How to be a good philosopher? Are there any rules?

Abstract: The paper represents an attempt to present and discuss a puzzle derived from Ludwig Wittgensteins rule-following considerations, aiming to avoid the existent ambiguities and conceptual misunderstandings. It is shown that the traditional logical approach to language is in principle unable to solve the puzzle. lthough this result was ta!en to balance the odds in the favour of a naturalistic approach to the aim of producing a philosophical theor" of natural language, it is also shown that this approach is e#uall" wrong-headed. few more general considerations with respect to the implications of this negative result are formulated in the end. 1. It has been said more than once about philosophical questions that they are meaningless. And it might be true of at least some questions usually asked by philosophers that they are so indeed. As philosophers, we cannot just rely on our own natural linguistic abilities and hope that this will do. e!eral e"amples ha!e been produced, if only in the philosophical literature of the last century, illustrating situations in which sentences that we would be inclined to consider meaningful pro!e to be, after a more careful e"amination, quite meaningless#. As a consequence, ha!ing a solid theory of language to pro!ide us with the criteria for distinguishing between meaningful and meaningless sentences could be regarded as a prerequisite task for a philosophical in!estigation. I belie!e that such a theory will rest at least on the following assumption$: %A& 'here is a set of rules that go!ern our linguistic acti!ities, such that, for any linguistic utterance, we are able to say if it breaks or agrees with some of these rules. (e can point out which rules were supposed to apply to the considered situation and which, if this is the case, ha!e not been correctly followed. In the second case, not only can we state that a
'his paper was supported by a research grant offered by the Ale"ander !on Humboldt *oundation between +ay and eptember $,,$ in the frame of a special program for scientific reco!ery in -alkans. *or useful comments and suggestions I am indebted to +ircea *lonta, Herbert chn.delbach, /eert 0eil and 0athrin /luer12agin.
) #

Adrian 2aul Iliescu %in 3(ittgenstein and the problem of nonsensical philosophical questions3, 4e!. filos., 567, 8, p. 89:18;,, -ucharest, #::< =in 4omanian>& discusses a few e"amples of this sort offered by (ittgenstein.
$

I ha!e elsewhere %namely, in /heorghe tefano!, 'he (ittgensteinian ?hallenge to the ?ontemporary 2hilosophical Approaches to 6anguage, doctoral thesis submitted at the @epartment of 2hilosophy, Ani!ersity of -ucharest, $,,,, =4omanian, unpublished>& considered other assumptions %like the one that we can distinguish between simple and comple" elements of language, between the sense and the force of an utterance, between language and metalanguage or between logical rules and semantical rules& and tried to show how they could be criticised from a (ittgensteinean point of !iew. 'hat enterprise was somewhat similar %keeping the proportions& to the one pursued by Hacker and -aker, #:<8. Howe!er, I belie!e that %A& goes deeper than any other assumption. Indeed, it is hard to see how an" normati!e semantical theory not assuming a !ersion of %A& would look like.

$ rule that was supposed to be followed has been broken, but we can also reasonably argue, by making appeal to the rule, that it so happened. Accordingly, the theory will aim to state the rules one must obey if one wishes to speak meaningfully. It will also pro!ide a procedure to apply the rules to each possible utterance, in order to see if it agrees with the rules or not. (ere we armed with the set of rules and the procedure to apply them correctly to each case, we would be able to detect, for any question we might want to raise, whether it is meaningful or not. 'hus, we could start a philosophical in!estigation being completely assured that the question we are trying to answer is not a meaningless one. 'he problem is that %A& is wrong and it should be rejected. 'he most radical criticism of %A& is fostered by the later (ittgensteinBs rule1following considerations. In what follows I will do little more %if any& than re1enact (ittgensteinCs arguments. (ith respect to this, I do not wish to get into disputes regarding what (ittgensteinBs rule1following considerations %4*?& really meanD. I would rather take it for granted that I understood (ittgenstein correctly. I am, howe!er, mostly willing to accept that I did not if the arguments presented here turn out to be feeble. -ut if someone accepts them, then I think the whole credit should be gi!en to (ittgenstein8. 'he structure of this paper is as follows. I will first show how (ittgensteinCs 4*? lead to the conclusion that if the rules are seen as some sort of entities %linguistic, abstract or perhaps mental&, then it is not possible to bridge the gap between them and our linguistic practice. 'his, I think, is enough to show that the traditional logicalEanalytical approach to language is in principle wrong1headed. I will not insist on this point because it seems to me that it has been lately understood, and the consequences of (ittgensteinCs arguments in this respect ha!e been gradually gaining acceptance. 'he last part of my paper will be dedicated to showing that an attempt to offer a naturalistic e"planation of the relation between semantic rules and linguistic acti!ities, usually presented as an alternati!e to the analytical approach, is equally mistaken. A few general conclusions will be presented in the end. 2. 6et us begin by looking at an e"ample. uppose that we ha!e two e"tremely simple languages9, 6# and 6$. Fach of them has only three sentences. *or the purpose of this presentation there is no point in getting into the internal
'here are at least four different readings of (ittgensteinCs 4*?. Gne reading has started with 0ripke, #:<$. A different interpretation was initiated by (right, #:<#, pp. ::1##;. A more recent one is due to +c@owell, #::$. %see also +c@owell, #::<, essay ## 1 3(ittgenstein on *ollowing a 4ule3.& Fach of these has generated !arious attempts to pro!ide a solution. *or a systematic summary of the debate generated by 0ripkeBs !ersion see, for instance, -oghossian, #:<:. ?rispin (rightBs !ersion of the argument has recei!ed % inter alia& different replies from both +c@owell and /areth F!ans %F!ans, #:<9&. 'he Hstandard readingI %i.e. the one that I hope not to distance !ery much from& is the one offered by Hacker and -aker, #:<9.
D

