Excursus on Wittgenstein ’ s Rule-Following Considerations

In this essay, I seek to demonstrate the interplay of philosophical voices – particularly, that of a platonist voice and a communityagreement-view voice – that drives Wittgenstein’s rule-following dialectic forward; and I argue that each voice succumbs to a particular form of dialectical oscillation that renders its response to the problem of rule-following philosophically inadequate. Finally, I suggest that, by seeing and taking stock of the dilemma in which these responses to the skeptical problem are caught, we can come to appreciate Wittgenstein’s own view of what might constitute a proper a response to the so-called problem of rule-following. This view can be preliminarily characterized by saying that Wittgenstein’s aim is to dissolve the temptation to philosophically rebut the skeptical challenge posed by the rule-following dialectic, an aim he achieves by revealing the semantic emptiness of the apparent sentences that raise the skeptical problem.


Introduction
There is a stretch of Wittgenstein's Philosophical Investigations (PI) known as the rule-following considerations.While within the secondary literature it is widely agreed that these considerations culminate in the paradox of §201,1 two questions remain open: First, where does the rule-following dialectic of PI begin?And second, how should the dialectic's culmination in §201's radical conclusion be understood?This essay takes issue with the answers given to both of these questions in much of the secondary literature.I argue, first, that the rule-following dialectic begins much earlier than most commentators have thought; and I argue, second, that the conclusion in which it culminates is one that Wittgenstein seeks to expose as unacceptable -indeed, unacceptable in such a way as to provide good reason to investigate what triggers the dialectic in the first place.I further contend that Wittgenstein does not offer a direct response to the problem of rule-following, but rather aims to demonstrate the inevitability of being led to a particular kind of philosophical dead-end once one has begun the dialectic.Thus, Wittgenstein responds to the problem only indirectly, by identifying those tacit assumptions that serve to lead us into the rule-following dialectic.
Once the indirect character of Wittgenstein's mode of response is made clear, it becomes possible to see much of what he is doing in the rule-following considerations in a new light.I ultimately argue, contrary to what most commentators conclude or assume, that Wittgenstein thinks there is no (real) problem of rulefollowing.Instead, he thinks the skeptical dialectic is ill-conceived.My account thus differs from two of the better known readings of Wittgenstein on rule-following: that of Saul Kripke (1982), who reads Wittgenstein as accepting the legitimacy of the problem of rule-following while offering a "skeptical solution" to it; and that of Gordon Baker and Peter Hacker (1984), who think (and think Wittgenstein thinks) that, although the skeptical dialectic reveals its premises to be false, it is fundamentally well posed and comprehensible.I argue instead that Wittgenstein thinks that the premises on which the paradox is seemingly grounded simply fail to mean anything when uttered as part of the skeptical dialectic on rule-following.These premises are, in other words, a kind of nonsense.
I take the scholarly contribution of this essay to be two-fold.The first contribution is exegetical: I aim to show that the interlocutors participating in the rule-following dialectic are more numerous than previous scholarship has noticed.Many philosophers have read Wittgenstein as grappling with a platonist interlocutor,2 and when another voice in the dialectic speaks against such a position, the second voice is often regarded as entirely representative of Wittgenstein's own final view of the matter. 3hus, Wittgenstein has been read as rejecting platonism wholesale while accepting a community agreement view of rule-following, the articulation of this latter view having been identified by many readers with one advanced by a voice that speaks in direct opposition to a platonist voice in PI. 4 I will show to the contrary that Wittgenstein's rule-following considerations encompass and reject not only a platonist response to the problem of rule-following, but also the community agreement view widely ascribed to Wittgenstein.Wittgenstein's actual aim in the rule-following considerations is to demonstrate that, while both of these views take themselves to be answering a shared, well-posed problem, this "problem" is a piece of nonsense that, at the outset of the investigation, we are unlikely to be able to recognize as such.Wittgenstein thinks that, since the rule-following dialectic is fundamentally ill-conceived, any account that accepts its skeptical challenge as it is originally posed will be unable to offer a satisfying response to it.
While arguing that Wittgenstein endorses neither platonism nor a community agreement view as a proper response to the issues raised in the rule-following considerations, I will also be concerned to show that Wittgenstein regards many of the remarks made by each of these apparently opposed interlocutors as (each in their own way) unobjectionable when construed in a certain manner.A preliminary characterization of the matter here at issue is that they are no longer to be construed as direct responses to the problem of rule-following.While rejecting his interlocutors' attempts to employ these remarks as direct responses to the skeptical problem posed by the rule-following dialectic, Wittgenstein does not thereby reject these forms of words themselves -or, for that matter, reject everything that an interlocutor trying to express such views might be moved to say.Indeed, he recognizes there to be something correct and platitudinous at the core of what each of his interlocutors is trying to express.What Wittgenstein seeks to show is what becomes of remarks that express truisms -how they come to be transformed into cases of philosophical confusion -when they are asked to serve as the basis for an answer to an ill-posed philosophical question.
The exegetical work undertaken in this essay therefore yields a second contribution, one that we would not be in a position to appreciate without first working through the entire architectonic of Wittgenstein's investigation of rule-following: it is the recovery of certain truths -indeed, platitudes -about normatively structured content that seem at first threatened and then later entirely lost to us when we are in the throes of the skeptical dialectic.I ultimately contend that there is something fundamentally incoherent about insisting on a certain kind of explanation of the meaningfulness of signs.5It is, in particular, the insistence on interposing an interpretation -or, more broadly, demanding some further normative construal -in every case of meaning or understanding that sparks the skeptical rule-following dialectic.And it is such a demand that is ultimately revealed by the investigation to be a "piece of plain nonsense" (PI §119), the culprit at the heart of the rule-following paradox.
