Jump to ContentJump to Main Navigation
The Practical TurnPragmatism in Britain in the Long Twentieth Century$

Cheryl Misak and Huw Price

Print publication date: 2017

Print ISBN-13: 9780197266168

Published to British Academy Scholarship Online: May 2018

DOI: 10.5871/bacad/9780197266168.001.0001

Show Summary Details
Page of

PRINTED FROM BRITISH ACADEMY SCHOLARSHIP ONLINE (www.britishacademy.universitypressscholarship.com). (c) Copyright British Academy, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in BASO for personal use. Subscriber: null; date: 08 April 2020

The ‘Middle’ Wittgenstein (and the ‘Later’ Ramsey) on the Pragmatist Conception of Truth

The ‘Middle’ Wittgenstein (and the ‘Later’ Ramsey) on the Pragmatist Conception of Truth

(p.29) 3 The ‘Middle’ Wittgenstein (and the ‘Later’ Ramsey) on the Pragmatist Conception of Truth1
The Practical Turn

Anna Boncompagni

British Academy

Abstract and Keywords

This chapter examines some remarks Ludwig Wittgenstein expressed on pragmatism in manuscripts and lectures during the first half of the 1930s. These remarks focus principally on the Jamesian conception of truth, according to which, roughly, a belief or a proposition is true if it is useful. Wittgenstein acknowledges that this conception is able to capture some characters of ordinary language, but at the same time, he criticizes some aspects of it, and his criticism strongly resembles Frank Ramsey’s attitude towards the same topics. In this sense, it is argued that Ramsey had a role both for Wittgenstein approaching pragmatism, and for the partly negative attitude with which he came to judge it. Yet, the two thinkers’ general perspectives diverge when it comes to the place of theory in philosophical activity.

Keywords:   Wittgenstein, Ramsey, William James, pragmatism, truth, ordinary language, usefulness


THIS CHAPTER OFFERS A BRIEF SUMMARY of what Wittgenstein had to say about pragmatism in his so-called ‘middle’ period, in the first half of the 1930s. As it will turn out, what he had to say focused on the pragmatist conception of truth, and particularly on William James’s version of it.

In what follows I will not dwell upon theories of truth, or on Wittgenstein’s general interpretation of pragmatism. In touching on these themes, my aim is limited to the highlighting of the relevance of Frank Ramsey both for Wittgenstein approaching the pragmatist notion of truth, and for the critical attitude with which he came to judge it.

The structure of the chapter rotates around four quotes from Wittgenstein’s manuscripts and lectures. In Quote I, 1930, from MSW 107,2 the subject of pragmatism appears for the first time in Wittgenstein’s writings. Quote I, 1932, contains a sharp criticism of pragmatism, which Wittgenstein expresses through a joke about how one might win a lottery. He makes a similar point during a lecture in 1935, Quote I, with an example about a mad physicist’s hypothesis. Both of these two remarks, in their content and in their irony, closely resemble Ramsey’s criticism of the Jamesian conception of truth. Quote IV, from a lecture or a conversation dating back to 1932, will ‘teach us differences’, as Wittgenstein (with Shakespeare) would put it, between the two thinkers’ general perspectives.

(p.30) Immediate experience and the hypothetical character of ordinary sentences

Wittgenstein’s and Ramsey’s approaches to truth intersect each other in many respects. One of them is the idea of redundancy; another one, strictly connected to the former, is the priority of the problem of sense over the problem of truth (see Glock, 2005: 58; 2013; Sullivan, 2005; Rumfitt, 2011). Yet, we shall leave these general themes aside, as this chapter investigates the matter from the point of view of the two thinkers’ relationship with pragmatism.

The first time Wittgenstein mentions pragmatism in his writings is on 20 January 1930, which happens to be the day after his friend Ramsey’s premature death. A couple of days before, Wittgenstein had noticed that his ‘previous entire conception of the proposition’ should be ‘rotated by a small angle’ (MSW 107: 247, 18 Jan. 1930). It is by working on this ‘small angle’ that Wittgenstein will consider the possibility of a pragmatist account of truth.