Jet (ittgenstein or (ittgensteineans would not agree perhaps with some of the things I am saying here. 'his is due to the fact that my strategy is a more tolerant one %I am inclined to accept as much as I can of the criticised position, if I can still show that it is wrong&.
8

If one is reluctant to accept them as fully de!eloped languages, then I could say that 6 # and 6$, together with the procedure of translating the sentences from 6 # to sentences from 6$ introduced below, form a particular language1game.
9

D structure of these sentences so, since both 6# and 6$ are finite, we could skip formation rules and define our languages in a descripti!e manner: 6# 1 KBAB, B-B, B?BL 6$ 1 KB#B, B$B, BDBL A te"t written in 6# will look like this: MA. ?. A. -. ?.CN 'he same te"t could be written in 6$ like this: M#. D. #. $. D.C Oow we state the following general rule for translating; from 6# into 6$: %4& 'he nth sentence from 6# is correctly translated into 6$ if it is replaced by the nth sentence from 6$. uppose now that we gi!e to a person trained to translate from 6# into 6$ the following te"t: %t#& M?. -. A. ?. A. -. ?.C Gur translator will start with the first three sentences as e"pected, by writing BD. $. #.B he will ne!ertheless continue in a most strange way: B#. $. D. D.B ?onfronted with the rule %4& she will answer that all the sentences were translated according to the rule, but she interpreted the phrase 3replacing a sentence by another3, as it appears in %4&, in a different manner. he might e"press her interpretation like this: %I#& *or any two formal languages which both contain m sentences, replacing the nth sentence from one language with the nth sentence from the other means writing instead of the nth sentence from the first language, alternately, once the nth sentence from the second language and once the {n+1}(mod m)th sentence from the second language. In order to pre!ent such de!iant interpretations we will make our intended interpretation e"plicit: %I#B& *or any two formal languages which both contain m sentences, replacing the nth sentence from one language with the nth sentence from the other means writing the nth sentence from the second language in the place of e!ery apparition of the nth sentence from the first language.
Here MAC, M-C, M?C and M#C, M$C, MDC are not symbols or shortcuts for the HrealI sentences of 6 # and 6$, but occurrences of the sentences themsel!es.
N

2erhaps one could doubt that (ittgenstein would regard translating from 6 # to 6$ as a language1game. -ut then we could de!ise another e"ample. (e may consider, for instance, a situation where only three orders %MAC, M-C and M?C& are gi!en and to understand them correctly is to react in a different way to each %one may think of transporting bricks from three different piles to a certain place, for instance&. It is not important here that the reactions are not within the realm of linguistic beha!iour anymore.
;

8 Oow our translator will retranslate %t#& correctly: BD. $. #. D. #. $. D.B (e make another try with the following te"t: %t$& H?. -. ?. A.I 'his, howe!er, produces an une"pected result: M#. D. $. D.C Again our translator claims that she agrees to the rule %4& and to the interpretation %I#B&. Gnly that this time she has interpreted %I#B& differently: %I$& (riting an e"pression in the place of another which appears within one string of symbols is to create a mirror image of the string and substitute the e"pression accordingly. 6et us stop here for a bit and turn to some more general considerations <. *or any two languages described as sets of sentences we could regard the translation rules as functions from the sentences of one set to the sentences of the other. 'he idea of correspondence between sentences must be somehow e"pressed in the rules. ome phrase marking the correspondence will appear in the formulation of the rule, playing the role the equality sign has in the definition of a function. 'his phrase could always be misinterpreted. (e can imagine an entire class of non1standard interpretations of the translation rule in the following way: *or two finite sets of sentences ha!ing n elements, #P n and C#P Cn, there are nn possible functions which associate elements from the second set to elements from the first set %of which, n! are bijections&. 6et f1 be the standard mapping function and m the number of possible functions. 'here are m! possible arrangements of the mapping functions in a row % m! is still a finite number&. 6et f1 P fm be such a series. 'he pattern of a non1standard interpretation, then, could be: %2& Ap to the ith occurrence of Sk the replacement should be made according to f1Q from the %iR#&th occurrence to the %iRj&th the replacement should be made according to f2 etc. ince i, j etc. are arbitrarily large numbers, it is ob!ious that there is an infinite number of possible interpretations of the translation rule:. Gther patterns of non1standard interpretations are also possible, of course. 'here are non1 standard interpretations of this sort, which will produce, in fact, the same result as f1#,. o there is no point in saying that the standard translation rule should be
In what follows I mainly e"pand on (ittgenstein, #:9D %from now on referred to as 2A& S<N and S#ND.
<