Still, it is necessary to work through the dialectic before we can come to appreciate the wrong turn we took at its outset.In order to recognize our mistake as such, we must see how every rejoinder to the problem of rule-following ultimately fails and how the set-up of the investigation itself precludes any possibility of meaningful response.Only then can we come to fully appreciate that it is the very framing of the rule-following dialectic that is senseless and is, as such, responsible for the ensuing paradox and its entanglements.

Framing the Dialectic
In PI, rumblings of the rule-following dialectic begin in §85. 6In this section, Wittgenstein ponders a signpost and queries: where does it say which way I am to follow it; whether in the direction of its finger or (for example) in the opposite one?-And if there were not a single signpost, but a sequence of signposts or chalk marks on the ground -is there only one way of interpreting them?-So I can say that the signpost does after all leave room for doubt.(PI §85) The thought motivating this series of questions and reflections appears to be an ordinary one: that the signpost is open to various interpretations and so leaves (a little) room to doubt its actual meaning.After all, it is possible to imagine circumstances under which the signpost would be said to point in the opposite direction.In so reflecting, we might come to think that, since in itself the signpost is passive, it is really we who give it meaning by interpreting it.The (seeming) platitude can be put like this: it is the signpost plus our interpretation of it that determines the way in which it points (its meaning).And a rule is very much like a signpost.Indeed, Wittgenstein opens §85 with this observation, writing, "A rule stands there like a signpost" (PI §85).When someone states a rule, the rule is presented in words, and these words are much like signposts in that they too may be taken in ways that are contrary to their actual, intended meaning.They are, in a sense, just noises, and we can easily imagine how they could have meanings other than their actual meanings (at least in suitably adjusted circumstances).
In order to bring the point out, let us imagine the following dialogue in which someone is introducing us to a way of counting (reciting a series of numbers) that he calls 'gimbling'.He says, "To gimble, one must count like this: three, five, seven, and so on".
"Ah", we say, "By gimbling you mean we are to count by odds (from three).So we shall continue the series like this: nine, eleven, thirteen, fifteen, and..." "No, that is not what gimbling means", he says."The series goes three, five, seven, eleven, thirteen, then seventeen... and so on." Finally, it clicks: he is reciting the series of prime numbers (from three) and so, we might say, is "counting by primes".To confirm our hypothesis, we say, "And so next comes nineteen, and then twenty-three, and then twenty-nine.Is that right?"At this point, he confirms that this is how one continues the series, and we conclude that we all know the meaning of the word 'gimble'.We see that, until we supplied it with an interpretation, 'gimble' was just squiggles on a page or a perturbation of the air, no more meaningful than swirls in a wooden desk or the squawking of a chair as it is dragged across the floor.Only in our regard for the sign as meaningful and then by our subsequent interpretation of it did it actually become so.The lesson is: the meaning of a sign does not consist in something found in the sign itself.That is, signs are not intrinsically meaningful, or meaning-bearing all on their own.It is by recognizing a sign as such and then through interpreting it that we are able to say, e.g., that to gimble is to count by primes.
In the example we imagined, we guessed the meaning of 'gimble' on our second try, but we need not have been so lucky.It would be easy to specify a different rule that also picked out the recited numbers.Thus, it might strike us that, though we made some limited confirmation of our interpretation of 'gimble', we did nothing that could count as proof of our interpretation.Just as the first numbers of the series permitted various interpretations, the further developed (but still finite) series by which we came to grasp 'gimbling' -and then the series by which we confirmed our grasp of that shorter series too -also must permit various (indeed, infinite) interpretations.
What has been said here of 'gimble' seems like it may be said of any word at all -for example, 'chair', 'yes', 'peacock', and 'yellow'and even any other sign (moving from the species word to the genus sign, which includes such nonverbal species as stoplights, hieroglyphs, raised middle fingers, and so forth).For, though we have more experience with these signs, and so have more information on which to base our hypotheses about, or interpretations of, these signs, it remains conceivable (if farfetched) that we have supplied any one of them with an incorrect interpretation.The problem may be put like this: the meaning of a sign must be more than a mere aggregation of the instances in which it has been (correctly) used since the sign itself must be indefinitely applicable -there is, for example, no limit on the number of things that we might rightly call a chair -but since any set of finite applications of a sign permits infinite interpretations, we cannot abstract with certainty any one interpretation of a sign (a rule governing its use) from the applications of it with which we are actually acquainted. 7t might now appear that, though it may (in theory) seem difficult to correctly interpret the meaning of a sign, it is not actually difficult to do so.As a practical matter, it usually does not take us long to ascertain, e.g., that 'gimbling' means 'counting by primes'.The problem, though, is deeper than that since interpretations too are given as signs and are thus just as open to misinterpretation as the signs they are supposed to interpret.It thus appears that every interpretation needs its own interpretation since their meanings are no more evident than that which they interpret.As Wittgenstein puts it: Interpretations are all, so to speak, too slippery to fix the meaning of a sign.
Considerations such as these drive us to ask, "How does an interpretation bring anything into accord with a rule?For, every interpretation requires another interpretation -and so on ad infinitum".In seeing that an interpretation can neither bring anything into accord with a rule, nor exclude anything from being in accord with a rule, we are threatened with being unable to make sense of the very distinction between applying a rule correctly and applying it incorrectly.There appears to simply be no difference between the case in which I use a sign rightly and the case in which I use it wrongly.For, if there is no adequate response to these worries, there is nothing that fixes the meaning of any sign, and then a sign can no more be said to mean this than it can be said to mean that.Thus, the epistemic worries with which we began the rule-following dialectic have changed form.No longer are we asking whether we can achieve certainty about the meaning of a sign.Now we are asking how there can even be such a thing as the meaning of a sign, and the apparent answer is: there cannot be such a thing.With this, the rule-following paradox has come fully into view.In reaching this "conclusion", though, we saw off the branch upon which we are sitting."Words have no meaning", we say, adding, "Of course, the words 'Words have no meaning' have no meaning either -and neither do these very words!" 9 Two of the most noteworthy responses to the rule-following dialectic (arising within PI but also existing in the secondary literature) might be termed "rule-following platonism" and "the community agreement view".We will now consider each response in turn, paying special attention to the form they take as they develop.Thus, Section 2 of this essay focuses primarily on explicating the structure of the platonist response to the skeptical problem.It will then be important to note that the basic structure of the platonist response recurs with the community agreement view.Establishing this overview of the dialectic -and especially of how the two most prominent responses to the skeptical problem it poses develop -will clear the way for understanding Wittgenstein's ultimate treatment of the problem of rule-following.

