From the WAB archives: A selection from the Bergen Wittgenstein archives' Working Papers and audio-visual materials

1 Introduction

Literary critics enjoy working on what are known as cruces: features or parts of texts especially hard to agree upon, bones of contention. In the English speaking world two famous ones are found in editions of Shakespeare’s sonnets: who is The Dark Lady and who hides behind the initials WH? In Wittgenstein’s corpus two stand out: in the Tractatus the role of the preface and closing paragraphs, in the Philosophical Investigations the so-called paradox about meaning and understanding in § 201. The debate about this latter crux increased sharply in volume after Saul Kripke published first his article and then his book twenty five years ago.¹ The majority of Wittgenstein scholars attacked Kripke, and with an odd fervour: he had forgotten the second part of § 201, the part where Wittgenstein shows that there is no paradox, that he is not a sceptic, etc. My purpose here is not to try and settle what Wittgenstein’s final position may look like, I have set myself two lesser tasks. The first is to find in earlier texts material relevant for the PI discussion, so we can see how Wittgenstein, over a long period, struggled with the problem; the second is to see how some critics of Kripke themselves do not pay close attention to § 201. The combined results, I believe, shift some ground in the debate between Kripke and a number of his critics.

§ 201 is part of a discussion in the PI that gathers pace with § 138, but clearly extends back to at least §§ 81-2, if not all the way back to § 19, and whose aim is to get a proper focus on meaning and understanding; meaning in both its structures, that is, both as in “the person x means…” and in “the expression s means…” § 201 itself is the direct conclusion to a debate stated crisply in § 198. There is good reason to consider § 198 the important paragraph. But the discussion extends beyond § 201, indeed well beyond PI, to several other texts by Wittgenstein.

The idea Wittgenstein combats is this: Understanding must always be linked to interpretation, the latter working as something coming between the mere expression and the meaningful use of it, interpretation as creating, or identifying, the relevant, active meaning. Wittgenstein can be seen as attacking the idea that in the use, and understanding, of a verbal expression we must interpret it, i.e. commit ourselves to a particular interpretative action, and that through this action we have given the verbal expression a meaning which shows how to understand it in that particular use. For Wittgenstein the combatting of this idea, under its guise of rule following, was an old preoccupation. We find him discussing it in several places, in texts valuable for understanding the relevant parts of PI. We shall shortly come to (some of) those other pieces of work.

Given the state of Wittgenstein’s writings, it is tempting, when trying to interpret him, to find sentences from here, there, and everywhere, to go passage hunting in the Nachlass jungle, in the words of Hanjo Glock. That, to my mind, would be a bad idea. Isolated remarks show us what he once wrote, but, unless they are put together by him, connected in a text, they do not show the conclusions to his thinking, what he means. It further settles the importance of his ideas on a topic if we find in his writings an enduring interest in that topic. I shall restrict myself to works published on paper.

Critics of Kripke says he took into account only the first part of § 201, viz:

“This was our paradox: no course of action could be determined by a rule, because every course of action can be made out to accord with the rule. The answer was: if any [course of] action can be made out to accord with the rule, then it can also be made out to conflict with it. And so there would be neither accord nor conflict here.” ²

2 Analysis of part one

First sentence: “This was our paradox: no course of action could be determined by a rule, because any course of action can be made out to accord with the rule.”

Paradox, first possibility: “No course of action could [can] be determined by a rule.”

Paradox, second possibility: whole sentence after colon.

Second sentence: “The answer was: if any [course of] action can be made out to accord with the rule, then it can also be made out to conflict with it.”

“Answer”: an intended denial is coming up of the paradox in the first sentence.

The second sentence itself, after colon, is a conditional. If accord and conflict were formally contradictory, then its form would be “if p then not p”, from this we first (mediately) conclude “if p then both p and not p”. Since “p and not p” is a contradiction, we finally conclude “not p”. However, conflict and accord can probably not be treated that way (see below).