Actually, a procedure could be de!ised to assign to each instance of %2& a real number written in the base n, where n is an arbitrarily large natural number.
:

6et the translation rule R1 be that up to the #, th occurrence of each sentence from 6# MAC should be replaced by M#C, M-C by $ and M?C by MDC and that the replacements should be made differently after that, no matter how. 'he translation rule R2 will stipulate that up to the #, th occurrence of each sentence from 6# we should write M$C for MAC, MDC for M-C and M#C for M?C,
#,

9 pri!ileged because it does not make the application of the function dependent on the number of occurrences of the symbols which are to be processed ##. 'here is no point either in trying to escape the difficulty by replacing the phrase e"pressing the correspondence relation in the formulation of the rule by the phrase used by the recalcitrant translator in his interpretation of the rule %the phrase Hthe replacement should be madeI in %2&, for instance&. *or, since the same phrase appearing in the conte"t of different sentences could be interpreted differently, nothing pre!ents our translator to interpret the phrase differently, now that it appears within a different sentence from his. And there is, of course, no point in saying that it is not plausible that one would take HTI to mean something so different from what it means to us. *inally, specifying that Heach occurrence of Sk must be translated in the same wayI is useless, since HeachI and Hthe same wayI could be interpreted differently from what we intend them to mean#$. It should ha!e become clear by now that regardless of our stipulation of the translation rules, the correct interpretation of the rules, the correct interpretation of the interpretation of the rules and so on, the translator might always interpret all our e"plicit stipulations in a way such that both the translations which seem to agree with our intended rules and those which seem to break our intended rules can be made to agree with the e"plicit stipulations#D. 2erhaps one might want to reply now that our parado" appears only because we take the translator not to understand the metalanguage in which we formulate the translation rules. 'he metalanguage that the translator uses differs from ours in !ery few respects, though. Actually, the differences regard just some phrases used to e"press correspondence relations in the context of the sentences stating the translation rules. Apart from that, nothing is different. Howe!er, we cannot point to these differences properly without making appeal to the semantic rules that go!ern the use of those e"pressions in metalanguage. And here we only ha!e an infinite regress. It is important to note, howe!er, that the initial case had nothing to do with matters concerning meaning directly. 'he fact that our translator understands the meaning of MAC, M-C M?C and M#C, M$C, MDC bears no rele!ance on the e"ample#8. (hat we can say is that he does not understand Hcorrectly translating from 6# to 6$I as we understand it#9. -ut I would like to suggest that, at least for now, we put matters concerning meaning
but after that MAC should be replaced by M#C, M-C by $ and M?C by MDC. And now we can define R3 by saying that up to the #,th occurrence, each symbol from 6# should be replaced by symbols from 6$ according to R1, and after that the replacement should be made according to R2. R3 produces e"actly the same results as %4&, of course.
##

'he source of this obser!ation is Oelson /oodmanCs MgrueEbleenC e"ample from /oodman, #:98.
#$

ee 2A S #<9.

ince i, j etc. may be arbitrarily large, the de!iant interpretation of the rule can account for both the past translations which seemed to accord to the rule and for the present translations, performed after the ith occurrence of the symbols to be translated %we could make interpretations dependent not on the number of occurrences of the symbols, but on the number of applications of the rule, too&.
#D

#8

(e could e!en imagine that he will always substitute M#C for MAC, M$C for M-C and MDC for M?C when asked to write a sentence from 6 $ which has the same meaning with a sentence from 6#. Oe!ertheless, he would not call this Htranslating from 6# to 6$I %and neither should we&.