Rule-Following Platonism
Since each interpretation with which a sign might be equipped may itself be misinterpreted and thus be in need of its own interpretation (and so on, impossibly, ad infinitum), the platonist concludes that there must be, as it were, a last interpretation -that with which we have supplied the signs when we finally, really do understand -which is itself more than mere dead signage.The last interpretation -the meaning itself -must be something that, when we grasp it, guarantees that we know how to apply the rule or use the sign correctly, without need of any further interpretation.Describing such a thought, Wittgenstein writes: What one wishes to say is: "Every sign is capable of interpretation; but the meaning mustn't be capable of interpretation.It is the last interpretation".(BBB 34) 10 The platonist agrees, in other words, with the rule-following skeptic that something must mediate between sign and meaning -and, furthermore, that it cannot be any ordinary interpretation that does so -but, unlike the skeptic, posits there to be something like a special "last interpretation" that effects such mediation.
There are two distinct stages in the development of the platonist response (stages that we will see mirrored by the community agreement view), the primary aim of each stage being to specify exactly what the posited "last interpretation" could be.The first we can call "mental platonism".Wittgenstein describes someone tempted by (what I am calling) mental platonism as: [thinking] that the action of language consists of two parts: an inorganic part, the handling of signs, and an organic part, which we 10 See also Z §56; §231.may call understanding these signs, meaning them, interpreting, thinking.(BBB 3) The rough idea is that the signs that are produced in the world are tools (of a sort) that we use to express the thing that is really meaningful (the "last interpretation") which is itself something mental.In thinking this through, though, it becomes apparent that nothing mental can do the trick, i.e., meet the desiderata required of the posited "last interpretation".Wittgenstein's objection, as it emerges in PI, is this: if the "last interpretation" is mental, there is necessarily a gap between it and its application in the world (i.e., between the internal and the external) which must be bridged.That is, if the "last interpretation" is mental -in one's head -then this mental "last interpretation" must still be translated into, or applied to, an actual application (e.g., I must pronounce the word or follow the signpost).So I may see in my mind (that) "A is followed by B", but in the application of this understanding -in which I follow A with B -there exists a gap that must yet be bridged by further normative construal, or a method of projection. 11By what can it be bridged?Well, it seems only another interpretation will do.But if this is so, then the platonist has not found his special "last interpretation".
Regarding such cases, Wittgenstein writes, "The application is still a criterion of understanding" (PI §146). 12That is, we still (must be able to) apply normative judgments to that which is "present in our minds".Thus, as we think through what is required of the "last interpretation" for it to succeed as a direct, head-on response to the problem of rule-following, we come to see that no mental item could do what the posited "last interpretation" must do.The only way for this account to be an account of rule-following is if we conceive of the special mental item as something that can be grasped correctly (or incorrectly).This means, however, that we remain firmly within the circle of the normative -mental platonism does no more than identify the source of the normativity of rule-following in the normativity of our (internal, mental) understanding. 13Thus, the skeptical problem that the platonist is trying to answer is merely shifted to another level -that of the mental -where it recurs in a form that is as demanding as ever.
Given the impossibility of a mental item being the posited "last interpretation", the platonist is pushed, in the second and final stage of the development of the platonist response, to what we can call "full-blown platonism".The "last interpretation", the meaning itself, cannot be a mental item for the reasons just rehearsed, and so, in an attempt to purge all traces of (human) normativity from his "last interpretation", the platonist posits fully platonic meanings, i.e., meanings that are neither worldly nor mental, but rather denizens of some supernatural realm. 14Meanings, it has come to seem, must stand completely outside of normal (worldly) cause-and-effect relationships and must somehow be abstract universals that exist beyond time and space.
To put the objection to this form of rule-following platonism as briefly as possible: The platonist, in order for his response to succeed, must explain how these fully platonic items can account for the occasions on which I mean or understand anything.But it is unclear how, at this point in the dialectic, such items can offer any account of (human) rule-following.For, according to the best epistemological theories, some form of causal interaction must obtain between the knower and that which is known. 15Thus, it would be necessary for these platonic items to causally interact with rule-followers existing in time and space.This possibility has already been ruled out, though, by our classification of these objects as platonic, i.e., as abstract objects located somewhere outside of time and space. 16As Matthias Haase puts it: … whatever such objects might be exactly, their minimal characterization already rules out their contributing to the resolution of our problem: as abstract objects they cannot causally interact with individuals in space and time and thus cannot explain the acts of individuals that are in space and time.(2012: 240)   The platonist, who sets out seeking a special "last interpretation" in response to the problem of rule-following, is forced at every juncture to further articulate his claims.He first conjectures a mental "last interpretation", but this proves inadequate in that, in order for the account to work, it is necessary to simply presuppose that the posited mental items have a normative character.Yet after attempting to purge his account of every trapping of terrestrial normativity, it becomes apparent that the posited item, being outside of time and space, cannot provide a satisfactory explanation of how rules can be followed in time and space.17Thus, the platonist faces a dilemma he cannot escape: The first horn is that, if the "last interpretation" -the meaning itselfto which he appeals already has an evident normative character (of mundane origin), then any attempt to ground (human) rulefollowing in such a thing will be open to the charge of circularity.The second horn of the dilemma is that, if the "last interpretation" to which he appeals has no (ordinary) normative character -if his "last interpretation" is completely supernatural -then, it is in no position to account for how actual people actually follow rules.

