A Resolute Later Wittgenstein

7/27/2019 A Resolute Later Wittgenstein

1/20

A RESOLUTE LATER WITTGENSTEIN?

GENIA SCHO NBAUMSFELD

Abstract: Resolute readings initially started life as a radical new approach toWittgensteins early philosophy, but are now starting to take root as a way of interpreting the later writings as wella trend exemplied by Stephen MulhallsWittgensteins Private Language (2007) as well as by Phil Hutchinsons Whatsthe Point of Elucidation? (2007) and Rom Harre s Grammatical Therapy andthe Third Wittgenstein (2008). The present article shows that there are neithergood philosophical nor compelling exegetical grounds for accepting a resolutereading of the later Wittgensteins work. It is possible to make sense of Wittgensteins philosophical method without either ascribing to him an incoher-ent conception of substantial nonsense or espousing the resolute readerspreferred option of nonsense austerity. If the interpretation here is correct, itallows us to recognize Wittgensteins radical break with the philosophicaltradition without having to characterize his achievements in purely therapeuticfashion.

Keywords: Wittgenstein, resolute reading, therapy, nonsense, philosophical method.

An ever-widening rift divides the world of Wittgenstein studies. Locatedon one side of the debate are the self-declared resolute readers, whocleave to some version of what James Conant (2007) calls Mono-Wittgensteiniansimthe idea that, roughly speaking, the early and thelater Wittgenstein were up to the same thing: namely, offering a therapythat will cure us of the illusion of meaning something where we reallymean nothing. 1 Located on the other side are what the resolute like to callthe standard readers, 2 who believe, rst, that although there is somecontinuity in places, there is signicant discontinuity between the earlyWittgenstein and his later self, and, second, that the later Wittgensteinaimed at more than mere therapy. Some important recent work publishedin this journal (Glock 2004b; Hutchinson 2007; Harre 2008) contributesto this discussion: Hans-Johann Glock argues for a traditional con-ception of Wittgenstein as a broadly analytic philosopher, while PhilHutchinson is an avid proponent of a variant of resolution that one

1

The term resolute reading was coined by Warren Goldfarb (1997).2 Hutchinson (2007, 693) calls them elucidatory readers.

r 2010 The AuthorMetaphilosophy r 2010 Metaphilosophy LLC and Blackwell Publishing LtdPublished by Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK, and350 Main Street, Malden, MA 02148, USAMETAPHILOSOPHYVol. 41, No. 5, October 20100026-1068

r 2010 The AuthorMetaphilosophy r 2010 Metaphilosophy LLC and Blackwell Publishing Ltd


2/20

might, for obvious reasons, want to call therapism; Rom Harre mightbe characterized as an advocate of mild therapism.

Resolute readings initially started life as a radical new approach toWittgensteins early work: rst presented by Cora Diamond and JamesConant, they gained currency as an attempt to save the Tractatus fromending in self-contradiction. 3 But the debate has not remained Tractatus -centred. As Conant points out, issues parallel to those which arise in theinterpretation of the Tractatus arise in connection with the interpretationof Wittgensteins later work as well (2004, 168). Stephen Mulhall inWittgensteins Private Language (2007) concurstaking his cue fromConant (2004), Mulhall offers the rst sustained attempt at providing aresolute reading of Wittgensteins Philosophical Investigations .4

Mulhall identies two features necessary for a reading to count asresolute: commitment to nonsense monismthe contention that fromthe point of view of logic there is only one kind of nonsense, that is, plaingibberish (2007, 2)and rejection of the idea that there is something wecannot do in philosophy (2007, 8). While I agree with resolute readersthat the limits of sense are not limitations fencing us off from anything inparticular, I contest their claim that acceptance of this fact requiresendorsement of nonsense monism on pain of falling prey to an alternativeconception of substantial nonsensethat is, the notion of pseudo-propositions that are determinately unintelligible or that specify athought that we cannot think (2007, 8). Such a conception is, indeed,

incoherentonly, nonsense monism doesnt follow from a rejection of it.More than the two interpretative options offered by resolute readers resolution or some kind of commitment to substantial nonsense are available here: as the present article will show, neither a resolutenor a substantial reading can in fact do justice to the complexities of Wittgensteins text. I shall argue, contra Mulhall and Conant, that theauthor of the Investigations does allow for more than one kind of nonsense, and, furthermore, that recognition of this fact does not, of itself, push one into substantiality.

Given that Mulhall has, to date, developed the most comprehensive

account of a resolute later Wittgenstein, his interpretation is the focalpoint of my discussion, but much of my critique is also aimed at Conantand Hutchinson. In the rst part of this article I critically examineMulhalls reasons for presenting a resolute reading and show that thereare no good grounds for accepting either side of the resolute readersinterpretative dichotomy; in the second part, I undermine the claim thatthe later Wittgenstein thought that all nonsense is plain gibberish. Finally,

3 See Conant 2000 and 2002, and Conant and Diamond 2004.4 Others sympathetic to, or actively endorsing, resolute readings include Stanley Cavell,

Alice Crary, Burton Dreben, David Finkelstein, Juliet Floyd, Phil Hutchinson, OskariKuusela, Matthew Ostrow, Rupert Read, Martin Stone, and Edward Witherspoon.


GENIA SCHO NBAUMSFELD650


3/20

in the third part, I discuss the implications of the present debate for anunderstanding of Wittgensteins philosophical method.

1

Mulhalls interpretative challenge in Wittgensteins Private Language is tomotivate the idea that in the Investigations , too, and not just in theTractatus , it is possible to distinguish between substantial and reso-lute readings. Prima facie this is not an easy task, as the laterWittgenstein does not present his reader with a Tractatus -type exegeticalconundrum: the Investigations does not declare itself, like the Tractatus ,to be nonsensical. But if it does not, what are the merits of readingWittgensteins later work in resolute fashion?