N and understanding apart and keep focussing on the relation between rules and their applications#N. Gur trouble, therefore, seems to be that it follows from the abo!e e"ample that we cannot distinguish between a correct translation and an incorrect translation by making appeal to the translation stipulations %rules, interpretations of the rules and the like&. -efore going any further, there is one more point I wish to make. In the e"ample abo!e there are a limited number of cases that a translator has to deal with. 'his shows that our problems are in no way related to the fact that in applying a rule we might encounter new, unheard of, situations#;. Oe!ertheless, each situation we would refer to as an Happlication of a ruleI is, in a sense, a new case#<. Gnly that this has not so much to do with the comple"ity of the acti!ities %!erbal or not& the rules regulate. 3. Gur problems seem to arise when we try to bridge the gap between rules and their applications. (e feel that a rule has to determine, so to speak, its correct applications in a necessary manner#:. -ut insofar as we take the rule to be a linguistic entity, i.e. a sentence, we cannot argue for this strong feeling of ours. Gne might think that all our trouble comes from that fact that we took rules to be linguistic entities. If we could regard the rule as something which precludes the need for further interpretation, the parado" should disappear. Indeed, it could be said that a sentence could only be the e"pression of a rule and not the rule itself. 'he rule itself could be perhaps thought of as an abstract entity, which, by its own nature, contains its interpretation. All the cases in accordance with the rule are, somehow, determined by it. 'he problem with this !iew is that there is no apparent way in which we could trace the relation between the rule, seen as an abstract object, i.e. not e"isting in space and time$,, and the actual situations in which we say that someone has followed the rule.
And this could be the case e!en if he alwa"s performs the translations as we do, for while we would be inclined to say that the rule we use is at least similar to %4&, he might be guided by a rule similar to 4D %see note ;&. A set of rules which produces the same result could be one specifying replacements for pairs of sentences %A.A. U #.#., A.-. U #.$., A? U #.D., -.A. U $.#., -.-. U $.$. etc.&. An e"tra rule will concern cases where we ha!e an une!en number of sentences: at the end of the te"t to be translated, when we meet the last sentence, we add MAC to it, translate the pair, and then erase the last M#C from the translated te"t %howe!er, one might complain that this is not using a different rule, but only a different procedure to apply the same ruleQ this is disputable&.
#9

(ittgenstein himself and most of his interpreters, most notably 0ripke, seem to mi" the question of the relation between a rule and its application with questions about meaning, understanding, the relation between the meaning of an e"pression and our linguistic beha!iour e"hibited in using it and so on %but see 2A SS#9N1#;<&. If the rules concerned are semantic rules, this is, I think, fully justified. Howe!er, the e"tent of (ittgensteinCs 4*? goes far beyond that. In this I agree with 4obert -random %see -random, #::8, p. #9&. 'he rules I am mostly concerned with go!ern the distinction between meaningful and meaningless sentences, but ha!e nothing to do with establishing the meaning of sentences or their parts %since they are supposed to be prior to any proper semantical matters&.
#N #;

(ittgensteinCs own e"amples of learning how to write down or to continue a series of numbers and 0ripkeCs e"ample about addition seem to suggest that. Gne can also de!elop an argument going this way from 2A S<,, for instance, but this is incidental to our discussion.
#<

'he e"tent to which the conte"t is making a contribution to a case is best presented in (ittgenstein, #:9N, 7I1D8 par. D18.
#:

ee, for instance, 2A S#<< par.$.

; 'he cases in accordance with the rule could not be causally determined by the rule. 'here has to be another sort of relation. In order to show what this relation consists in, we must talk of some correspondence. And here a !ersion of 2latoCs Hthird manI argument will surely apply. Another strategy would be to speak of the rule as a mental object. 'he problem with our recalcitrant translator seems to be, on this !iew, that he has not grasped the translation rule yet. 'he rule is not Hin his mindI, so to speak. (ere the rule Hin his mindI, there would be no more questions of interpreting or applying it wrongly. (hat the abo!e e"ample shows is at best that we cannot Hput the rule in the mind of the translatorI by using rational arguments. 'his does not make using rational arguments unnecessary, of course. (e could show to someone who has already grasped the rule that he had broken it in one particular case by using such arguments. 6et us ask now what sort of mental object the rule is. uppose it is a mental representation. *or the case presented abo!e, grasping the rule might be, for instance, ha!ing the representation of a translation table. 'he table will perhaps look like this: A ?

V# V$ VD

(e still ha!e the problem of using the table in actual cases. 'he three arrows could be regarded as a single symbol, which pro!ides a substitution rule for MAC, M-C, M?C and M#C, M$C, MDC$#. As a single symbol, it may stand for an" substitution rule, of course. Gne might try to answer to this problem by saying that we actually ha!e three different representations, one for each row of the table, which are kept apart from each other in our mind. I will put aside for now the objection that if the three representations are kept apart from each other there is no indication that they ha!e something in common, which they should, since they relate to the same rule. 6etCs accept for now that things are this way. In order to pre!ent further misinterpretations, we could e!en eliminate the arrows. 'he three representations will then look like this: A# -$ ?D Gne problem with MA#C, M-$C and M?DC is that they contain no hint that they ha!e anything to do with a rule. -ut e!en if we accept that the person who represents MA#C to herself somehow knows that MA#C is a rule, she still might get the rule wrong. 'here are many ways in which one can make substitutions while ha!ing in mind the three representations from abo!e. 'he string MA. ?. -. ?C, for instance, could produce any of the following results: MA. ???. --. ???.C M#. DDD. $$. DDD.C
'his also means that we cannot point directly to the rule and say to our recalcitrant translator: H'his is the rule you were supposed to obeyI.
$,

ee, for instance, 2A S N, for the idea that there is no absolute distinction between simples and comple"es.
$#