The Community Agreement View
In this section, we will consider a further reaction to the rulefollowing dialectic -and, in particular, to rule-following platonism -known as the community agreement view (which I will equally call "communitarianism").What I will discuss under this heading is a philosophical response to the rule-following paradox that gets voiced in a number of passages in PI (especially § §202-242); and according to many commentators, 18 these passages should be read not only as reacting and responding to the platonist response, but also as representing Wittgenstein's own view of what constitutes a proper response to the problem of rule-following.I will argue to the contrary that the communitarian response no more corresponds to Wittgenstein's own view of the matter than does the platonist response.For the moment, though, I will simply concentrate on describing how the communitarian response arises within Wittgenstein's rule-following dialectic.
The communitarian reasons that, since the problem with signs seems to be their endless misinterpretability, what is needed is something that cannot be misinterpreted.Indeed, the platonist and the communitarian are thus far in agreement, but in reacting to platonism's failure to get an unmisinterpretable "last interpretation" into view, the communitarian posits something else that is supposed to be equally indefeasible -namely, the supposed bedrock of community agreement.The communitarian gives voice to a picture on which meanings are fixed by some form of community agreement (or communal expectation, or community approval, et cetera).It is, roughly, how everyone in a community takes a particular sign that determines its meaning, or the way in which everybody expects a particular rule to be applied that determines its application.So, when we ask, "What does that sign mean?" we ostensibly find out, in the most basic kind of case, by checking to see how the community as a whole responds to it -the community's de facto consensus is supposed to give us what we need to know (i.e. the relevant normative standard).
When further considering what the communitarian response might amount to, we will see that there are various -and extremely different -ways in which it can be spelled out.We said that, on this picture, a sign's meaning (and our understanding of it) rests on community agreement.How, then, should we understand this agreement?We can ask: does it mean what the community should agree on (or what it thinks it should agree on)?Or does it mean what the community actually agrees on?Is it possible for an individual's use of a sign to deviate from what the community agrees to be the proper use of that sign -so that not every use of a sign can be considered constitutive of the community's agreement?Or is community agreement simply a set of brute regularities found in the behavior of the members of a community?In pressing such questions, we will find there to be a waffle at the heart of the communitarian response (a waffle not unlike what we saw in the platonist response).The communitarian response oscillates between (1) a version of it that seems philosophically unhelpful in responding to the problem of rule-following and (2) a version of it that ultimately proves unintelligible because it attempts to accommodate the skeptic's own understanding of the problem at hand.
We saw earlier that, as it attempts to explain the normativity of rule-following, the platonist response oscillates between appealing to (1) something that itself presupposes a normative structure -i.e., mental understanding -and appealing to (2) the merely factual (and ultimately unhelpful) presence of an extraordinary, platonic "last interpretation".So too, if the communitarian appeals to the community's understanding of how a term should be used -if he considers only those cases in which a sign is used properly and so considers only correct uses of a sign in order to get into view the cases that are supposed to constitute a community's agreementhe never exits the circle of the normative.Instead, he only pushes the skeptical problem back one step, shifting it from the level of the individual to the level of the community.(I will call the version of the community agreement view that succumbs to this fallacy the "intentional variant" of the response.)So, just as the platonist was forced to purge all traces of (terrestrial) normativity from his account of the regress-stopping "last interpretation", the communitarian too must purge his own regress-stoppercommunity agreement -of all traces of normativity in order to avoid giving a merely circular account.The communitarian must characterize community agreement in such a way that his account seems to explain the normative in terms of something that, considered in and of itself, is entirely non-normative.Thus, in the second stage of the communitarian dialectic, we come upon a communitarian who insists not only that he is appealing to community agreement, but that he is in doing so appealing to a brute, empirical fact -namely, that of a certain kind of regularity arising within the relevant community.He must ultimately claim that this underlying factual agreement somehow fixes normative standards for the uses of signs and applications of rules."Somehow fixes" -how?It is this question that the second form of communitarianism must try to answer.(I will call the version of the response that tries to answer this question the "resolute variant".) On the intentional variant of the response, the community has expectations, intentions, and so forth -all in some unexceptional sense -and these forms of agreement are thought to determine how words and rules are to be projected into novel contexts. 19It is an intuitive, minimally theoretical way of hearing the communitarian response and thus a natural place to begin trying to understand it.We will see in the end, though, that communitarianism's intentional variant cannot provide a direct answer to the problem of rule-following.
In our first pass at communitarianism (i.e., its intentional variant), we regard the community as (truly) forming a consensus, approving of certain actions, expecting certain applications, and so forth.But insofar as we so regard the community, thereby analogizing the community's intentional acts and states to the individual's, these very things are subject to possible misinterpretations -just in the way signs are.Suppose a community expects that, when instructed to count by twos, its members eventually continue the series 1000, 1002, 1004, and so on.Now we may ask, "What is the difference between the community that expects the further eventual continuation of the series to be 2000, 2002, 2004, and the community that expects the series to eventually be continued 2000, 2004, 2008?What fact establishes that the community's expectation entails this application and not that one?"Nothing seems to justify interpreting the community's expectation in one way or the other -and so, once again, we encounter a regress of interpretations.
Evidently, the groundwork for objecting to the intentional variant of communitarianism has already been laid.If, due to the possibility of misinterpretation, an individual cannot determine the meaning of a sign by his lone intention, then neither can a group of individuals by their collective intention.If, in other words, we think of community agreement as intentional, interpretation -i.e., further normative construal -is required for the account to work.Thus, the intentional variant of communitarianism only pushes the skeptical challenge to another level (as did mental platonism) where the problem of rule-following recurs, entirely undiminished.