Mulhall takes his interpretative cues for promoting a resolute read-ing of the Investigations from 374The great difculty here is not torepresent the matter as if there were something one couldnt doand 500When a sentence is called senseless, it is not as it were its sensethat is senseless. But a combination of words is being excluded from thelanguage, withdrawn from circulation (Mulhall 2007, 9). If we dontheed Wittgensteins warnings here, while at the same time regarding hisnotion of grammar as a way of recalling us to the distinction betweensense and nonsense, argues Mulhall (2007, 9), then we might be temptedto give this a substantial spin; we might end up regarding grammar, like

logical syntax in the Tractatus , as prohibitive and as preventing us fromarticulating something that is, nevertheless, in some sense, perfectlyintelligible. So we might be seduced into thinking that we can getintimations of what lies beyond the limits that grammar has demarcated.

But this problem may well strike one as spurious: it clearly betrays amisunderstanding to confuse the limits of sense with limitations orexclusion from a speciable domainone might just as well regardgrammar in the ordinary (linguistic) sense as imposing limitations onones expressive capacities! Nor does Mulhall make it clear why arealization of the fact that there is indeed nothingno thing that lies

beyond these limits should be incompatible with appreciating grammarsprohibitive role. The rules of chess, for example, precisely in virtue of allowing certain moves, prohibit others. But it would be confused to glossthis as, say, the rules of chess limiting my ability to play chess, sincewithout the rules, there would be no such thing as chess in the rstplace.

Perhaps it is Mulhalls resolute conception of nonsense that ispreventing him from appreciating this point. For, according to resolutereaders, nonsense is not the result of violating established criteria for theuse of words; rather, nonsense occurs because we have not yet establishedcriteria for the use of the offending expression. As Mulhall puts it inInheritance and Originality (2001), which he regards as a prequel to


A RESOLUTE LATER WITTGENSTEIN? 651


4/20

Wittgensteins Private Language (2007, 13): A meaningless [nonsensical]utterance is one to which no sense has yet been given, not one whichcannot have meaning because of the sense that it (or its component terms)have already been given (2001, 6869). But if so, then it would seem tofollow that grammar cant rule anything out, since that would allegedlyinvolve ruling out something in particular , and this is not possible, giventhat, according to resolute readers, no sense has yet been assigned to thestring of words in question. Mulhall uses the following example toillustrate this idea:

Michael Dummett has offered Chairman Mao is rare as a piece of substantial nonsense, because he claims it attempts to conjoin a proper name(which can take only rst-level functions as arguments) with a second-level

function (which can take only rst-level functions as arguments). But if it isessential to a symbols being a proper name that it take rst-level functions asarguments, then we can treat Chairman Mao as a proper name in thiscontext only if we treat is rare as a rst-level function rather than a second-level function (say, as meaning tender or sensitive). And by the sametoken, if it is essential to a symbols being a second-level function that it takerst-level functions as arguments, then we can treat is rare as a second-levelfunction in this context only if we treat Chairman Mao as a rst-levelfunction rather than a proper name (perhaps on the model of a brutalpolitician). Either way of parsing the signs is perfectly feasiblewe need onlyto determine a suitable meaning for the complementary component in each

case; but each way presupposes an interpretation of the string as a whole whichexcludes the other. So treating it as substantial nonsense involves hoveringbetween two feasible but incompatible ways of treating the string, without eversettling on either. (2007, 34) 5

This argument relies on a counter-intuitive suppressed premise: thethought that it is only possible to identify the meaning of a sub-propositional expression if this occurs within a sentence that has a sense.Not only is this principle extremely implausible in its own right, there isalso no evidence at all that the author of the Philosophical Investigationswould have accepted it. Consider, for example, the following passage:

What does it mean to say that the is in The rose is red has a differentmeaning from the is in twice two is four? If it is answered that it meansthat different rules are valid for these two words, we can say that we have onlyone word here.And if I am now attending to the grammatical rules, thenthese just do allow the use of the word is in both connexions.But the rulewhich shows that the word is has different meanings in these sentences is theone allowing us to replace the word is in the second sentence by the sign of equality and prohibiting this substitution in the rst sentence. (Wittgenstein1992, 558; translation emended)

5 For similar arguments see Conant 2005, Diamond 2005, and Witherspoon 2000.




5/20

This section makes it quite clear that Wittgenstein thinks that there aregrammatical rules which determine the correct and the incorrect applica-tions of words. That is to say, Wittgenstein thinks that grammar prohibitsthe formation of the construction the rose equals (or is equivalent to)red. This contradicts Mulhalls contention that nonsense cannot be theresult of attempting to conjoin intelligible words in unintelligible ways(2007, 9). We can combine intelligible words in unintelligible ways, butthe result of this wont be a proposition that is determinately unin-telligible, since there is no such thing, but a meaningless string of words.

But from the fact that there is no such thing as a determinatelyunintelligible proposition, it does not follow, as resolute readers sup-pose, that we cannot identify the words that make up the meaninglessstring. We can see, for instance, that in the expression the rose equalsred the word equals is being misused, just as we can see that inDummetts example Chairman Mao and the expression is rare arebeing misused, without having to attribute a meaningsome sort of nonsensical senseto the expression as a whole (since it has none). Allthat is required for identication to occur is to know the meanings of the uses ofthese expressions in ordinary contexts. 6 Once I havemastered the rules for the uses of these expressions, I can tell straightaway that Chairman Mao is rare makes no sense, just as I can seestraight off, if I have mastered chess, that moving the rook diagonallyacross the board is not, cheating aside, a possible move in this game.

In other words, once the rules for the use of the expressions ChairmanMao and is rare are in place, it follows that the combination Chair-man Mao is rare can make no sense. That is to say, it is precisely becauseof the kinds of meanings (uses) that Chairman Mao and is rare havein other, normal, contexts, that the construction Chairman Mao israre is nonsensical. Hence, pace Mulhall, the grammatical rules doconstitute a given, impersonal source of authority (2007, 66), as oncethey are in place, certain formations will be ruled out as inadmissible inadvance on pain of unintelligibility. 7

Hutchinson (2007) runs an even more radical line than Mulhall: not

only does a word not have a meaning outside a particular context of use,it (whatever it is) doesnt even qualify as a word unless it is used.Hutchinson claims: There is no such thing as a word outside someparticular use. . . . For a word to be is for a word to be used . Languagedoes not exist external to its use by us in the world. As I have alreadynoted, language cannot, in McDowells phrase, be viewed from sideways

6 See also Hacker 2003, Glock 2004a, and Scho nbaumsfeld 2007.7 And it is important to note here, as Hacker (2003, 1920) says, that if one were to

assign to a signicant word or phrase a meaning in contexts from which it is excludedasin assign to is rare in Chairman Mao is rare, as Mulhall suggests, the meaning of

tender or sensitivethen one would have changed its meaning. So one would, asWittgenstein noted, be talking of something else.