This is a very difficult sentence (to put it mildly), since it probably plays on the presence of incompatible rules created by different understandings of the rule: however, no one rule is such that a course of action both accords with and is in conflict with that rule, on one and the same interpretation of the rule. The sentence should probably read: if a rule can be so interpreted that a given [course of] action will accord with it, then the rule can also be so (differently) interpreted that the [course of] action will conflict with it.

Third sentence: “And so there would be neither accord nor conflict here.”

This seems to draw a conclusion from the second sentence. The mediate conclusion we draw is: “if any [course of] action can be made out to accord with the rule, then it can be made out both to accord and to conflict with the rule”. The consequent of this mediate conclusion is a conjunction “it [sc. any course of action] can be made both to accord with the rule and to conflict with it”. The reasoning from the whole of the mediate conclusion to the third sentence as a final conclusion does not go through unless we add two assumptions. The first is that the consequent of the mediate conclusion is considered an impossibility, even if it is not a formal contradiction. The second assumption, actually a second premiss, is: if any [course of] action can be made out to conflict with the rule, then it can also be made out to accord with it. This gives a mediate conclusion with the same consequent as that of the mediate conclusion drawn from the second sentence. The two mediate conclusions tie accord and conflict together: we cannot have the one without the other. If there is accord or conflict, then there is both accord and conflict. In order to reach the third sentence we deny the intolerable consequent of this sentence in italics, this gets us the denial of the antecedent, which denial is the third sentence. It is this third sentence which then, in part 2, is seen as unacceptable, thus leading to the rejection of the paradox. It is worth noting that in order to reject the paradox, Wittgenstein must see accordance with a rule as a necessary component of being determined by a rule. And, outside the specific interpretational framework, we do not get the second sentence.³

We have three sentences, whose relationships are not easy to grasp. The first is an explanation, giving a reason for the conclusion that actions are not determined by rules, which conclusion is perhaps the paradox, since it contradicts common beliefs, if rightly understood. An alternative, which seems syntactically correct, is that the whole first sentence, after the colon, is the paradox. The second sentence is a conditional, with the reason from the first sentence as its antecedent. The third sentence is a categorical proposition which draws a conclusion from the second. This conclusion denies the truth of the reason given in the first. So, presumably, the conclusion of the first sentence has not been grounded in a truth. Therefore, that we have a paradox has not been successfully argued for. Seeing the last two sentences as a rebuttal of the first makes sense of the opening of the second part of § 201, which then is a criticism of the rebuttal. That rebuttal is anyway feeble: the third sentence is as much a reason for the conclusion in the first sentence as is the reason actually given there. One difficulty I have with an analysis that places the paradox in the first sentence is that the conditional is the closest we have as paradoxical, both in itself and in its inferential products. As the substantial part of the attempted rebuttal, it is the engine producing the paradox given out in the first sentence. In my further remarks nothing shall be made of the fact that I find this first part difficult, and that I am not at all certain as to how it should be analyzed. These problems should not impugn any of the positions I take.

Kripke’s great merit was not the main structural idea, that he got from Nelson Goodman, but the fact that he employed it in this context.

What did Goodman do? His project was to seek an analysis that justified inductive reasoning, conceived as the formation of general beliefs, or statements, on the basis of particular judgements of the form “This x is F”. The reasoning would then go: “This x is F, also that x and that one and.., so all of x-kind are F”.

Goodman described the attempt at justification as a failure, because any set of particular judgements would justify, to an equal extent, any number of mutually inconsistent general statements.

The machinery Goodman employed was ingenious. He introduced an imagined historical process of delivering a series of particular judgments, say “This emerald is green”, and in this process we cut a temporal point, let us call it a “node”. A node then selects a set of particular judgements, which set can play the role of evidential basis for a general belief. In our example, let us say that all emeralds examined so far were judged to be green, so they are similar in that respect. In that situation, mark the moment as our node: can we use the evidential basis thus shaped to justify advancing the general conclusion that emeralds are green? No, said Goodman, and for the following reason. Let us introduce a predicate “grue”: objects are grue just in case they are either examined before the node and found to be green, or examined after the node and found to be blue. This permits us to say that an emerald is grue if examined after the node and found to be blue. (Goodman’s original formulation draws a different distinction from the one I give here.) The conclusion Goodman draws is that the evidential basis in our example supports to an equal degree the inconsistent hypotheses that all emeralds are green and that all emeralds are grue. (And to an equal degree the mutually exclusive predictions that the emerald next examined is green and that it is blue.)