< MA#. ?D. -$. ?D.C MA. -. ?. ?.C$$ M#?-?. AD-D. A?$?C etc. (e e"pected the translator to read MAC, M-C, M?C and M#C, M$C, MDC appearing in his mental representations as iconic symbols for the signs she has to work with. It is ob!ious that she did not do so and is not at all clear why she should ha!e done that. And here there still is a correspondence relation we seem to ha!e o!erlooked, namely that between the elements of our representation %MAC, M#C, M$C as images in our mind& and the components of the cases to which our representations ought to apply %MAC, M#C, M$C written on a piece of paper&. 'his amounts to the conclusion that grasping the rule is not ha!ing a mental representation. 6et us now substitute the concept of Ha sign looking like MACI for the mental representation of MAC, the concept of Ha sign looking like M#CI for the mental representation of M#C and so on. 'he person who has grasped the rule that MAC should be replaced by M#C, then, must possess, apart from mental representations of MAC and M#C, the concept of a sign which looks like MAC and the concept of a sign which looks like M#C. he must also possess the concept of Hsubstituting one sign for anotherI. And now the rule could be regarded as a proposition formulated in the language of thought: %416G'& ubstitute the sign, which looks like M#C for the sign which looks like MAC. 'he fact that this is a thought and she possesses the right concepts, we think, pre!ents any misinterpretations of the rule. -ut here we are in no better position than when we spoke of the rule as an abstract object. Oeither an abstract object, nor a mental one has feelers to touch the reality with $D. 'his is perhaps inherent to the way we usually speak of objects and relations: one object does not contain the relation it has with another. 4. 6et us now think that we ha!e a large enough set of cards. ome of them ha!e MAC written on one side and M#C on the other. Gthers ha!e M-C written on one side and M$C on the other. 'hose with M?C written on one side will display MDC when turned o!er. A te"t written in 6# by using the cards could be correctly translated into 6$ by turning all the cards to the other side. I will further assume that this is done mechanically. 2erhaps we could say that the mechanism formed by the cards and the means to turn them o!er is correctly translating from 6# to 6$. 'he translation rules are somehow physically instantiated in this mechanism, and the relation between the rules and their correct applications is a causal relation. And now it may occur to us that we are not in a !ery different position. Gur brains are, in a sense, such mechanisms, though far more complicated. 'his is the starting point of a naturalist approach. 'he problem with our puWWle, as it may occur to the naturalist, was due to the
$$

'his result is e"plained by the fact that the translator takes MA#C, M-$C and M?DC to mean something like a sorting rule: H4ewrite the series of symbols such that, in the end, all the ACs will come first, all the -Cs will come in the second place and all the ?Cs in the third.I
$D

'he earlier (ittgenstein thought that of pictures %see (ittgenstein, #:N#, $.#9#9&.

: fact that we were trying to get from rules to their correct applications by using a Hlogical pathI, so to speak. 'his choice was enforced upon us by our preference for the method of conceptual analysis. Jet, this was not the only a!ailable alternati!e. Another way of getting from the translation rule of our e"ample to the set of appropriate translations would be that of stating the causal laws which describe the functioning of the mechanism translating from 6# into 6$$8. 'he approach proceeds with the remark that grasping a rule is some sort of ability or disposition$9. Howe!er, it will be conceded that when we speak of dispositions or abilities we actually speak of certain processes going on in our brains, processes which cause us to beha!e as we do when we act according to the rules. F"plaining the relation between rules and acting according to rules is nothing else but constructing an empirical theory about those processes and the relations between them and our beha!iour. 6et us now try to figure out how a theory e"plaining the relation between %4& and correctly translating from 6# to 6$ would look like. Gne model for the facts the theory must apply to is this: 'ranslation Input V 'ranslating +echanism V 'ranslated Gutput 'he theory, then, will include some general principles or laws that go!ern the functioning of the 'ranslating +echanism. *rom these laws, with the aid of some deri!ation principles, we should be able to obtain predictions of the form: %2& Ander the circumstances ?#XPX?n, if state I obtains in the mechanism, then state G will also occur. MI C and MG C are two states of the 'ranslating +echanism, which correspond to a 'ranslation Input and to a 'ranslated Gutput respecti!ely$N. Here it is clear that our e"ample helps us to keep things simple: there is no need to speak of an infinite number of states occurring in the mechanism$;. 'here are three I states and three G states. 'he ?#,P,?n conditions are required in order to Hcut the causal relationsI between the functioning of the mechanism and e!erything else going on in the uni!erse %my head included&$<.
$8

ee, for instance, 2ylyshyn, #::< for this !iew. ee the first two sentences of 2A S#8:.