Before examining the passage, let me briefly summarize Wittgenstein’s attitude towards his ‘previous conception’, as it emerges from his writings at that time. In 1929, he is looking back at the Tractatus and reinterpreting it in a phenomenological way.3 According to this partly new point of view, a proposition, portraying a fact, is verified, and is therefore true, if the portrayed fact occurs in the flux of experience. A primary or phenomenological language consisting in propositions of this sort is ideally able to capture immediate experience. But Wittgenstein’s aspiration to a Tractarian-phenomenological language encounters significant difficulties. Ramsey highlights two of them. First, the claim that elementary propositions must be independent is untenable. I will not linger on this point, as it is not central for our purposes. Second, some generalizations, namely those concerning infinite cases, such as ‘All men are mortal’, cannot be accounted for in terms of conjunctions and disjunctions of propositions. Paralleling the ‘later’ Ramsey’s distinction between genuine generalizations and variable hypotheticals, Wittgenstein will distinguish between what he called ‘laziness dots’, through which for instance we represent the alphabet by the expression ‘A, B, C, …’, and the dots which cannot be replaced by any finite enumeration, as when we represent cardinal numbers by ‘1, 2, 3, …’ (Moore, 1993: 90), a case which in some respects is akin to ‘All people die before they are 200’ (example given in BT: 249). We will return to this issue shortly.

Another difficulty with the phenomenological perspective, only in part addressed by Ramsey (see NPPM: 55), is that although it may appear as the only language capable of catching immediate experience, it is in fact extremely complex and distant from ordinary language, primarily because, in the end, it does not (p.31) satisfy our needs. A phenomenological language is not useful. This is one of the main points that emerge in our first quote from Wittgenstein, to which we now turn.

(I1) When I say ‘There is a chair over there’, this sentence refers to a series of expectations. I believe I could go there, perceive the chair and sit on it, I believe it is made of wood and I expect it to have a certain hardness, inflammability etc. If some of these expectations are mistaken, I will see it as proof for retaining that there was no chair there. Here one sees the access [Zugang] to the pragmatist [pragmatistichen] conception of true and false. A sentence is true as long as it proves to be useful.

(MSW 107: 248)4

This interesting passage has unfortunately received scarce attention in literature.5 A sentence like ‘There is a chair over there’—Wittgenstein says—does not refer to the world of immediate perceptions, which is the object of what Wittgenstein elsewhere calls primary language, the language of the primary system, or the language of visual space. ‘There is a chair over there’ is an expression belonging to secondary language, the language of the secondary system, the language of physical space. This kind of sentence generates an expectation, which, in turn, is connected with ways of acting, in such a way that if the expectation is not fulfilled by the expected reactions of the objects, the sentence will be declared mistaken. It is interesting to see that Wittgenstein began to use the two expressions ‘das erste System’ and ‘das zweite System’ out of the blue, without defining them, in the first half of 1929 (MSW 105: 84).6 Someone who did introduce these expressions and define them was Ramsey, in his paper ‘Theories’ (T), written in 1929 as well.7 Ramsey and Wittgenstein were concerned with the same problems.

The theme of the relationship between immediate experience and the language of science had been at the centre of the debate in philosophy of science for a few decades, with thinkers such as Ernst Mach, Rudolf Carnap, and others taking part in it.8 (Carnap and Norman Campbell seem to have been particularly relevant (p.32) for the development of Ramsey’s perspective.9) The approaches of Wittgenstein and Ramsey do not completely overlap. In the first place, while Ramsey was working on the connection between primary language and the language of theories, Wittgenstein was working on the connection between primary language and ordinary language. This difference is not overwhelming: ordinary language and the language of theories are ultimately of the same sort, as they both refer to the world of physical objects or of physics, in opposition to the world of sense data (see Marion, 1998: 135). But there is another difficulty in comparing the two approaches. Ramsey invented a method thanks to which the two languages could be connected, and affirmed that, in the end, the secondary language is fictitious and could be dispensed with (although at the cost of high complexity). Wittgenstein eventually concluded that the perspective of a phenomenological language should be abandoned (see MSW 107: 205, 25 Nov. 1929 (PR: §1b); MSW 108: 29, 21 Dec. 1929 (PR: §53a); BT: 320; BBB: 45–8), and that the tension between the two alleged different worlds, the world of sense data and the world of physical objects, is already resolved within ordinary language. ‘There is no need of a theory to reconcile what we know about sense data and what we believe about physical objects’, he stated, ‘because part of what we mean by saying that a penny is round is that we see it as elliptical in such and such conditions’ (LEE: 69).

Yet, in spite of the differences, Ramsey’s and Wittgenstein’s reflections on these themes each bear signs of the presence of the other. One of the significant similarities is that both describe the secondary language as having a pragmatic aspect (MSW 106: 102–4; PR: §230a), that is, it is a more manageable mode of expression, for practical purposes, than the primary language, although it is only for the primary language that the notions of true and false propositions, strictly speaking, apply. This will be clearer if we proceed with the quote from MSW 107.

(I2) Every sentence we utter in everyday life appears to have the character of a hypothesis.

A hypothesis is a logical structure. That is, a symbol for which certain rules of representation hold.

The point of talking of sense data and immediate experience is that we are looking for a non-hypothetical representation.