The general lesson so far is that by putting nodes where we want, we can always construe inconsistent general statements, all equally supported by the observed facts.⁴

The main point here is that judgements about particular cases settle whether the specimen objects are similar, or not. Blue and green, grue and bleen, they all sort objects into classes of partially identical objects. Since the deep underlying puzzle in Wittgenstein’s discussion of the paradox is the fact that conceptual deployment relies on, or establishes, objectual similarity, Goodman’s formal point is of great interest. Another reminder Goodman gives us is that though much of Wittgenstein’s discussions are directly focussed on rule-following, the underlying issue is of meaning, how to conceive the meaning of linguistic expressions.

This is the main idea Kripke takes from Goodman, only with abstract examples: number series. But the idea of node works equally well, if not better, with number series. The reason is that with sense properties such as green and blue, grue and bleen, there is an issue about naturalness and observability. That issue is not active with algebraic operations on natural numbers, where the worry would be about the complexity of algebraic equations. I would say that Kripke’s employment of Goodman’s idea in the analysis of the first part of § 201 is brilliant, – the more so because he could have taken most of it from Wittgenstein himself. I shall come to that in a moment. A difference between Kripke and Goodman is that while Goodman looks at judgements, Kripke looks at rule-governed actions: additions are actions done according to the rule of addition, but so are quadditions actions done according to the rule of quaddition. I think, with reservations, that Kripke’s reading of this part of the text is correct. When people blame him for forgetting the rest of § 201 I believe he simply didn’t see it as providing a solution to what I think was his initial problem: on what basis can we tell whether we add or quadd? Kripke goes on to claim that there are no discoverable facts deciding one way or the other. So, the issue is not epistemological, but metaphysical, as some like to say. I shall not discuss that distinction.

I said above that the topic of rule-following was of long standing in Wittgenstein’s oeuvre, we find discussions in The Blue Book, The Brown Book, and theBig Typescript. There are also remarks of value in Remarks on the Foundations of Mathematics. The first three texts are of a particular interest: The Blue Book from 1933–4 was virtually a publication, so it must carry the weight such do. Likewise, but to a smaller extent, with The Brown Book from 1934-5. (An additional attraction of these texts is that they are, since not broken up into short remarks, more accessible introductions to Wittgenstein.) Big Typescript, from the same period as The Blue Book, 1933, was revised by Wittgenstein up until 1937, according to the editors of the scholarly edition. It thus postdates the first draft of PI from late 1936-37, which draft however does not contain § 198.⁵ PI § 198 seems to be from 1944, with § 201 somewhat later. So, there is a gap of about six or seven years between those remarks in PI and previous ones.⁶

That Wittgenstein’s thinking supports Kripke’s reading of the first part of § 201 is brought out in passages from The Blue Book, page 12-4:

“We must distinguish between what one might call “a process being in accordance with a rule”, and, “a process involving a rule” (in the above sense)”

Wittgenstein has just explained the above sense as what we have when the expression of a rule forms part of the processes of understanding, obeying, etc. He then continues by giving an arithmetical example where the calculations are in accordance with the rule of squaring, but “also in accordance with any number of other rules”. In order to say that a rule was involved the symbol of the rule has to form a part of the calculation.