$9 $N

Gne might wish the theory to contain also an e"planation of the causal relation between the input and the output facts %i.e. signs written on paper, for instance& and the I and G states. 'hat part should e"plain, for instance, what happens when the translator mistakenly takes a M-C in handwriting for a M?C and translates it by MDC. Howe!er, I am not concerned with this sort of situations here.
$;

ee note #9. 'he rule of addition or the rule of writing the series of e!en numbers still apply for numbers so large that we would ne!er be able to represent, gi!en the finite number of neurons in our brains. -ut I do not think that this counts as a strong critique of the naturalist !iew %It would be !ery easy to speak of rules for performing addition or obtaining the ne"t e!en number for numbers written in unary base. And we could apply these rules e!en to numbers that we do not read entirely.&
$<

Is our theory supposed to account for neural processes that correspond to Hwilling to translate correctly from 6# to 6$I? I think not. Jet, it is ob!ious that e!en if the ability to translate correctly is instantiated in my brain and the I corresponding to reading MAC

#, 'here is a question that could be raised now: How general are the laws of the theory? (e could think of many different mechanisms, apart from our brains, performing what we would call Htranslations from 6# to 6$I. (e could use water pipes, strings, electric wires etc. in order to build such mechanisms. 'he 'ranslation Input will be coded differently, for each such mechanism, of course. If we build a different theory for each, then we should say that there are different rules for Htranslating from 6# to 6$ with a computerI, Htranslating by using a human brainI and so on. Gn the other hand, it is ob!ious that the functioning of a machine using water is go!erned by different physical laws from those concerning the functioning of an electrical machine. 'he idea of coding the input and the output brings some problems too: we could think of e"actly the same states of a mechanism coding a different input or output. *or instance, let 6D be a language which contains only three sentences: MIC, MIIC, and MIIIC. It is clear that e"actly the same theory that e"plains the functioning of our first mechanism using cards to perform the translation will work as an e"planation of a translation from 6D to 6$. hould we say that we only ha!e one translation rule or two? I will suppose, for the sake of the argument, that the laws of our theory apply only to human brains$: and that we could somehow circum!ent the idea of coding by speaking of the causal relation between the input and the output, on one hand, and the brain states, on the other. 'hus, the occurrences of the sentences from 6D will produce different states in our brain from those produced by the occurrences of the sentences from 6#. -ut we might still feel that there is something not completely in order here. How is it possible that we could deri!e only three predictions %connecting the state of reading MAC with the state of writing down M#C, the state of reading M-C with that of writing down M$C, and the state of reading M?C with the state of writing down MDC& from our theory? (ere the languages 6# and 6$ one sentence bigger, would we ha!e to construct a completely different theory? 'o this one could answer that perhaps there are two kind of laws or principles: some general principles which will perhaps e"plain all the cases of the Mwriting something instead of something elseC kind, and some particular laws which, when added to those general principles, produce only the three predictions we were looking forD,. uppose now that we ha!e completed this theory. Its structure could be roughly represented like this: ' T Y/6, 26, @2, +Z where /6 is the set of general laws of the theory, 26 is the set of its particular laws, @2 is the set of deri!ation principles and + is the meta1theoretical
obtains, I could write down MDC just because I do not want to translate MAC correctly. o, one of the conditions should specify that I want to perform the translations correctly %i.e. that the state corresponding to that occurs in my brain, perhaps in some other region than the one where the translating module is located&.
$:

Gne could say, for instance, that the theory regards Hconsciously translating from 6 # to 6$I, although it does not necessarily ha!e to e"plain the part regarding consciousness. -eing conscious about what one is doing could be regarded as one of the conditions ? #,P,? n and the e"planation for that could be left as a task to a different theory.
D,

omeone might think here of problems like: MIs the functioning of the brain of a person who knows to replace symbols and learns to translate from 6 # to 6$ similar to that of the brain of a person who learns to replace symbols b" learning to translate from 6# to 6$?C %compare this with 2A SS #9N1#9<&

## assertion that the theory offers an e"planation for what is to perform a correct translation from 6# to 6$. +aybe one would like to say that the rule %4& was replaced in this theory by the laws from the 26 set. 'his, I think, is not a !ery good mo!e, since it instantly brings forward questions related to the distinction between norms and natural laws. o perhaps is better not to try to pin the rules on any assertion of the theoryD#. 'he consistent naturalist will probably say that it is not the case to think of the rule as a linguistic entity anymore and therefore it makes not sense to speak of a distinction between two types of statements. 'hat is a fair enough remark, I think. Jet, there might be a problem with this naturalist account D$. ince ' is an empirical theory, it should be at least in principle refutable. 6et us try to find out what would count as an empirical infirmation of '. 6et us think of a translator who translates correctly from 6# to 6$. 'he states described by the theory occur in his brain, according to 'Cs laws and all, the conditions ?#,P,?n are satisfied and the I corresponding to reading MAC also obtains. Jet he makes something which we would usually call a mistake. 'his happens perhaps due to a disturbing factor which the theory did not account for. 'here are two different ways in which we could modify our theory to fit the empirical facts. Gne is to add the condition that the disturbing factor does not occur to the conditions ?#,P,?n. Another possibility is to e"tend the laws of the theory such as to co!er these cases and predict the translator reactions in these situations too. Gnly that in this situation our theory will not be a theory of Hcorrectly translating from 6# to 6$I anymoreDD. -ut why do we say that? 'his was not an infirmation of +. ince a correct translation is whate!er the theory describes, why not adopt the second choice? And if we do not adopt it, is it not because we make appeal to the rule? 6et us now imagine a different situation. Gur translator makes the same mistake. And now we realise that this too can be accounted for by the theory. (e ha!e thought that there are only three predictions that could be deri!ed from ', but we applied the principles from the @2 set incorrectly. And we realise that the translatorCs beha!iour could be also predicted by our theory. o there is no empirical infirmation here. (e ha!e disco!ered that the e"planatory power of our theory is greater than what we ha!e thought. till, we wish to change our theory. (hy is that so? 2erhaps our problem can be pointed at directly in the following way. If ' is to be an empirical theory, then + too, since it is a part of it, must be refutable by the e"perienceD8. -ut how can we speak of an empirical infirmation of + without making appeal to the non1naturalised concept of the translation
And neither on the theoryCs predictions. Indeed, it is quite easy to see that stating a causal connection between reading MAC and writing M#C is no replacement for a translation rule, seen as a norm, since speaking of what will or what would happen, were some other things to happen too, is completely different from speaking of what ought to happen. Gne might say, for instance, HF!en if you are brain1dead, it still applies that you must translate MAC by M#C.I
D#