6/20

on, in the sense in which we cannot stand outside language in order thatwe might talk about language (2007, 706). This is very confused. Indeed,we cannot use language to get outside language tout court , in the sense,perhaps, in which, if we are not resolute readers, Wittgenstein himself attempted to get outside language in the Tractatus , and which is alsothe sense McDowell (who is being misused by Hutchinson here) has inmind, that is, by trying to adopt some sort of transcendental perspec-tive on language and the world. But this does not imply, as Hutchinsonmistakenly seems to assume, that we cannot use language to say some-thing about how language functions. That is to say, it is an error to believe,as Hutchinson does, that if we reject such a transcendental perspective,then it just follows that there is no such thing as a word outside someparticular use. For to think that there is such a thing as a word outside particular contexts of use is not in the least the same as thinking there issuch a thing as a word outside all contexts of use, and only the latter wouldqualify as adopting an external perspective on language. Hence, at best,all that Hutchinson has shown is that there is no such thing as a wordoutside languageoutside all contexts of usebut no reader of Wittgen-stein, resolute or otherwise, would disagree with that .

It is surprising that resolute readers are so sanguine about acceptingthe notion that words by themselves have no meaning, for, among otherthings, this conception has the effect of turning native speakers intoradical interpreters. That is to say, on the resolute conception, I can never

understand a sentence straight off (or recognize its misuse), almost as if Ididnt have a command of my own language but were like an explorergoing into a foreign land trying to interpret the signs of the natives. Of course in different contexts the same words may mean different things (or,indeed, nothing at all), but, as a rule of thumb, if I know what x , y,and plus mean, I shall know what x plus y means (and if it meansnothing, like the rose plus red, I shall know this too). So, contra theresolute readers, I dont rst need to interrogate (Conant and Diamond2004, 64) the sentence to see whether it makes sense or not.

Naturally, it is quite a lot harder to spot nonsensical utterances in

philosophy, and it is therefore the role of Wittgensteins grammaticalremarks to remind us of what we happen, in a particular context, to havelost sight of. For example, Wittgenstein says: It makes sense to say aboutother people that they doubt whether I am in pain; but not to say it aboutmyself (1992, 246). In other words, although we all, as competentlanguage users, intuitively feel that there is something wrong with sayingI doubt whether I am in pain, it may take a grammatical investigationto reveal why this is so. And this, pace Mulhall (2007, 59), implies thatthere is a perfectly intelligible way in which one can convey grammaticalinformation to someone who already possesses it.

Let us now look at the Wittgenstein passages immediately precedingthe one Mulhall quotes in support of his interpretation, in order to




7/20

ascertain whether anything that Wittgenstein says there might lendcredence to a resolute reading of Philosophical Investigations 500.In 499 we read: To say This combination of words makes no senseexcludes it from the sphere of language and thereby bounds the domain of language. But when one draws a boundary it may be for various kinds of reason. Prima facie, far from supporting Mulhalls reading, theseremarks sound like a direct contradiction of it. Conant is aware of thisproblem:

Certain passages in the later work, however, in which Wittgenstein speaks, forexample, of excluding certain combinations of words from our language,might seem to contradict this [a resolute interpretation], thus inviting a readingof Wittgenstein along the following lines: certain combinations of words are to

be identied as impermissible on the grounds that these combinations violatethe principles governing which combinations of words are grammatically well-formed. It is precisely such a reading of his work that a resolute reader of Wittgenstein will hear him seeking to fend off in 500 of PI: When a sentenceis called senseless, it is not as it were its sense that is senseless. (2004, 186). 8

But this, surely, is much too quick. For, rst, if this were in fact thereading that Wittgenstein were rejecting, then why would he repeat , in thevery same passage in which he is, according to Conant, doing the putativefending off, the selfsame sentence that Conant claims might be contra-dicting a resolute interpretation? 9 In other words, if Conant were right,why would Wittgenstein reiterate, in 500, that a combination of words isbeing excluded from the language, withdrawn from circulation ? Even inthe most charitable of moods, this would hardly make much exegeticalsense. Second, the offending sentence is also to be found in 499, andgiven that this directly precedes 500, where, as already noted, it occursagain, we have two very good reasons for thinking that Wittgenstein is notrepudiating the view Conant says he is: the heuristic reason thatWittgensteins remarks follow on naturally from each other, the laterremarks commenting, and expanding, on the previous ones, as well as theunignorable fact of the repetition itself.

We can now turn the tables on a resolute reading of 500. Giventhat Wittgenstein, in 497, speaks again of rules of grammar 10 and, in 499, of grammar bounding the domain of language, it seems that 500 is warning us precisely against a resolute interpretation of theimplications of this. In other words, what Wittgenstein is fending off at

8 See also Conant (1998, 246).9 And the statement when a sentence is called senseless, it is not as it were its sense that

is senseless taken by itself is not sufcient to clinch a resolute reading, since it is perfectlycompatible with the view Conant claims Wittgenstein is trying to head off. See the nextparagraph.

10

The rules of grammar may be called arbitrary, if that is to mean that the aim of thegrammar is nothing but that of the language.




8/20

500 is not the view that, as Conant puts it, certain combinations of words are to be identied as impermissible on the grounds that thesecombinations violate the principles governing which combinations of words are grammatically well-formedsince, as 497 and 499, interalia, clearly show, this is the conception Wittgenstein is endorsingbutrather, to put it anachronistically, the resolute readers idea that accep-tance of this conception (that there is such a thing as infringing the rulesof grammar) results in the production of sentences with senselesssenses. So Wittgensteins own view cannot, pace Conant and Mulhall,be usefully characterized as either resolute or substantial, for, if I amright, he is propounding a conception of grammar that resolute readerswould reject, while, at the same time, warning us not to construe this asimplying commitment to some form of substantial nonsense. Hence, wehave, so far, been given no good reason for coming down on either side of the interpretative dichotomy that resolute readers confront us with: asubstantial reading of the relevant passages of the Investigations is justas much to be rejected as a resolute one. There are more than just thetwo alternatives offered by resolute readers here.