One can imagine the communitarian, in reacting to the considerations we have just made, further developing his response as follows: "It is what the community actually expects and sanctions -not how we interpret its expectations -that matters".In other words, we need a resolute understanding of community agreement, one that does not simply ascribe the very same kinds of intentions, expectations, beliefs, et cetera -so problematic at the level of the individual -to the community at large.As we attempt, in the coming pages, to articulate the resolute variant of the communitarian response, we will see that the view eludes us as we try to focus in on it.In working through the dialectic in this wayfirst, through the intentional variant; then, the resolute variant -we can see there to be something deeply confused in the communitarian response.There is (1) its desire for some set of plain, brute facts to ground rule-following and (2) the inability of those very facts to have even the minimal (normative) structure required for such an account to be coherent.Together, (1) and ( 2) create an unacceptable dilemma for communitarianism.
On the resolute variant of the community agreement view, 20 community agreement is regarded as consisting of a set of brute 20 Kripke (1982) puts forward a community agreement view of rule-following that has been widely understood as the view that we are about to consider, but the view I think Kripke actually articulates is not the one in which I am interested here.So, while I do not take myself to be targeting Kripke per se, I am targeting the received Kripke.For people who read Kripke as putting forward the view that I target under the heading of "resolute communitarianism", see Baker and Hacker (1984: 4), Bloor (1997: 60-64), Paul Boghossian (1989: 519), Anandi Hattiangadi (2007: 65-66), Kusch (2006), McDowell  (1992), John Searle (2002: 252-253), Williams (1999: 162-163), and Wright (1989: 234).For people who take issue with this reading of Kripke, see Alex Byrne (1996) and George Wilson (1998).
facts.What is supposed to fix the right way of using a sign or applying a rule is limited to what the community has actually sanctioned or approved of or the ways in which the community has actually used a sign (regardless of whether we would ordinarily say that certain of these uses were incorrect).The idea is that, if the community has actually sanctioned a particular application of a rule or use of a sign (or actually uses it that way), then, and only then, can we say that to be the correct way of applying it.It is supposed that, on this specification of it, the community's agreement -which is supposed to provide standards for the use of signs -is of such a nature that it requires neither interpretation nor normative construal, thus providing an unimpeachable account of rulefollowing.
While the resolute communitarian concedes there is something right about the arguments of the rule-following skeptic -and so casts his account of rule-following in terms that are supposed to accommodate the skeptic's understanding of the problem -he tries to resuscitate our grip on rule-following.The resolute communitarian gives up on intentional meaning-facts -these, he thinks, have been shown to not really be facts at all -and instead grounds his account of rule-following in what he sees as truly nothing more than mere, brute facts, finally jettisoning anything that might require interpretation (and thereby set off a regress) and replacing it with something that we are supposed to have no choice but to acquiesce in.Crispin Wright offers the following characterization of this variant of communitarianism (which he attributes to Kripke): According to Kripke's Wittgenstein, all our discourse concerning meaning, understanding, content and cognate notions fails of strict factuality -says nothing literally true or false -and is saved from vacuity only by a 'Sceptical Solution', a set of proposals for rehabilitating meaning-talk in ways that prescind from the assignment to it of any fact-stating role.(1989: 234)   In other words, though we have given up on a certain kind of account of meaning, we putatively recover a simulacrum of our original conception of rule-following.This is achieved by substituting, in our assessments of the uses of signs, the applications of rules, and so forth, "approved by the community" for "true" or "correct" and "disapproved by the community" for the opposite.We might say: we identify a certain kind of logicthough not the one we expected to find -in our practice of ascribing truth and falsity to sentences and so get back some semblance of "meaning-talk".Of this variant (which he too ascribes to Kripke), Paul Boghossian writes, "The proposed account is, in effect, a global non-factualism: sentence significance is construed quite generally in assertion-theoretic terms" (1989: 519).Again, instead of getting what we would ordinarily imagine to be facts of meaning, we get an analysis on which an application of a rule will be deemed "correct" or "incorrect" based on whether the community has actually displayed agreement with respect to that particular application of the rule.
It is important to note the following consequence of the resolute communitarian's claims: While some rules -indeed the most interesting ones -are ordinarily thought to determine an infinite number of recursive iterations, it is difficult to see how any actually existing community (whose historical extent is limited in time and space) could ever react to or employ (and thereby establish a genuinely normative standard for) infinite iterations of a rule.Thus, on communitarianism's resolute variant -according to which facts of community agreement are limited to what the community has actually had occasion to agree on or the ways in which the community members have actually used a sign at some point in the past -there can be no fact about whether a particular, seemingly potential, iteration of a rule is in accordance with that rule if the community has not actually agreed that the rule is to be so applied.
By pressing the right questions, the underlying incoherence of this account can be quickly brought to light.So far, in our elucidation of this variant of communitarianism, we have allowed ourselves to take the following idea for granted: that there is some comprehensible thing that we can count as that of which the community approves with regard to its usage of a particular application of a rule.For example, we have taken for granted that the first steps of the instruction "add two" can be revealed as "2, 4, 6", even if there may be no answer as to how the series is to be continued beyond some further iteration of it.But now we ask: what rule determines that the first three members of the series are always "2, 4, 6"?What if, in the month of November, the instruction "add two" calls for "duck, duck, duck" to be the first three members of its series? 21ven if a rule had a merely finite number of instances, we would ordinarily take it to have some generality, e.g., take it to be able to settle that the first three members of the series are "2, 4, 6" on indefinitely many occasions.At this point, however, we are simply reencountering a problem with which we were previously confronted (and had imagined ourselves to be avoiding).The resolute communitarian said that, for certain further expansions of a series, the question of "the right way to go on" has no application.Yet the problem we were trying to avoid by making this claim equally applies to the question of how the first three instances of a series -e.g., "2, 4, 6" -can be determined with any generality whatsoever.Thus, the problem of rule-following has here destroyed not only the possibility of further expanding the series, but also the possibility of establishing a way of repeating the beginning of the series.