In what follows, the background of my remarks is contained in 2A S#9< and 2A SS #:D1 #:9.
D$ DD

It could be, perhaps, the theory of Htranslating correctly from 6 # to 6$ and making mistakes when you are !ery tiredI.
D8

'his does not mean that it should be refutable considered in isolation from other principles of the theory. Actually, because + makes reference to all the other statements of the theory, it is only in conjunction with all of them that its sense could be rendered properly.

#$ rule? -eing a meta1theoretical statement, + states a relation between the rest of the theory and the non1naturalised rule. It is ob!ious that we cannot gi!e + up. -ut what can we make of it? 'he naturalistCs reply, as I imagine it, could run as follows: H(hat guides our in!estigation is the model of a mechanism which ne!er makes mistakes. Gur theories are empirical in the sense that they are only appro"imations of such a model. 'he more factors which could affect the correct functioning of the translating mechanism we disco!er and eliminate, the better our appro"imations of the ideal model are. o + states the relation between our theory and this ideal model. (e donCt make appeal to a non1naturalised rule, but to an ideal model within which the causal relation between the processes instantiating the rule and those counting as applications of the rule is ne!er affected by disturbing factors.I 'his account, howe!er, seems to contradict our common !iew about the scientific progress. (e use to think that science ad!ances by offering models that are less ideal and more close to the facts D9. In this !ein, one may think that ?lassical +echanics is a more abstract model of the physical interactions than the 4elati!ity 'heory, which offers a more accurate description of the factsDN. (hy would anyone want to go the other way round? -ut I will lea!e that aside for now. 6et us imagine that we ha!e two such ideal models. Gne is that of a regular translator who ne!er makes mistakes. 'he other one is a model of one of our de!iant translators who ne!er di!erges from the way he is performing the translation. (hy do we choose to be guided by the first and not by the last model? How could the naturalist moti!ate his choice of the first model without speaking of the translation rule? And e!en if he could somehow moti!ate his choice, there still is a problem left. Gn the new reading, what + says is that there is a relation between '[K+L and the ideal model. ince the ideal model is not a physical object, it follows that the relation cannot be a causal one, which makes it ob!ious that + could not be empirically infirmed. If the naturalist is not bothered by that, it is only because he does not realise that now he is in e"actly the same position as the conceptual analyst. Oamely, if he wants to hold on +, he must speak of the relation between an abstract object %the ideal model& and actual situations and processes. If he only wants to speak of the functioning of our brain, then nothing will be objected to that. -ut then he will not be in a position to claim that he has answered a naturalised !ersion of the question regarding the relation between rules and their applications. 5. If it is true that the both adumbrated approaches fail to account for the relation between rules and their applications, then it is hard to see how a mi"ture of the two will succeed to do the job. And it is useless, of course, to try to state rules if we cannot relate them to their applications properly. (hat is wrong with %A&, then? 2erhaps it is wrong to speak of rules and their application as if they were two different things. Anyway, I am not going to try

I think +a" 2lanck was e"pressing this pre1kuhnian !iew !ery nicely. And e!en if this !iew is perhaps obsolete, I do not see how bringing a more sophisticated account of the scientific progress could change the odds in the fa!our of the naturalist.
D9

In taking this to be our common !iew about the relation between these two theories I deliberately disregard more learned accounts of this matter like those offered by Finstein and Heissenberg. I do not think, howe!er, that this !itiates my argument.
DN