More promising might be the thought that substantial readings of the Investigations involve, as Mulhall suggests, attributing to Wittgensteinan implicit philosophical theory of the (now grammatical) conditions of sensequite as if our everyday abilities to distinguish sense fromnonsense require at the very least a philosophical grounding or founda-

tion (perhaps a criterial semantics, or a theory of language-games, or ananthropology of the human form of life) (2007, 9). There certainly arephilosophers who ascribe just such views to Wittgenstein and who arepuzzled by the very idea of not advancing theses in philosophy. 11 But notemulating them does not require resolution either, as one can acceptthe conception of grammar outlined above without being forced, in virtueof this, to ascribe theories to Wittgenstein. 12 In fact, I take it that this isthe kind of reading that Hacker has offeredConants wilful misreadingof him as substantial notwithstanding (2004, 16869).

The problems so far considered are compounded by the fact that the

meaning of substantial itself seems to shift throughout Mulhalls text.That is to say, as we read on beyond the introduction, it becomes clearthat a substantial reading of the Investigations is not just a conception

11 These are what Hutchinson calls the doctrinal readers (2007, 693).12 By a theory or theoretical explanation, Wittgenstein means the hypothetical construc-

tions of natural science: It was true to say that our considerations must not be scienticones. . . . And we may not advance any kind of theory. There must not be anythinghypothetical in our considerations. We must do away with all explanation , and descriptionalone must take its place (1992, 109). Here the relevant contrast is between explanationconceived scientically (hypothetically) and explanation by description. That is to say, I

think it is obvious from the context that Wittgenstein is not rejecting all possible forms of explanation (because a description could also be an explanation).




9/20

that has succumbed to some form of substantial nonsense, or thatattributes theoretical concerns to Wittgensteinno, it seems to compriseany reading (including Mulhalls own from pre-resolute times!) thatregards Wittgenstein as advancing a non-empty view, or some form of argument. For example, any reading of Wittgensteins famous privatediarist passage ( 258) that attempts to show that private ostensivedenition is logically impossible is characterized as substantial byMulhall. For, according to a resolute reading, this passage does notshow us that something (noting the occurrence of our pains) is logicallyimpossible in the kind of context the diarist stipulates; it shows us that,although we thought that we were imagining someone noting theoccurrence of his sensations, we were not imagining anything. . . . Witt-gensteins aim is to get us to see that the tale of the diarist amounts onlyto someone making a mark, or uttering an inarticulate sound (Mulhall2007, 99). But, surely, if Wittgenstein is right, and private ostensivedenition is logically impossible, then it just follows from that that all theprivate diarist is doing is, in the end, nothing in particular. So, thesubstantial readers conclusion doesnt, in this respect, differ from theresolute one. But, if so, why prefer resolution?

Mulhall anticipates this type of objection and responds that if onesinterlocutor is convinced that her empty words articulate an insight, thensimply to oppose or dismiss them (by directly invoking a grammaticalarticulation that they appear to violate) would fail to acknowledge the

fact that she will necessarily respond to that invocation from within herconviction (2007, 83). Such a response would, in other words, fail to takethe interlocutor seriously as a subject. This is not convincing, however,for no serious (Wittgensteinian) philosopher would go about attemptingto defuse the notion of a private language by starting with the thoughtthat this is a piece of nonsense. Rather, it needs to be shown that it is that is, it needs to be shown by giving arguments (reasons) why, if it does,this form of words violates a grammatical articulationand so justquoting chapter and verse of grammar can never be sufcient. Butthen, despite his now attributing substantiality to his former self

(Mulhall 2007, 96), neither would the author of On Being in the World (1990) have thought it sufcient. So, again, there seems no good reasonhere to prefer a resolute reading.

2

Whatever the merits of attributing a resolute conception of nonsense tothe Tractatus ,13 nowhere, in the Investigations , does Wittgenstein say thathe believes that there is no difference between philosophical nonsense and

13

This is a question I shall not be addressing in this article. For a critique of the view see,for example, Hacker 2000, Proops 2001, Williams 2004, and Scho nbaumsfeld 2007.




10/20

plain gibberish. 14 It is consequently not surprising that Conant (2004,186), for example, has to cite as evidence a passage from Wittgensteinsunpublished 1935 Lecture on Personal Experience:

Different kinds of nonsense . Though it is nonsense to say I feel his pain, this isdifferent from inserting into an English sentence a meaningless word, sayabracadabra (compare Moore last year on Scott kept a runcible atAbbotsford) and from saying a string of nonsense words. Every word inthis sentence is English, and we shall be inclined to say that the sentence has ameaning. The sentence with the nonsense word or the string of nonsense wordscan be discarded from our language, but if we discard from our language Ifeel Smiths toothache that is quite different. The second seems nonsense, weare tempted to say, because of some truth about the nature of things or thenature of the world. We have discovered in some way that pains and

personality do not t together in such a way that I can feel his pain.Thetask will be to show that there is in fact no difference between these two casesof nonsense, though there is a psychological distinction, in that we are inclinedto say the one and be puzzled by it and not the other. We constantly hoverbetween regarding it as sense and regarding it as nonsense, and hence thetrouble arises.

Conant takes this passage as unambiguously supporting his case, andrelates it to his reading of the private language sections of theInvestigations in the following way. The standard interpretation of theseremarks, Conant claims, is to show that the very idea of a privatelanguage is the idea of something which we can rule out because of thekind of sense that the locution private language has antecedently beengiven. It takes it that there is something determinate which the philoso-pher wants to mean by the locution private language and that that isnonsense (2004, 187). But Wittgensteins point, according to Conant, isto show that such a conception makes no sense, and, hence, that what wetook to be an instance of substantial nonsensea particular somethingthat cannot becollapses into mere nonsense (plain gibberish).