But now it seems that the representation loses all its value if the hypothetical element is dropped, because then the proposition does not point to the future any more, but it is, as it were, self-satisfied and hence without any value.

(MSW 107: 249, partly in PR: §226a, 226b [modified translation])

(p.33) The non-hypothetical, hence certain representation, guaranteed by the language of sense data, has no value because it is of no use, and it is of no use because it does not point to the future; that is, it does not generate expectations. It captures immediate experience, but this is useless. On the other hand, a hypothesis has an instrumental or pragmatic value, as ‘the sense of a sentence is its purpose’ (MSW 107: 249; cf. MSW 107: 234; PR: §15c). But since it always remains open, it can never be completely verified. In this sense, as Wittgenstein concludes, ‘there’s no truth or falsity for it’ (MSW 107: 250, PR §226b).

Wittgenstein’s distinction between propositions and hypotheses, which is envisioned here, and particularly his subsequent characterization of hypotheses as ‘law[s] for forming propositions’ (MSW 107: 283, 4 Feb. 1930; PR: §228d),10 strongly resembles Ramsey’s characterization of variable hypotheticals or causal laws as ‘not judgments but rules for judging’, which ‘form the system with which the speaker meets the future’ (GPC: 149).11 (As noticed by many commentators, both these notions are most likely influenced by Hermann Weyl’s (1921) definition of the universal quantifiers as instructions for judgements.12) It is hard to avoid the impression that it was Ramsey’s criticism of the Tractatus’s account of generalizations, leading to the concept of the variable hypothetical,13 together with Ramsey’s distinction between primary and secondary systems, that inspired Wittgenstein’s thinking in these months. Even if we prefer not to speak of direct influences, we can be sure that on these issues as well as on other topics the two thinkers worked side by side, read the same authors, and debated extensively, although not always reaching the same conclusions. Since Wittgenstein’s first mention of pragmatism appears in this conceptual context, and given Ramsey’s acquaintance with pragmatism,14 we can safely conclude that pragmatism itself was a subject of debate between them.

Usefulness and the ‘eingreifen’ of beliefs and propositions

What did Wittgenstein mean, in Quote I, by ‘a sentence is true as long as it proves to be useful’, a formulation that he attributed to the pragmatist approach? First of (p.34) all, one must keep in mind that, like many of his contemporaries, Wittgenstein identifies pragmatism chiefly with James; hence the remarks he addresses to the pragmatist conception of truth are in fact addressed to the Jamesian position. In the context of the remark, his point is that truth is something thanks to which we can proceed with the minimum of impediments in our life: in this sense, truth is usefulness. To put it differently, if we expect something to be the case, and this something is the case, then our picture of what is the case works, and we use it properly, as a map, as it were.15 A belief is therefore true if we can handle it as an instrument in our practical activities, if it facilitates and does not obstruct the way we do things, if it guides our actions properly. Wittgenstein sometimes uses an interesting word to express this idea: the verb eingreifen, in its mechanical meaning of engaging, meshing, like cogwheels and gear wheels can engage or mesh.16 The absence of eingreifen is of course what characterizes a ‘wheel turning idly’, in Wittgenstein’s more famous example (PR: §1c; see also §12c and PI: §§132, 127). When a true belief eingreift, it engages us, we are moved, led, guided by it in the right way. In turn, when we believe a proposition, we make use of it, and we commit ourselves to use the same words with the same meaning in the future. In a similar vein, Ramsey described belief as ‘a disposition … to reassert the sentence or assert other sentences with the same meaning on suitable occasions’ (NPPM: 40, Apr. 1928). In Wittgenstein’s words:

‘I mean something by the proposition’ is similar in form to: ‘This proposition is useful’, or ‘This proposition engages with my life’.

BT: 5 (modified translation; orig. in MSW 110: 293, 5 July 1931).17

And in G. E. Moore’s rendering of Wittgenstein’s words:

One point on which he insisted several times … was that if a word which I use is to have meaning, I must ‘commit myself’ by its use. And he explained what he meant by this by saying ‘If I commit myself, that means that if I use, e.g., “green” in this case, I have to use it in others’, adding ‘If you commit yourself, there are consequences’.

(Moore, 1993: 52–3)

In this sense, properly, a true proposition or a true belief is what works, to use a typical Jamesian slogan. Although Wittgenstein, in a strict sense, is dealing not with truth but with meaning, in a broader sense I think that there is a phase in his thought in which he seems sympathetic to the idea of truth as usefulness, once usefulness is understood in terms of the eingreifen of words (PR: §15c; BT: 63, 81; LEA: 162). (p.35) In this perspective, roughly, correspondence itself has to do with accordance or harmony between behaviour and reality. A wide concept of utility can connect belief, as expressed in actions or in dispositions to act, and reality, in such a way that if an action is in accordance with reality, the underlying belief is useful, therefore true—or: true and therefore useful.