“Some one teaches me to square cardinal numbers, he writes down the row
     1    2    3    4,
and asks me to square them. (I will, in this case again, replace any processes happening ‘in the mind’ by processes of calculation on the paper.) Suppose, underneath the first row of numbers, I then write:
     1    4    9    16.
What I wrote is in accordance with the general rule of squaring; but it obviously is also in accordance with any number of other rules; and among these it is not more in accordance with one than with another. In the sense in which before we talked about a rule being involved in a process, no rule was involved in this. Supposing that in order to get to my results I calculated 1x1, 2x2, 3x3, 4x4 (that is, in this case wrote down the calculations); these would again be in accordance with any number of rules. Supposing, on the other hand, in order to get my results I had written down what you may call “the rule of squaring”, say algebraically. In this case this rule was involved in a sense in which no other rule was.

We shall say that the rule is involved in the understanding, obeying, etc., if, as I should like to express it, the symbol of the rule forms part of the calculation. (As we are not interested in where the processes of thinking, calculating, take place, we can for our purposes imagine the calculations being done entirely on paper. We are not concerned with the difference: internal, external.)”

Another major point comes out here:

“Let us go back to the statement that thinking essentially consists in operating with signs. My point was that it is liable to mislead us if we say “thinking is a mental activity”.”

The question what kind of an activity thinking is is analogous to this:

“Where does thinking take place?” We can answer: on paper, in our head, in the mind.

None of these statements of locality give the locality of thinking……It is correct to say that thinking is an activity of our writing hand, of our larynx, of our head, and of our mind, so long as we understand the grammar of these statements.” (pp. 15-6)

Just above, he had said this: “We are not concerned with the difference: internal, external”. This ‘externalization’ of the mental is a dominant feature in his thinking, we find remarks to that effect in other places. When he later says:

“The meaning [of a phrase] is not a mental accompaniment to the expression” (The Blue Book p. 65),

I see that as telling us two things. Firstly, it prefigures the remark about nothing coming between the sign and the fact, and secondly, the way to discuss matters of meaning is to locate meaningfulness in external features of meaningful activity.

As this shows, Kripke could have got his main idea from The Blue Book. This makes it sound as if we have three texts, Kripke’s, Goodman’s and The Blue Book, and that the paradox of § 201 fits straight into those texts. More exactly, that Kripke’s use of Goodman fits The Blue Book, and that this latter fully explains the first part of § 201. But is that true? The Blue Book introduces the distinction between involvement and accordance, and treats it as a crucial difference, in that rule-involvement will stop the regress of interpretations. It is hard to see how Kripke can take involvement to achieve that. We cannot justify a further development of a number series by referring to the algebraic equation we supposedly used because all we have is the way we interpreted that equation in the past, basically all we have is an evidential basis left behind. So Kripke must deny the importance of Wittgenstein’s distinction in The Blue Book. But Kripke’s set-up of the problem, with his employment of “my external symbolic representation”, on page 7 of his book, is almost straight out of Wittgenstein’s The Blue Book notion of how involvement is present. (However, Kripke’s “my internal mental representation” needs to be handled with greater care.) While Kripke can have no use for The Blue Book notion of involvement what of Wittgenstein himself, the author of it? Does he keep the distinction in later works?

In The Brown Book, page 112-4 we find important changes compared to what we have been discussing. Here is a long quotation:

“[T]he row A wrote down was 2, 4, 6, 8. B looks at it, and says “Of course I can go on”, and continues the series of even numbers. Or he says nothing, and just goes on. Perhaps when looking at the row 2, 4, 6, 8 which A had written down, he had some sensation, or sensations, often accompanying such words as “That’s easy!” A sensation of this kind is for instance, the experience of a slight, quick intake of breath, What one might call a slight start. Now, should we say that the proposition “B can continue the series”, means that one of the occurrences just described takes place? Isn’t it clear that the statement “B can continue…” is not the same as the statement that the formula an = n2 + n – 1 comes into B’s mind? This occurrence might have been all that actually took place…..If a parrot had uttered the formula, we should not have said that he could continue the series. – Therefore, we are inclined to say “to be able to …” must mean more than just uttering the formula – and in fact more than any one of the the occurrences we have described. And this, we go on, shows that saying the formula was only a symptom of B’s being able to go on, and that it was not the ability of going on itself. Now what is misleading in this is that we seem to intimate that there is one peculiar activity, process, or state called “being able to go on” which somehow is hidden from our eyes but manifests itself in those occurrences which we call symptoms…It is true, “B can continue…” is not the same as to say “B says the formula…”, but it doesn’t follow from this that the expression “B can continue…” refers to an activity other than that of saying the formula, in the way in which “B says the formula” refers to the well-known activity…..To say the phrase “B can continue...” is correctly used when prompted by such occurrences as described in 62), 63), 64) but that these occurrrences justify its use only under certain circumstances (e.g. when experience has shown certain connections) is not to say that the sentence “B can continue…” is short for the sentence which describes all these circumstances, i.e. the whole situation which is the background of our game……..We can say: The expression “B can continue the series” is used under different circumstances to make different distinctions.”