If we apply these insights to the 1935 passage, the following pictureemerges. Wittgenstein wants to disabuse the philosopher of the view thatthe locution I feel Smiths toothache has a sensethere is somethingdeterminate the philosopher wants to mean by itand it is this sensethat turns out to be nonsense. But, according to Conants reading of Wittgenstein, there is at best a psychological distinction between I feelSmiths toothache and abracadabra.

It is not so clear, however, that the remarks in question really dosupport Conants reading. For the point of Wittgensteins comments isnot to show that what the philosopher took to be substantial nonsense is

14 Even the resolute readers oft-quoted remarkMy aim is: to teach you to pass froma piece of disguised nonsense to something that is patent nonsense (Wittgenstein 1992,

464)is inconclusive, as patent nonsense need not necessarily be the same as plaingibberish.




11/20

really plain nonsensethe trouble arises, Wittgenstein says, because weconstantly hover between taking the locution in question as sense orregarding it as nonsense, not , pace Conant, between regarding it assubstantial nonsense or plain gibberishrather, what Wittgenstein wantsto get the philosopher to see is that we exclude from the language I feelSmiths toothache for conceptual or grammatical reasons, and not, asthe philosopher imagines, for metaphysical ones. In other words, it is not,as Wittgenstein says, because of some truth about the nature of things orthe nature of the world that the expression I feel Smiths toothache isnonsensesay, because we have discovered that pains and personalitydo not t together in such a way that I can feel his pain. The point,therefore, of saying there is no difference between saying abracadabraand I feel Smiths toothache is not to show that the second stringcontains nonsense words, as Conant thinks, but rather to prevent thephilosopher from supposing that the reason why we discard the latterfrom the language is to exclude an impossible possibilitynamely, thepossible state of affairs of my feeling Smiths toothache. The very ideaof my feeling his toothache is senseless, however, for, if I could, as it were,feel it, then this would eo ipso make it my toothache, not Smiths (and tosay this is to make a grammatical remark). Consequently, the signicanceof this passage consists in weaning the philosopher away from the ideathat a rule of grammar functions like a metaphysical prohibition.

Naturally, once the philosopher has been brought to realize that I feel

Smiths toothache is nonsense, that is, that it can be discarded from thelanguage because it is not a move in the game, the same thing followsfrom it as from abracadabrato wit, nothing. 15 In this respect, there isindeed no difference between the two strings. But this does not imply, asresolute readers seem to maintain, that therefore the two strings are thesame in every other way too . For nothing follows from a tautology such as p or not p either, but this fact does not turn it into gibberish (it is, atbest, and, as the author of the Tractatus thought, senseless, that is to say,it asserts nothing).

In other words, the reason why Wittgenstein says that there is in fact

no difference between these two cases of nonsense is in order toemphasize that nothing follows from my not being able to feel Smithstoothachea form of words is withdrawn from circulation, that is all and this, precisely, in order to head off the metaphysical reading that istempted to construe a strings being nonsense as a kind of super-falsehood: somethings being impossible because of some truth aboutthe nature of things. And if one were to construe nonsense thus, thensomething would follow from an expressions being nonsensicalnamely,the necessary truth of its negation. It is this that Wittgensteins remarksare supposed to alert the reader to, and not, contra Conant, that I feel

15 Except if we are magicians and are signalling that the conjuring trick has taken place.




12/20

Smiths toothache and abracadabra are logically indistinguishable something already signalled by the fact that Wittgenstein italicizesdifferent kinds of nonsense at the beginning of the passage and says,Though it is nonsense to say I feel his pain, this is different frominserting into an English sentence a meaningless word.

Of course Conant would continue to maintain that I feel Smithstoothache differs only psychologically from abracadabra. But hereone may legitimately wonder what this distinctive psychological kind of nonsense (Mulhall (2007, 5) is supposed to be that resolute readersappeal to. Given that, as Mulhall warns, it is vital to note that whatmakes it [the nonsense] illuminating is not anything about the nonsenseitselfnothing intrinsic to it, as it weresince logically speaking it has nointrinsic structure (2007, 5), it remains mysterious how, exactly, thestring manages to bring it about that one is philosophically tempted by it.If this is not, in any way, a function of the strings content or logicalstructure, it would seem to have to accomplish this feat by pure magic, forno one is so much as tempted by piggle wiggle or abracadabra.

It is at this point in the discussion that resolute readers tend to startspeaking of imaginatively inhabiting the interlocutors perspective(Mulhall 2007, 82). But this just raises the same question again in differentguise: What, if not content, is it that constrains our imaginative acts of identication? If it is true, as resolute readers seem to suppose, that onecannot even identify the meaning of a word unless it occurs within the

context of a sentence that has a sense, then no constraints at all seemavailable here. And if so, then even wheeling in a Tractarian sign/symboldistinction will not help, for if all I am taken in by is mere signs (notsymbols)signs not parsed in any particular waythen I can conceiv-ably take them to mean anything I please, and hence, it will again bedown to pure chance whether what I take them to mean bears anyrelation to the confusion Wittgenstein wants to dispel.

Even if we grant resolute readers that what we are taken in by is thefact that the nonsensical string jingles like or supercially resemblesa genuine sentence, the question arises, what, precisely, it is that makes

the nonsensical string look or sound like a meaningful one in the rstplace. Can a resolute reader really help himself to a notion of resemblancewithout having to grant, at the same time, that the relevant string isactually composed of words , words that generally have a meaning (even if in this particular context they have none)? For in order for a notion of resemblance to get off the ground, the nonsensical string must havesomething in common with a genuine sentence. And what might this be, if not the fact that the string is composed of words we can recognize?

Piggle wiggle does not resemble any kind of sensical linguisticconstruction, and therefore we cant operate with it. But nonsensicalstrings of the relevant kind cannot be like that if we are, as resolutereaders must maintain, to be taken in by them. So the nonsensical




13/20

string must dupe us by sounding and looking like real words in agrammatically well-formed sentence. But if the words in the mean-ingless string sound and look like real wordsand are not just plaingibberish like piggle wigglethen what is to stop us from saying they arereal words employed in ways contrary to the rules for their correct use?After all, one might say that if something waddles like a duck and quackslike a duck, then it most probably is a duck. And if the only thingpreventing resolute readers from granting this is the fact that the string asa whole is nonsense, then this is simply to beg the question.

