty is not to represent the matter as if
there were something
one
couldn’t
do’ (§§371 and 374, first and third emphasis in original, second emphasis added).

34
Aristotle recognized such a distinction: see his
Metaphysics
, Bk Δ,
Ch. 17
.

35
Cf. 6.12 and 6.124.

36
Cf. 5.61.

37
Sullivan (
1996
), pp. 197–198. See in particular the wonderful n. 9 (p. 213) in which he makes clear how the organization of the 6s reflects topics that are of central concern to Kant. See Sullivan (
1996
) more generally, and Sullivan (
2002
), esp. §3.3, for his detailed defence of this reading.

Note that, although Wittgenstein aligns himself with Frege in rejecting the view that arithmetic consists of synthetic
a priori
truths, his conception of arithmetic is nevertheless very different from Frege’s. While conceding that mathematics is ‘a logical method’ (6.2), he regards mathematical propositions as strictly nonsensical, hence not even as senseless (6.2 again; and cf. 4.241ff. and 5.53ff.; see also Potter (
2000
),
Ch. 6
). This is a useful corrective, very pertinent to what will come later, to the idea that Wittgenstein sees ‘nonsensical’ as a term of unmitigated opprobrium. In
this
case, at least, Wittgenstein is well aware of the use to which nonsense can be put: see 6.211. (Note that I part company in some of these comments with Conant and Diamond (
2004
), n. 59, where they deny that Wittgenstein regards mathematical propositions as nonsensical; but I am not sure what their alternative is.)

38
But see McGinn (
2006
), pp. 269–270, for a somewhat different reading of this.

39
Cf. Geach (
1976
).

40
We shall return to the idea of what can be shown in the next section.

41
Cf. n. 11.

42
Cf. the parenthetical sentence in 3.328, the last sentence of 4.002, and the first sentence of 5.5563.

43
I say it ‘may’ suffer from an analogue of the Fregean problem because, even where nonsense is concerned, ‘the outward form of the clothing is not designed to reveal the form of the body’ (4.002).

44
Cf. Peter Sullivan’s discussion of Wittgenstein’s associated problem with logical category distinctions in Sullivan (
2003
), pp. 217ff. For Wittgenstein, the logical category to which anything belongs is determined by the quantifier within whose range it lies. But this means that the very use of the word ‘anything’ in this formulation of the doctrine – ‘the logical category to which
anything
belongs’ – cannot have its intended generality. It is impossible to generalize about things of different logical categories (as indeed this very admonishment purports to do).

There is a related problem for Frege when he claims in Frege (
1997a
), p. 140/p. 18 in the original German, that ‘an object is
anything
that is not a function’ (emphasis added).

45
‘Almost’ every turn? What are the exceptions? This question is far from straightforward. For one thing the answer may be relative to the reader (cf. Conant and Diamond (
2004
), n. 102). But on any account of the matter the exceptions are liable to be few, isolated, and miscellaneous. Putative examples, to supplement the sentence from 6.371 already cited, are the remark about notation at 5.531, the parenthetical comment about the ways of seeing the diagram in 5.5423, the reference to Frege in the first sentence of 6.232, the historical remark in the first sentence of 6.372, and the claims about philosophical methodology in 6.53. (Note: it is not uncontroversial that there
are
any exceptions. For an argument that there are not, see Morris (
2008
), pp. 343ff.)

46
I use ‘ineffable’ to mean ‘incapable of being expressed in words’. This leaves open the possibility of something ineffable’s being expressed in some other way, say in actions or in music. We shall see the significance of this possibility in the next section.

47
This is P.M.S. Hacker’s term: see Hacker (
1986
), p. 18.

48
For examples of something more or less traditional, see Anscombe (
1971
); Hacker (
1986
), Chs 1–4; and Pears (
1987
), Pt II.

49
And it makes Wittgenstein’s project in the
Tractatus
not only much more like Kant’s in the ‘Transcendental Dialectic’, but also much more like his own in his later work: see esp. §2 of the next chapter. See also the essays in Part II of Crary and Read (
2000
), entitled ‘The
Tractatus
as Forerunner of Wittgenstein’s Later Writings’.

50
For examples of something more or less new, see Conant (
1989
); Diamond (
1991e
); and Kremer (
2001
). For a powerful recoil in favour of a traditional reading, see Hacker (
2000
). Poised somewhere in between are Reid (
1998
); McGinn (
1999
); Proops (
2001
); Sullivan (
2002
); Morris (
2008
),
Ch. 7
, §E; and Morris and Dodd (
2009
). Marie McGinn develops her ideas at much greater length in McGinn (
2006
). For a quite different reading, see Moyal-Sharrock (
2007
). For an excellent discussion, see Costello (
2004
).

51
What follows summarizes the argument of Moore (2003b), from which I have borrowed some material.

52
Ramsey did direct this quip at the
Tractatus
, but there is an issue about whether he himself (as opposed to the countless people who have subsequently appropriated his quip) intended to direct it specifically at the idea that there are things that cannot be put into words; see Diamond (
2011
).

53
Cf. Conant (
1991
), p. 159; (
2000
), p. 198; and Diamond (
2000
), pp. 150–151.

54
See further §8.

55
I should emphasize that I offer this as much in a spirit of reconstruction as by way of serious exegesis of the exegesis. Indeed, I think that there is need for a corrective on both sides of the debate, precisely because of a shared tendency to overlook the possibility that there are things other than truths that are non-trivially ineffable: see e.g. Conant (
1991
), p. 160; (
2004
), p. 171; and Hacker (
2000
), p. 368. (Among those who show awareness of this possibility are Michael Kremer, in Kremer (
2001
), §IX, and Peter Sullivan, in Sullivan (
2002
), §2.4.)

Note: if the idea of ineffable understanding is to be divorced from the idea of ineffable truth, then clearly there must be understanding which is not understanding of any truth. Here there is an echo of a suggestion that first surfaced in
Ch. 2
, §6, that there is ineffable knowledge which is not knowledge of any truth but rather practical knowledge. It is a particularly loud echo if, as I believe, ineffable states of understanding are themselves states of practical knowledge; see Moore (
1997a
), pp. 161 and 183ff.