The most important point coming out of these passages is that the presence, in the process, of the expression of a rule does not necessarily mean that the rule is involved, in the sense of The Blue Book, the presence may serve other purposes. Also, but less clearly advocated, the absence of the expression of the rule does not mean that the rule is not involved. Said otherwise: in order to follow a rule it is not necessary that the expression of the rule is present in the activity.

On this second point, whether the presence of the rule is required, the text does not strike me as clear when it says that more than mere presence is necessary. So, my reading of Wittgenstein’s p. 113, may be disputed with regard to this second claim. Let me therefore leave the story for a moment to make these remarks: Wittgenstein does not say that the presence of the rule formula in the process in unnecessary. But, we may speculate on the matter. To claim that its presence is not required, would be to put into question what he means by a process of calculating. I believe Wittgenstein did speculate precisely on that matter: is the presence of the rule formula necessary, what sort of process do we have in cases of understanding? And, if we see such speculation expressed in e.g. the example in PI § 506, of the man told to turn right, then it started early, since the example is also found in the older part of theBig Typescript (see p. 16). In one sense, it is obvious that the rule needs to be present, the rule must determine the outcome. But, remember, we are here talking about the rule formula. Wittgenstein’s final position is the common sense one that the rule must be present, but not as a formula, among the formulas of the calculating process.

However, the first point, about the insuffiency of the presence of the rule, is the important one. One consequence of this change is that the involvement – accordance distinction from The Blue Book has been dropped. Wittgenstein needs another way of accounting for how actions are rule-governed. The focus is on being able to go on. A point stressed is that we are misled into thinking that there must be a process hidden from our eyes called “being able to go on”. Behind the formula may be the circumstances under which it is uttered.

Now, coming up to PI, Wittgenstein introduces a variant of the distinction from The Blue Book. We now have a difference between a rule determining a course of action and a course of action being in accordance with the rule. He also uses the expression “to obey a rule”. This must be squared with the main points where The Brown Book disagrees with The Blue Book.

It would be laughable to suggest that Kripke had read only this part of 201. But he worked with that first part and obviously must have found the second part unilluminating with respect to what the first set up. Personally, I find Kripkes work valuable, while also agreeing with those critics of his who claimed he misinterpreted Wittgenstein’s thinking and indeed missed his whole philosophy. For that, see Cora Diamond’s essay about looking for rules in the right place.⁷ Here is an extract from Remarks on the Foundations of Mathematics, p. 414:

“But how then does the teacher interpret the rule for the pupil? (For he is certainly supposed to give it a particular interpretation.) – Well, how but by means of words and training?”

And if the pupil reacts to it thus and thus; he possesses the rule inwardly.

But this is important, namely that this reaction, which is our guarantee of understanding, presupposes as a surrounding particular circumstances, particular forms of life and speech. (As there is no such thing as a facial expression without a face.)

This, in a nutshell, is close to what Diamond claims Kripke missed about the place of rules.

3 Criticism of Kripke

Critics of Kripke blame him for forgetting this second part of § 201:

“It can be seen that there is a misunderstanding here from the mere fact that in the course of our argument we give one interpretation after another; as if each one contented us at least for a moment, until we thought of another standing behind it.

What this shews is that there is a way of grasping a rule which is not an interpretation, but which is exhibited in what we call “obeying the rule” and “going against it” in actual cases.”