Not only have resolute readers sawn off the branch on which they weresitting, Wittgenstein himself is content to allow that a partial under-standing of a question or sentence is possible, even if it should turn out,on further reection, that no clear sense can, in the end, be assigned to it.Consider, for example, the following remark: It seems clear that weunderstand the meaning of the question: Does the sequence 777 occur inthe development of p ? It is an English sentence; it can be shown what itmeans for 415 to occur in the development of p and similar things. Well,as far as such explanations reach, so far reaches, one can say, onesunderstanding of this question (Wittgenstein 1992, 516; translationemended). In other words, far from claiming that the question Does thesequence 777 occur in the development of p ? is no different from sayingabracadabra, Wittgenstein allows, pace the resolute readers, that thisquestion is intelligible to the extent that one can, precisely, recognize the

words sequence, p , development, and so on, in it; that it soundssimilar to the question Does 415 occur in the development of p ? and soonall of which is necessary, if one is, in the end, to come to understand(should one be persuaded by Wittgensteins argument) why the originalquestion cannot be given a clear sense after all (cf. also Glock 2004a, 242).

So, one cannot help feeling that, when all is said and done, resolutereaders cannot really mean what they say. 16 For they must either besmuggling some content into otherwise plain gibberishin which case,why should we accept that we are dealing with gibberish at all?or elsestop insisting that plain nonsense can in some unspeciable psycholo-

gical sense (which is neither a function of content nor logical relations)be understood, in which case, of course, there is also nothing further tosay. 17 But if, for all the world, it seems as though the sentences the

16 The following sentence, which I have no idea how to reconcile with the rest of whatgoes on in Mulhalls book, would seem to lend support to this idea: On his [Wittgensteins]taxonomy, not all nonsense is devoid of sense, or meaning, or point (2007, 122). How,exactly, is the thought that not all nonsense is devoid of sense compatible with a resoluteconception of nonsense?

17 This exactly parallels the problem resolute readers have as regards the Tractatus . Theyeither fall into self-contradiction by covertly smuggling content into apparently plain

nonsense, or leave the reader with no materials at all from which to construct a (Tractarian)ladder.




14/20

resolute readers regard as nonsense are composed of intelligible words,why should we believe them when they claim they arent?

3

In the light of all the objections so far considered, perhaps the onlyimaginable selling point of a resolute reading is the fact that it is genuinelydifcult to understand the role that reductio -type arguments play in theInvestigations . For, as Mulhall asks:

How could any reductio argument deliver a genuine conclusion, by revealingthe sheer nonsensicality of its apparent starting-point? And if reductio argu-ments really are legitimate means for gaining intellectual insight, then theposition in which they leave us is surely no more uncomfortable than thatoffered to us by the Tractatus , with its concluding claim that a criterion forunderstanding its author is the realization that every elucidatory word of hismust be recognized as simply nonsensical. . . . Perhaps, when our concern iswith the limits of sense, there is no other way of acknowledging them, and of inviting others to share that acknowledgement. And perhaps it is a criterion of properly understanding our difculties here that we apprehend the latentnonsensicality [my italics] of the previous sentence. . . . No doubt, our falsesense of understanding here was undergirded . . . by Wittgensteins use of theword beetle in articulating his putative analogy to a fantasy about painand painas opposed, say, to brillig, or $%&, or n . (2007, 137)

Although I agree with Mulhall that the position we are in may stand inneed of further clarication, I dont believe that a satisfactory response tothe problem can be just to write off as plain nonsense both whatWittgenstein says about pain and beetles in the Investigations and most of what we say about what Wittgenstein is up to in such passages.But I dont see how this unpalatable consequence can be avoided if thefantasy about pain is really no different, as Mulhall says, from $%&,for, if so, then our articulation of that fantasy will also turn out to be$%&a fact that Mulhall seems to be acknowledging when he saysthat we must apprehend the latent nonsensicality of the previous

sentence. Given that the fantasy about pain pervades quite a bit of Mulhalls text, however, it will not just be the previous sentence thatwill reveal itself to be sheer nonsense, but more signicant parts of thebook too. Consequently, we need to nd an alternative way of makingsense of Wittgensteins philosophical method.

Anthony Kenny believes that because Wittgenstein is committed to theview that the philosophers dogma is not a genuine proposition fromwhich other things might follow, but only a piece of nonsense in disguise,there can be no room for argumentation within Wittgensteins philosophi-cal method, since it is impossible to make a piece of nonsense a premise inan argument (2004, 175). Consequently, Kenny contends, Hacker, forexample, must be wrong to ascribe arguments to Wittgenstein, for, if




15/20

something is in fact nonsense, it cannot follow from an argument bydeductive inference. But if so, says Kenny, we may well be puzzled aboutwhat kind of following (2004, 175) Hacker is talking about when he says,for instance, that it follows from the private language argument thatsolipsism and idealism are misguided philosophies. That said, Kennyimmediately goes on to attach a proviso to this claim:

A reason why we might think that there is room for argument even inWittgensteinian therapy is that the philosophical treatment of a problemmay well involve the use of words like so, therefore and because whichare characteristic of genuine inference. But that is due to the demands of thetherapeutic procedure. The misguided philosopher believes that his dogma is agenuine proposition. To cure him of that illusion we have to humour him: wehave to take his pseudo-proposition seriously by treating it as if it was a

genuine proposition and drawing consequences from it. Of course theseconsequences will themselves be pseudo-propositions and only pseudo-con-sequences. . . . The therapeutic procedure is not, however, a mere incantation.It must obey the laws of logic. What follows from the pseudo-propositionmust be what would really follow from it if it were a genuine proposition.(2004, 175)

But now it seems that Kenny is hoist on his own petard. For he previouslycastigated Hacker for helping himself, when attributing arguments toWittgenstein, to a non-deductive notion of following, while himself appealing, in this passage, to what sounds suspiciously like the resolute

readers idea of apparent logical relationsrelations that wouldobtain if only the pseudo-proposition were a genuine proposition. And if something is wrong, as Kenny seems to think, with Hackers conceptionof following, then surely his own invocation of pseudo-following what would really follow from [a proposition] if it were a genuinepropositioncannot fare any better. For how, if we allegedly cannotmake sense of the idea of non-deductive following, are we to under-stand the logical relations supposedly at work in pseudo-following?So, if Kenny is right about Hacker, then his own account can hardly bethought to be immune to similar criticism.