We can press this problem further by asking, "What rule determines what counts as a repetition of a series -say, '2, 4, 6'even barring the possibility of the rule calling for a putatively different first three members in the month of November -say, 'duck, duck, duck'?"In other words, what counts as a bare repetition of "2, 4, 6" -and how could community agreement ever establish such a thing?The problem is not, as it might initially appear to be, that I might not know that what seems to me to be a repetition of a particular series will also be regarded by the community as such a repetition.The actual, far more baffling, problem is: our talk of "the same" makes no sense here -for, on this resolute variant of communitarianism, "the same" is supposed to be constituted by the community's agreement.If, however, a community's agreement is no more than brute behavior occurring 21 Wittgenstein considers such a problem in PI §214.on distinct occasions separated by time and space, it seems there is nothing that binds together what the community approves of on one occasion and what it approves of on another occasion.There is, in other words, no way to conceive of the community's approval as it occurs on one occasion as potentially reaching to some other (spatio-temporally distinct) occasion, thereby sanctioning another use of a sign.Thus, the problem is not that it is difficult for me to know some fact here -rather, there is just no fact here for me to know.
At this point, we have discussed two problems which should be distinguished.First, without actual community approval of this particular instance of a sign, or this very application of a rule, there is no telling whether the community approves of it.Second, even if the community does seem to approve or disapprove of a particular application of a rule, there is still nothing to fund the claim that this application of the rule is the same as some other previous application of a rule. 22Nor, of course, is there anything to fund the claim that the rule being applied here and now is the same as some other previously applied rule.Thus, while it may be provisionally granted that the community could come along and approve of this very application -"2, 4, 6" -of this instruction -"add two" -such approval would have no bearing on whether this -"2, 4, 6," -is an application of this -"add two".Neither these two distinct applications nor these two instructions can be said to be the same as any others.Such claims have no meaning here.
In the preceding description of the problem, we have still allowed ourselves to employ the concept sign in order to speak of distinct occurrences of signs.In thinking through the implications of the problem, though, we lose our grip on the very idea of a sign as some distinct "thing".For, our concept of a sign, no matter how semantically inert we might think signs to be, still must be the concept of something that can be identified as recurring on 22 These points will sound familiar to readers of Warren Goldfarb's "Kripke on Wittgenstein on Rules" (1985).Of such an account of rule-following, he writes: "At best it can draw on actual, face-to-face occasions of acquiescence of persons to each other" (1985: 484).I will go on to argue that this best-case scenario for resolute communitarianism does not obtain.separate occasions.If, however, we lose our grip on the very idea of a rule -viz., on the very idea that two instances of something can indeed be instances of the same thing -then we simultaneously lose our grip on the concept of a sign altogether.This applies not only to our ability to assign the same meaning to two distinct occurrences of a sign, but also to our ability to recognize two distinct occurrences of a sign as being occurrences of the same sign.We seem to have abolished the concept of same not only at the level of meanings, but also at the level of the signs themselves.
So far, we have been exploring some difficulties of the resolute variant of communitarianism by focusing on the question: "Of what does the community approve?" -i.e., what is the thing at which a community's approval is aimed?There is, though, at this point in the dialectic, something equally unintelligible in the idea that there is something -a distinct, general capacity -that the community exercises in approving of the use of a sign.We can bring this out by focusing on the question: "What is the community's approval?"For, how am I to say that the community approves of this application of a rule, or that they agree on this use of a sign?Is approval to be expressed in the same way as it was in the past?But what counts as "the same" here?The community's approval cannot tell me since it is exactly what counts as their approval or agreement that I am asking after.Even if, then, I got the community to look at the series written here -"2, 4, 6" -I would not be able to know whether the community approved or disapproved of it (or had taken no notice of it, or regarded it as art, or...) without there being some way in which one could recognize that two exercises of the community's capacity for approval were two expressions of the same attitude. 23At this late stage of inquiry, anything which we might want to pick out as a bit of linguistic behavior can issue in nothing more than mere noise, and all forms of supposed linguistic agency have been reduced to brute motion.
In communitarianism's initial construal -i.e., on the intentional variant considered earlier -it does not adequately break with our ordinary ways of thinking about rule-following and so is open to the charge of circularity.Then, as the communitarian takes the full measure of the skeptical problem and tries to articulate a resolute understanding of communitarianism, his own conception of what he is appealing to dissolves into incoherence.The movement we have seen -that between the latently inadequate intentional variant and the patently incoherent resolute variant -is representative of a confusion, present from the beginning, in the communitarian's way of recoiling from platonism.As we traverse the dialectic enacted by the communitarian, passing from the intentional variant to the resolute one, we come to appreciate that the communitarian is faced with a dilemma (remarkably similar to the platonist's).The first horn of the dilemma is this: (1) if the facts of community agreement that he appeals to already have an evident normative character, then any attempt to ground rule-following in them will be open to the charge of circularity.The second horn is this: (2) if he appeals to brute facts that have no evident structure at all, then these facts are in no position to ground anything.In seeking to escape the first horn of the dilemma, the communitarian runs right into the second.

The Form of Wittgenstein's Treatment
A primary aim of Wittgenstein's rule-following considerations is to evoke how, in applying a certain picture in our philosophical thinking on rule-following, it can come to be therein misapplied.While this (mis)application of the picture may strike us as obvious or pre-philosophical -i.e., as something that we do not mark as an "application" of a "picture" at all -it is this tacit application of it that is the "decisive move in the [philosophical] conjuring trick" (PI §308), i.e., the unmarked sleight of hand that generates the rulefollowing paradox.The picture of which I am speaking -the one that gets misapplied -is that with which the rule-following considerations begin in §85: "A rule stands there like a signpost".In itself, there is nothing wrong with this picture (it is, after all, just a picture).It is in trying to make certain uses of this picture -in misapplying it, in imagining ourselves to be gleaning certain philosophical insights from it -that we wander unknowingly into the skeptical framing of the rule-following dialectic.What, then, was philosophically fateful in the way in which the problem came to be unwittingly framed?