The misunderstanding referred to seems to be the assumption shared by the paradox and the attempted rebuttal. It is clear throughout Wittgenstein’s texts that the threat to be removed is that of loss of meaning and impossibility of following rules. Since the concept of interpretation underpins the threat, Wittgenstein needs to detach understanding, or grasping meaning, from interpretation, that much is pretty clear. (It is worth noting that something Wittgenstein claims here is that (merely) presenting in thought the regress makes us aware that there is another way of grasping a rule than by interpreting it: “What this shews”!)

4 Criticism of some critics of Kripke

Now finally, to some critics of Kripke. I shall take a quick look at the way three of them handle § 201: the three are David Pears, Peter Winch and John McDowell.

My general complaint about all three critics is that they ignore this, the third part of § 201:

“Hence there is an inclination to say: every action according to the rule is an interpretation. But we ought to restrict the term “interpretation” to the substitution of one expression of the rule for another.”

(Coming on the heels of the last sentence of part two, I find it difficult to make sense of the “hence”.) There are two ways of understanding this injunction, the first is: we, perhaps including Wittgenstein, have not, but should, read “interpretation” that way. Such a reading is difficult to accept. Since that conception of interpretation is behind the paradox, how could it not have been in use? And, if it were so, that it had not been in use, why ought we adopt it? Would Wittgenstein be telling us to stay with the paradox? That makes nonsense of the fact that he is here involved in an attempt to avoid the paradox. The second way is that Wittgenstein himself uses the word that way, and claims that, in such contexts, it is the correct way. Given his long standing practice of so using the word, I am compelled to read his remark as telling us what an interpretation is in such contexts. Wittgenstein is not saying: look, a mistake has been made and a destructive regress has been created; the villain is the employment of a notion of interpretation, drop any such and the mistake is rectified. The mistake was made employing the specific notion of interpretation Wittgenstein is telling us we have to use in such contexts, and which was used to generate the regress. If one does not heed what Wittgenstein says here, and feels free to employ another concept of interpretation, then it is hard to see how one can understand what he means by “way of grasping a rule which is not an interpretation”.

Pears, one author who does register the third part of § 201, has this to say:

“An ‘interpretation’ is an analysis expressing the meaning of the original sentence in different words.” (The False Prison, footnote p. 467)

Note the scare quotes round “interpretation”, Pears does not take seriously what Wittgenstein says. However, contrary to Pears’ contention: the rule expression itself is an expression of meaning, the interpretation, being another expression, most likely is an expression of another meaning, however much its proponent believes otherwise. The replacement of one interpretation by another replaces one meaning by another, thus creating a sequence of connected meanings. (We may construe the relationship among the members of such a sequence as constituted by the fact that each produces a segment (evidential basis) of one and the same path, which segment includes the previous one, and extends it by picking a choice offered at a node; a sequence thus develops as one among the different possible paths in the web of nodes.) This is the regress problem that threatens the existence of meaning, and which Wittgenstein diagnoses as showing the misunderstanding. If Pears was correct, there would first of all be no relevant regress, and we may ask: what would we achieve by seeking yet another expression of the meaning? The answer to that question is provided by another simple question and its answer: if one expression doesn’t satisfy us, give us the meaning, how can any?⁸ The answer is obvious: none will. (This as a general, theoretical, claim: empirically we often find that rephrasing makes the original point understandable to the audience.) An intolerable outcome, but not Wittgenstein’s target. It can however be argued that Pears sees through to the conclusion, a regress will be generated by absence of rule-determination, that is, by lack of meaning.