Kennys reading therefore seems to parallel the dilemma faced byresolute readers in the previous section: on the one hand Wittgensteinstherapeutic procedure must not end up collapsing into mere incantation; onthe other, it seems impossible to avoid this consequence if nonsense cannotgure in an argument, or is literally plain gibberish. To avoid the rst horn,resolute readers appeal to a psychological conception of nonsense, whileKenny invokes the notion of pseudo-following, but neither of theseoptions is, for the reasons given in this article, very promising.

Interestingly, Wittgenstein in the Investigations doesnt actually speakvery often of a constructions being nonsensical. He tends, rather, tospeak of a picture lacking a clear application. This might help us get abetter sense of what he is up to when he tries to show why we should




16/20

withdraw a combination of words from circulation, or why a form of words has no clear use. For it is this that reductio arguments, within thecontext of Wittgensteins later philosophy, really come down to. Con-sider, for example, the following remark in 352 that occurs within thecontext of whether it makes sense to ascribe pains to a stove:

Here it happens that our thinking plays us a queer trick. We want, that is, toquote the law of excluded middle and to say Either such an image is in hismind, or it is not [either the stove has a pain or it does not]; there is no thirdpossibility.We encounter this queer argument also in other regions of philosophy. In the decimal expansion of p either the group 777 occurs, orit does notthere is no third possibility. That is to say: God seesbut wedont know. But what does that mean?We use a picture; the picture of avisible series which one person sees the whole of and another not. The law of

excluded middle says here: It must either look like this, or like that. So itreallyand this is a truismsays nothing at all, but gives us a picture. And theproblem ought now to be: does reality accord with the picture or not? And thispicture seems to determine what we have to do, what to look for, and how but it does not do so, just because we do not know how it is to be applied.

I would like to suggest that to say this combination of words makes nosense is analogous to saying this picture has no application. But apicture, even if senseless, obviously isnt gibberish. For it suggests, asWittgenstein says, a particular application to usand it can only do thisif it is not mere gobbledygookbut then, when we actually try to apply

the picture in this way, this turns out not to be possible. So, for example,one might think that in order to understand the question of whether astove has pains, it is sufcient simply to imagine that the stove has whatI have when I am in pain (either the stove has a pain or it does not).Wittgenstein might then invite his interlocutor to specify what havingamounts to in this context, and it would quickly emerge that thephilosopher construes the grammar of having a pain as functioninglike the grammar of being in possession of some kind of object, say, abeetle, albeit it one intrinsically inaccessible to anyone else. And so itperhaps makes sense to stipulate that the same private object is for ever

locked away inside the stove as it is locked away inside me. Thus the la-tent nonsense gradually becomes ever more patent ( 464), and Wittgen-stein has loosened the grip of the picture that held us captive ( 115) (inthis case the inner object picture of pain). This process of transforma-tion could be characterized as the attempt, as it were, to model somethingfrom the picture, in order to make us see that this cant be done ( 512):

It looks as if we could say: Word-language allows of senseless combinationsof words, but the language of imagining does not allow us to imagine anythingsenseless.Hence, too, the language of drawing doesnt allow of senselessdrawings? Suppose they were drawings from which bodies were supposed to be

modelled. In this case some drawings make sense, some not.What if Iimagine senseless combinations of words?




17/20

Following this analogy, I would suggest that philosophical nonsenseasenseless combination of words that, rst of all, seems to make senseis,in many ways, like a senseless drawing; a drawing by Escher, for example.And just as one could dispel a childs puzzlement about Eschers hands,for instance, if it was asked to construct a model from them, soWittgenstein dispels philosophical illusion by asking the philosopher toapplymodel a body fromthe picture he is tempted by, in order toshow him that no such thing is possible ( 517): The question arises:Cant we be mistaken in thinking that we understand a question? Formany mathematical proofs do lead us to say that we cannot imaginesomething which we believed we could imagine. (E.g. the construction of the heptagon.) They lead us to revise what counts as the domain of theimaginable.

In other words, what Wittgenstein seems to be saying here is that themathematical proof transforms our concept of imaginability; imposes, asit were, a conceptual limit on it. For once the proof has been found (thegrammatical rule articulated), nothing anyone says or does will count asan instance of imagining the construction of a heptagon (or, indeed, of aprivate language)just as someone who, in chess, claims to becastling with a bishop will not be deemed to have castled, even if shemoves her bishop horizontally across the chess board and swaps its placewith the kings.

If this account is correct, then Kenny and the resolute readers have, for

different reasons, got Wittgenstein wrong. For it is only if we want toreserve the word argument for showing a claim to be either true or falsethat we need balk, like Kenny, at the idea that Wittgenstein usesarguments to show how a particular combination of words does notmake sense. 18 Naturally, I cannot deductively prove that something isnonsense, as this is a patently incoherent ideaand, in this much, I can,indeed, not make a piece of nonsense a premise in an argument. What Ican do, however, is seek to make the nonsense patent by arguing againstthe philosophical preconceptions (false pictures) that attract the philoso-pher to the nonsensical sentence in the rst place. Once these are

undermined, the philosopher will himself be brought to see that his wordsonly seemed to add up to a genuine claim.Hutchinson therefore subjects Wittgenstein to considerable misreading

when he claims that Wittgenstein attempts to break the grip of a pictureby facilitating [his] interlocutors realization that other pictures areequally valid (2007, 694). For if a picture lacks a clear application, ithardly makes any sense to think that other pictures might be just as valid,

18 Even if this means having to modify, as Glock (2004a, 243) says, the standardassumption that everything that stands in logical relations with something meaningful is

itself meaningful. See also McManus 2006, 137. If this upsets our philosophical intuitions, tospeak with Wittgenstein, it is perhaps high time they were upset.




18/20

since the false picture is precisely not valid. Loosening its grip isconsequently a matter not of presenting other valid pictures but, as Ihave argued, of showing how, and why, the false picture has led us intoconfusion in the rst place. 19 For, quite apart from anything else, whywould anyone in the grip of a Fodorian conception of the mind, say, carethat there are other pictures of the mind around, unless it could beshown that these other pictures are superior?