At the outset of the investigation, we have the picture of a rule such that it "stands there like a signpost".We have, in other words, noticed a certain way in which a rule might be said to be passivei.e., in that it "stands there" or sits there on the page -and thus far have merely found a picture that suits us.This picture comes to be misapplied when we imagine we have derived a certain philosophical insight from it -namely, that the signs are dead, totally inert or lifeless, and always in need of interpretation in order to come alive. 24That is, the mistake is to insist, once having noticed how signs (really do) "stand there", that in every case the ostensibly dead signs require further normative construal.
In seeing how the whole dialectic hangs together -how all the direct rejoinders to the problem of rule-following lead nowhere in the end -we can come to appreciate that the rule-following paradox is unavoidable once we try to think of signs in this way.The inevitability of paradox, once seen, finally makes evident that, in trying to construe our picture of the sign sitting on the page, or "[standing] there like a signpost", such that we imagine ourselves to have seen that -really -every sign is dead, we are speaking nonsense.For, we are unable to get into view what it would really mean for the signs to be well and truly dead.If we take this starting point to be obligatory, we move gradually further through the rulefollowing dialectic until we finally feel compelled to endorse a "conclusion" that is unintelligible in that it purports to deprive us of the very capacity we must exercise in asserting it.
In §201, Wittgenstein writes that, in adducing considerations such as those so far raised, "what we hereby show is that there is a way of grasping a rule which is not an interpretation" (PI §201).In other words, Wittgenstein takes the paradox to reveal that there is something wrong in the way of thinking that has led to it: namely, that way of thinking on which it seems as if the only way to grasp a rule -to move from sign to meaning, or rule to application -is via interpretation.Since trying to look at things in this way leads to paradox, Wittgenstein reasons, we should reject it.Thus, §201 can appear to have the form of a reductio ad absurdum.It would be better, though, to call it a revelatio absurdi (a revelation of absurdity).For, what we see is not that a premise is false, but rather that a certain way of applying a picture leads us to nonsense.In one sense, then, we do not reject anything at all but instead see the error of trying to look at things in a certain way. 25Our manner of applying the picture does not have enough sense to be rejected as false -there is, at this intellectual crossroads, nothing firm for us to jettison.
At one point, Wittgenstein writes: the fundamental fact is that we lay down rules, a technique, for playing a game, and that then, when we follow the rules, things don't turn out as we had assumed.So that we are, as it were, entangled in our own rules.(PI §125) And this is just what we have seen come to pass in the unfolding of the inquiry into rule-following.We make a certain application of our picture -we impose a requirement on rule-following according to which every case of it is supposed to involve an interpretation (thus setting up how the game is to be played) -and then, in following out the ensuing dialectic, find that we become "entangled in our own rules".In seeing past our imposition of this requirement, we can reflect and come to appreciate that, after imposing it, anything we might call on can only come too late to be of any help in responding to the problem of rule-following in the manner in which it has been raised.By the time we find that there is a need for some special regress-stopper to save us from a regress of interpretations, we have already acquiesced in the skeptical framing of the investigation and thereby succumbed to the inevitability of its paradoxical conclusion.We thus find ourselves unable to make sense of the very rules we laid down at the outset of the dialectic.
In Wittgenstein's revelatio absurdi, the veneer of sense is removed from formulations such as "Every sign is dead and in need of interpretation" -and for us, where once stood "disguised nonsense" now stands "patent nonsense". 26We finally come to appreciate that, when we previously took ourselves to be asking after the implications of a sign's meaning always being determined by an interpretation, we had nothing clear in mind about which we were asking -our entire investigation was centered on the mere illusion of sense.
Only by fully indulging in the temptation of our imagined insight about the "dead" signs do we eventually come to see our way past this temptation.This requires tracing the rule-following dialectic in its entirety, following it through to every dead end until the inquiry finally becomes manifestly paradoxical. 27Thus, Wittgenstein writes: In philosophizing we may not terminate a disease of thought.It must run its natural course, and slow cure is all important.(Z §382) In seeing there to be no direct, head-on answer to the problem posed at the outset of the rule-following dialectic, we are supposed to come to see that the framing of the problem was where the mistake was made.We trace our way back to the very start of the dialectic and there uncover a use of words that is "disguised nonsense".Only when we come to ask ourselves, in regard to, e.g., the words "A rule stands there like a signpost", questions such as "When would I say such a thing -and what, in saying such a thing, would I actually mean?" do we come to see that, at the beginning the inquiry, we are yet to assign any meaning to our words -they are, as Wittgenstein puts it elsewhere, "idling" or "on holiday" (PI §132; §38). 26Of this method, Wittgenstein writes, "My aim is: to teach you to pass from a piece of disguised nonsense to something that is patent nonsense" (PI  §464, 1958 version). 27In this way, the later Wittgenstein's method in the rule-following dialectic is similar to early Wittgenstein's in the Tractatus Logico-Philosophicus, described in §6.54 as follows: "My propositions are elucidatory in this way: he who understands me finally recognizes them as senseless, when he has climbed out through them, on them, over them.(He must so to speak throw away the ladder, after he has climbed up on it.)He must surmount these propositions; then he sees the world rightly." We can now see that Wittgenstein is not speaking hyperbolically, but instead quite literally, when he writes that investigations such as these do not discover falsehood at the heart of a philosophical investigation, but rather that the "results of philosophy are the discovery of some piece of plain nonsense" (PI §119).Instead of attempting to solve the problem of rule-following by finding a false premise or invalid inference to jettison from our thinking on the topic, Wittgenstein's strategy is of an entirely different sort.We are to eventually come to realize something we are incapable of realizing when the dialectic is first entered: that, when we initiate the inquiry, we fail to assign any meaning to certain of our words and sentences, an insight that only becomes apparent to us once we have seen that every direct response to the problem of rule-following issues -and must issue -in a paradoxical dead-end.By showing where it leads, Wittgenstein seeks to demonstrate first that we fail to mean anything at the end of the rule-following dialectic when we reach its "conclusion", and then also that we fail to say anything meaningful as we traverse even its earlier -and at first seemingly more intelligible -stages.It ultimately turns out that, in negotiating the entirety of the dialectic, we merely participate in the illusion that the sentences we speak as part of it are meaningful.