Winch, in his essay on Kripke “Facts and superfacts”, does not quote the third part of § 201, but inserts his own, or rather what he sees as Kripke’s, gloss of “interpretation” as “representation”. (Kripke, on page 7 of his book, when laying out what he sees as the problem, uses two phrases, quoted earlier, “my external symbolic representation” and “my internal mental representation”.) This gloss Winch gives in his insistence that Kripke’s formulation of the issue commits the mistake Wittgenstein argues against in the second part. (Quoted from Trying to make sense, p. 58) ⁹ The gloss is introduced when Winch claims that Kripke misunderstands this second part of § 201. If Winch had paid heed to the third part of § 201, and how the point stated there fits the way Wittgenstein generates the regress, I think he would have had to say something else about Kripke’s misunderstanding. Winch would have needed to clarify whether Kripke, on page 7 of his book, wrote about the first or the second part of of § 201. If Kripke addressed the first part, what then? After all, Wittgenstein does not misunderstand himself when setting out the premisses for the paradox. (Winch himself says that Kripke was identifying the issue when using the expression.) One may say this without raising the issue of whether “representation” is a suitable gloss of Wittgenstein’s use of “interpretation”. Actually, it is not clear what charge Winch levels against Kripke. If Winch is happy with rewriting the second part of § 201 with “representation” in place for “interpretation”, then the charge is simply that Kripke failed to pick up Wittgenstein’s claim about the grasping of meaning without interpreting (= representing). That is not a very likely charge against Kripke; on the other hand, if Winch objected to Kripke’s gloss, it is strange he adopted it without comment. Here is a fuller extract from Winch:

“To convince ourselves that Wittgenstein is not attacking a straw man we need look no further than the passage from page 7 of Kripke’s book that I quoted earlier, where it is said that I grasp a rule ‘by means of my external symbolic representation and my internal mental representation’. Here, in the very identification of what is under discussion, we find the misunderstanding Wittgenstein characterizes by saying that ‘we give one interpretation [read ‘representation’] after another; as if each contented us at least for a moment, until we thought of yet another standing behind it (section 201).”
(Trying to make sense, pp. 57-8)

Looking at McDowell’s essay “Meaning and Intentionality in Wittgenstein’s Later Philosophy”¹⁰ a few features are striking. McDowell uses Crispin Wright as a foil, and the discussion is carried out on Wright’s turf. This means that McDowell ‘internalizes’ the outer expression, and the arguments against Wright and Kripke¹¹ concentrate on a conception of mental objects as static, lifeless, inert. Thus they mimic lifeless external expressions, dead signs. By going along with this set-up McDowell reverses Wittgenstein’s move from the inner to the outer. I suppose he may use Wittgenstein’s own reason: we are not concerned with finding a place where thinking takes place. But Wittgenstein made his move in order to get a proper understanding of the inner and how it works. He doesn’t operate with a false picture of the inner, such as the one McDowell starts off with and then leaves behind in his debate with Kripke and Wright. Alone among the three critics I cannot see that McDowell either quotes or raises as an issue Wittgenstein’s conception of interpretation. It is strange to read McDowell writing about interpreting (dead) items in the mind, or interpreting meanings, and, tellingly, writing about mental content. Wittgenstein does not invite us to think about the threatening regress that way, nor find the way to avoid it by starting out thus.

To put it very bluntly: what Pears, Winch and McDowell don’t seem to realize by ignoring Wittgenstein’s instruction about what “interpretation” means, is that Wittgenstein has long since left behind their opinions, whatever they are, about what interpretation is in the relevant contexts. Wittgenstein’s problem is to find out how meaning is possible, given that there is nothing between the sign and the fact. We have to give up what is said in § 94:

“The tendency to assume a pure intermediary between the propositional signs and the facts.”

Since the meaning does not come between the sign and the facts, that is the world, where, and what, is it? There is no occasion to discuss that issue now, but if I were to look for relevant texts, here are some relevant ones from PI:

§ 432 “Every sign by itself seems dead. What gives it life? –In use it is alive. Is life breathed into it there? – Or is the use its life?”

§ 503 “If I give anyone an order I feel it to be quite enough to give him signs. And I should never say: this is only words, and I have got to get behind the words…”

§ 504 “But if you say: “How am I to know what he means, when I see nothing but the signs he gives?” then I say: “How is he to know what he means, when he has nothing but the signs either?””

We do not interpret signs, in that special sense, we use them and react to them.