Finally, it also betrays a misunderstanding to hold, as Hutchinson(2007, 695) does, that Wittgenstein thought that philosophical problemsare existential or personal problems. Although passionate about theethical dimension of philosophy, 20 Wittgenstein believed that philosophi-cal problems are problems to which anyone can be susceptible, since theylie in our language and language repeats them to us inexorably (1992, 115). Hence, philosophical problems are not personal in the sense inwhich nancial problems, for instance, are personal problems. For asWittgenstein once put it: Language has the same traps ready foreveryone; the immense network of easily trodden false paths [ dasungeheure Netz gut gangbarer Irrwege ]. And thus we see one person afteranother walking down the same paths and we already know where he willmake a turn. . . . Therefore, wherever false paths branch off I ought toput up signs to help in getting past the dangerous spots (2005, 312). 21 If resolute readers were right, and Wittgensteins philosophical method wereentirely patient-specic, what would be the point of putting up such

signs?

Department of PhilosophyUniversity of SouthamptonHigheld Southampton SO17 1BJ United Kingdom [email protected]

Acknowledgments

I would like to thank Denis McManus, Aaron Ridley, and DanielWhiting for comments on earlier drafts of this article.

19 Of course there are instances where Wittgenstein wants to change our perspective byshowing us another picture, but these will typically not be occasions where he is concernedto undermine the picture that held us captive. For a more detailed discussion of this seeHacker 2007, 1069.

20 For a development of this notion, see, for example, Scho nbaumsfeld 2007.21 I dont think this quotation can be dismissed just because it is from The Big Typescript ,

as it clearly tallies with much of what Wittgenstein says in the Investigations (especially 115). Furthermore, Conant also relies on material that dates roughly from the same period.


mailto:[email protected]:[email protected]


19/20

References

Ammereller, Erich, and Eugen Fischer, eds. 2004. Wittgenstein at Work .

London: Routledge.Conant, James. 1998. Wittgenstein on Meaning and Use. Philosophical

Investigations 21, no. 3:22250. . 2000. Elucidation and Nonsense in Frege and Early Wittgen-

stein. In Crary and Read 2000, 174217. . 2002. The Method of the Tractatus . In Erich Reck, ed., From

Frege to Wittgenstein , 374462. Oxford: Oxford University Press. . 2004. Why Worry About the Tractatus ? In Barry Stocker, ed.,

Post-Analytic Tractatus , 16792. Aldershot: Ashgate. . 2005. Two Conceptions of Die U berwindung der Metaphysik . In

T. G. McCarthy and S. C. Stidd, eds., Wittgenstein in America , 1361.Oxford: Oxford University Press. . 2007. Mild Mono-Wittgensteinianism. In Crary 2007, 31142.Conant, James, and Cora Diamond. 2004. On Reading the Tractatus

Resolutely: Reply to Meredith Williams and Peter Sullivan. In Ko lbeland Weiss 2004, 4699.

Crary, Alice, ed. 2007. Wittgenstein and the Moral Life: Essays in Honourof Cora Diamond . Cambridge, Mass.: MIT Press.

Crary, Alice, and Rupert Read, eds. 2000. The New Wittgenstein .London: Routledge.

Diamond, Cora. 2005. Logical Syntax in Wittgensteins Tractatus .Philosophical Quarterly 55:7889.

Glock, Hans-Johann. 2004a. All Kinds of Nonsense. In Ammerellerand Fischer 2004, 22143.

. 2004b. Was Wittgenstein an Analytic Philosopher? Metaphilo-sophy 35, no. 4:41944.

Goldfarb, Warren. 1997. Metaphysics and Nonsense: On Cora Dia-monds The Realistic Spirit . Journal of Philosophical Research 22:57 73.

Hacker, P. M. S. 2000. Was He Trying to Whistle It? In Crary andRead 2000, 35387.

. 2003. Wittgenstein, Carnap and the New American Wittgenstein-ians. Philosophical Quarterly 53:123.

. 2007. Gordon Bakers Late Reading of Wittgenstein. In GuyKahane, Edward Kanterian, and Oskari Kuusela, eds., Wittgensteinand His Interpreters , 88122. Oxford: Blackwell.

Kenny, Anthony. 2004. Philosophy only states what everyone admits.In Ammereller and Fischer 2004, 17382.

Ko lbel, Max, and Bernhard Weiss, eds., Wittgensteins Lasting Signi-cance . London: Routledge.

Harre , Rom. 2008. Grammatical Therapy and the Third Wittgenstein.Metaphilosophy 39, nos. 45:48491.




20/20

Hutchinson, Phil. 2007. Whats the Point of Elucidation? Metaphilo-sophy 38, no. 5:691713.

McManus, Denis. 2006. The Enchantment of Words . Oxford: OxfordUniversity Press.

Mulhall, Stephen. 1990. On Being in the World: Wittgenstein and Heidegger on Seeing Aspects . London: Routledge.

. 2001. Inheritance and Originality . Oxford: Oxford UniversityPress.

. 2007. Wittgensteins Private Language . Oxford: Oxford UniversityPress.

Proops, Ian. 2001. The New Wittgenstein: A Critique. EuropeanJournal of Philosophy 9, no. 3:375404.

Scho nbaumsfeld, Genia. 2007. A Confusion of the Spheres: Kierkegaard and Wittgenstein on Philosophy and Religion . Oxford: Oxford Uni-versity Press.

Williams, Meredith. 2004. Nonsense and Cosmic Exile: The AustereReading of the Tractatus . In Ko lbel and Weiss 2004, 631.

Witherspoon, Edward. 2000. Carnap and Wittgenstein. In Crary andRead 2000, 31549.

Wittgenstein, Ludwig. 1992. Philosophical Investigations . Trans. G. E. M.Anscombe. Oxford: Blackwell.

. 2005. The Big Typescript . Ed. and trans. C. Grant Luckhardt andMaximilian A. E. Aue. Oxford: Blackwell.



A Resolute Later Wittgenstein

Documents

Transcript of A Resolute Later Wittgenstein