#D to ad!ance a solution to our problem here but only to e!aluate the import of the arguments presented abo!e. 'o pre!ent any misunderstandings, it does not follow from the abo!e presentation that we cannot communicate, compute, that our social institutions are illusory or anything of the kind. Oor do I think that any relati!ist conclusions are supported by this account alone. (hat the arguments from abo!e amount to is only that we cannot produce normati!e theoriesD;. Again, this is not to say that there are no !alues, norms or rules. 'hey can be taught, transmitted, e!en enforced. 'hey sometimes collide. 'here is nothing wrong with normati!e talk either, as long as we do not take normati!e talk for what it is not, namely a descripti!e account of norms, rules and !alues D<. 4ejection of theoriWing about norms and !alues might be taken by some to lead to some sort of irrationalism. ince we cannot pro!ide theories and arguments about what is morally, prudentially, practically or in some other way reasonable to do in this or that situation, one may think, it is as if we ha!e no reasons at all. I do not agree with this !iew. @iscussing these matters is, howe!er, far beyond the topic of my paper. \ It seems quite radical, to say the least, to maintain, as (ittgenstein usually does, that all the traditional philosophical questions are meaningless. It is difficult, ne!ertheless, to remo!e the suspicion that some of them might actually be nothing more than nonsensical puWWles. (e do not ha!e solid, ultimate criteria to distinguish between meaningful and meaningless sentences. And it might be the case that is in principle not possible to ha!e such criteria. 'rying to disco!er by philosophical analysis the rules that go!ern the distinction and to state them is doomed to be a isyphean acti!ity. ?onstructing scientific theories which e"plain what happens in our brains when we learn to use the language or when we actually distinguish between meaningful and meaningless sentences could be informati!e, but will not help us a single bit to remo!e the shadow of suspicion cast upon our philosophical acti!ities. +ost philosophers, I do hope, will not think that ignoring 0ant and plunging into traditional metaphysics would be !ery wise. And some would perhaps agree that ignoring the later (ittgenstein or interpreting him such that his !iews do not endanger our philosophical habits is not ad!isable either.

'his is the conclusion of (ittgenstein, #:N#, of course. I donCt think that the later (ittgenstein has changed his !iew in this respect.
D;

'here are cases in which we would say that normati!e talk is inappropriate e!en if no one would take it for theoriWing. (e might feel, for instance, that a sentence like H ince you are a !ery close friend of mine, you ought to e"press condolences to me for the death of my motherI should never be uttered.
D<

#8

References
-oghossian, 2aul, #:<:, 3'he rule1following considerations3, $ind, pp. 9,;1 98: -random, 4obert, #::8, Making it !"#icit, ?ambridge, +A: Har!ard Ani!ersity 2ress F!ans, /areth, #:<9, $o#ected %a"ers, G"ford: ?larendon 2ress, pp. D$$1D8$, 3 emantic 'heory and 'acit 0nowledge3 /oodman, Oelson, #:98, &act' &iction' and &orecast, ?ambridge, +A: Har!ard Ani!ersity 2ress Hacker ,2. +. . and -aker, /. 2., #:<8, (ang)age' Sense and *onsense, G"ford: -lackwell Hacker, 2. +. . and -aker, /. 2., #:<9, R)#es' +rammar and *ecessit,.n .na#,tica# $ommentar, on t/e %/i#oso"/ica# 0n1estigations , G"ford: -lackwell 0ripke, aul, #:<$, 2ittgenstein on r)#es and "ri1ate #ang)age, ?ambridge: Har!ard Ani!ersity 2ress +c@owell, ]ohn, #::$, 3+eaning and Intentionality in (ittgensteinBs 6ater 2hilosophy3, $idwest %tudies in &hilosoph", !ol. 57II, pp. 8,19$ +c@owell, ]ohn, #::<, Mind' 3a#)e and Rea#it,, ?ambridge, +assachusetts X 6ondon: Har!ard Ani!ersity 2ress, pp. $$#1$N$ 2ylyshyn, ^., #::<, article on 3?ogniti!e architecture3 in 4outledge Fncyclopedia of 2hilosophy, 6ondon: 4outledge (ittgenstein, 6., #:9D, %/i#oso"/ica# 0n1estigations, edited by /. H. !on (right, 4. 4hees, /. F. +. Anscombe, translated by /. F. +. Anscombe, G"ford: -asil -lackwell (ittgenstein, 6., #:9N, Remarks on 4/e &o)ndation of Mat/ematics '(emer!ungen )ber die *rundlagen der $atemati!+' edited by /. H. !on (right, 4. 4hees, /. F. +. Anscombe, translated by /. F. +. Anscombe, G"ford: -asil -lackwell (ittgenstein, 6., #:N#, 4ractat)s #ogico5"/i#oso"/ic)s, 6ondon: 4outledge X 0egan 2aul (right, ?rispin, #:<#, 34ule1*ollowing, Gbjecti!ity and the 'heory of +eaning3, in HoltWman, te!en H. X 6eich, ?hristopher +. %eds.&, 2ittgenstein- to &o##o6 a R)#e, 6ondon: 4outledge X 0egan 2aul, pp. ::1 ##;.