The emptiness of our words is therefore something that will only strike us when we regard the investigation of rule-following retrospectively.For, it is only after the dialectic has been fully worked through, when it has become manifestly paradoxical and we have seen that all our direct rejoinders to it lead nowhere in the end, that we are able to uncover its source in the "piece of plain nonsense" that lies hidden at the start of the inquiry, at the point of its initial framing.It is then that we finally discover that nonsense does not break out and enter the dialectic at some late stage, but is instead there from the beginning. 28The very set-up of the investigation, itself being senseless, is the real culprit of its manifestly paradoxical result.
At the outset of the investigation -and in our subsequent responses to the challenges therein raised -we try to make sense of a too-simplistic account of our capacity for understanding.Wittgenstein seeks to show how we can neither take certain recondite exercises of our capacity for understanding -e.g., our ability to recognize a sign as a sign or our ability to interpret -to be primitive exercises that are intelligible as self-standing capacities, nor build up an account of following a rule from these ostensibly more basic capacities.Instead of seeing our capacity to interpret signs as the most basic expression of our capacity for understanding them, Wittgenstein aims to demonstrate how our capacities for interpretation and for recognizing signs as signs presuppose a more basic form of exercising our capacity for understanding.The various capacities we seek to appeal to in the rule-following dialectic -the capacity to interpret, to recognize a sign as a sign, to recognize a particular occurrence of a sign as one in which it is merely a dead sign, to agree in our use of a sign, et cetera -are all parts of a single complex capacity whose most basic exercise is one in which we grasp a rule or the meaning of a sign without interpretation.If we lack this capacity, then, of course, we can understand nothing.If we have it, then we can understand things and, in some cases, understand them immediately (i.e., without interpretation or supplemental normative construal).
By following the rule-following dialectic to the point of paradox, we can see the necessity of a kind of return, a need to reflect on its origin.Indeed, this is what brings us to the second part of Wittgenstein's treatment of the problem of rule-following in which he investigates the enormity of what we overlooked in our original framing of the problem (though a discussion of this part of Wittgenstein's treatment would take us well beyond the confines of this essay). 29Thus, when we reach the end of §201, we have not 29 Wittgenstein elucidates our capacity for understanding and the way in which it is tied to other human capacities -e.g., our capacity to partake in a common practice -in other parts of PI.Indeed, starting at §202 -i.e., immediately after the paradox of §201 -and continuing to approximately §242, Wittgenstein undertakes an investigation of these reached the end of Wittgenstein's treatment of rule-following -it is instead merely the point at which it is supposed to become clear to us that a wrong turn was taken somewhere earlier in the dialectic.We therefore mistake the spirit of the concluding remarks of that section if we seek in them Wittgenstein's "solution" to the problem of rule-following.What he there offers are instead reminders of certain truisms that, in the light of our earlier wrong turn, have come to seem paradoxical.
One such example from §201 is this: That there is a misunderstanding here is shown by the mere fact that in this chain of reasoning we place one interpretation behind another, as if each one contented us at least for a moment, until we thought of yet another lying behind it.(PI §201) This remark is meant to make vivid that, when we call upon interpretations in ordinary contexts, it is only ever to clarify what is to be done or understood in the face of some unclarity in the situation.When, however, we take every situation to be of such a sort, we unmoor the concept of interpretation from the languagegame in which it is at home and turn it into something mysterious.We thereby make it seem as if an interpretation performs some extraordinary feat in a manner that we can no longer comprehend.
Wittgenstein's final piece of advice in §201 takes the form of a suggestion.He writes that "one should speak of interpretation only when one expression of a rule is substituted for another" (PI §201) 30 -that is, one should speak of interpretation when, in topics.Arguably, one of the reasons for the popularity of the communitarian interpretation of Wittgenstein is that it notices all of these appeals to practice, institution, custom, and so on, that come after §201 but then mistakes Wittgenstein's appeals to them for an attempt to answer the problem of rule-following head-on.These sections are part of an investigation aimed at understanding our form of life and the intricate fabric from which it is woven (complexities which have been set aside as irrelevant in the platonist and communitarian attempts to answer the rule-following paradox head-on); and while before §201 Wittgenstein considers various direct responses to the problem of rulefollowing, after §201 Wittgenstein's focus is instead on elucidating our single complex capacity for understanding and the place it occupies in the weave of our lives.For two commentators who take up § §202-242 in the spirit in which I am urging they be treated, see Cora Diamond's "Rules: Looking in the Right Place" (1989) and Stroud's " Wittgenstein and Logical Necessity" (1965). 30See also BBB 3. everyday life, we actually do speak of interpretation.If we were to so restrict our use of the term, then interpretation would no longer seem to be something that takes us from dead signs to living ones, i.e., from a total non-employment of signs to an employment of them.Rather, it would only be something that clarifies one employment of signs in light of another, and that would not be so mysterious at all. 31 every interpretation hangs in the air together with what it interprets, and cannot give it any support.Interpretations by themselves do not determine meaning.(PI §198) 8