As for Ramsey, in ‘Facts and Propositions’ (1927) he uses his well-known example of the chicken:

[I]t is … possible to say that a chicken believes a certain sort of caterpillar to be poisonous, and mean by that merely that it abstains from eating such caterpillars on account of unpleasant experiences connected with them … [I]n regard to this kind of belief the pragmatist view was correct, i.e…. the relation between the chicken’s behaviour and the objective factors was that the actions were such as to be useful if, and only if, the caterpillars were actually poisonous. Thus any set of actions for whose utility p is a necessary and sufficient condition might be called a belief that p, and so would be true if p, i.e. if they are useful.

(FP: 40)

It must be acknowledged that in this paper Ramsey is interested not in this kind of animal belief, for which he states that the pragmatist vision was correct, but in beliefs expressed in words (see Methven, 2015: 142–5). Yet, in analysing the content of beliefs expressed in words, Ramsey again refers to their causal role in behavior (FP: 44), hence highlighting not only the difference but also the continuity between the two kinds of belief. In particular, the conception expressed in the quoted lines is clearly not so distant from Wittgenstein’s idea of usefulness as eingreifen. In Ramsey’s perspective, too, when someone believes something, she is committed to some consequences, namely she will admit some possibilities of action and exclude other possibilities (cf. Rumfitt, 2011).

Wittgenstein never completely abandoned the idea of usefulness, but later criticized it, sometimes harshly, especially when simplified and taken to mean, brutally, that whatever it might be useful to believe will be true in virtue of its being useful, or to imply that everything human beings do, they do because it is useful.18 ‘Why do humans think? Because it has proved its worth?’ he asks himself in BT. ‘Why do they make calculations for boilers and not leave the thickness of their walls to chance?’ He continues: ‘And yet one can say that thinking has proved its worth. That there are fewer boiler explosions now than before … So sometimes one does think because it has proved its worth’ (BT: 179–80).19

It is within this composite conceptual framework that Quote I is embedded.

(p.36) Lotteries, mad physicists, and whether it is reasonable to believe in hell

In February 1932, Wittgenstein wrote a note in MSW 113, later copied in MSW 114 and in BT, with minor variations. While in its original formulation and in the BT the note does not refer to pragmatism, in the MSW 114 version it does. I quote MSW 114 (using PG for the translation) and add the explanation of the joke which Wittgenstein himself added in handwriting in a blank page of the BT.

(II1) If I want to carve a block of wood into a particular shape any cut that gives it the right shape is a good one. But I don’t call an argument a good argument just because it has the consequences I want (Pragmatism). I may call a calculation wrong even if the actions based on its result have led to the desired end. (Compare the joke ‘I’ve hit the jackpot and he wants to give me lessons!’.)

(PG: 185, from MSW 114: 158)

The joke runs as follows:

(II2) A tells B that he won the jackpot in the lottery: he saw a crate lying in the street and on it the numbers 5 and 7. He calculated that 5 × 7 is 64 and filled in 64. B: But 5 × 7 is not 64! A: I win the jackpot and he wants to give me lessons.

(BT: 185)

Pragmatism here is taken to be the position that an argument is right if it has the desired consequences, regardless of objective correctness. Before commenting, let us examine also Quote I, which in some aspects is close to this one.

In 1935, during a lecture regarding symbols, use, and application, Wittgenstein discusses the definition of a proposition as anything that can be either true or false. In this game, in order to know whether a sentence is a proposition, we ask: is this something which can be true or false? But Wittgenstein adds another question: is it possible to play this game, not with a proposition, but with the name of an object, say, ‘table’ or ‘apple’? Is it possible to negate an apple? ‘What corresponds to our idea of negation is the use we make of it,’ he says. This reflection leads him to think about the different uses of terms like ‘use’, ‘application’, ‘practice’, ‘usefulness’. Here is what he says:

(III) A distinction is to be made between use and application. When I talked of ‘Apple’ being used as an order, we understood this because we have here a practical application, which is useful. Whether an application has a use in practice depends on the kind of life we lead. The pragmatic criterion of the truth of a proposition is its usefulness in practice. But the person who says this has in mind one particular use of ‘useful’: its use in the lab, say, to predict the future. But if a mad physicist were to offer a prize for a completely wrong hypothesis, then a person whose hypothesis had a distribution of confirmations like this

The ‘Middle’ Wittgenstein (and the ‘Later’ Ramsey) on the Pragmatist Conception of Truth

(p.37) instead of along the curve, would find it useful although it was useless for prediction.

(LEA: 142)

The pragmatic criterion of truth is here identified with the idea of ‘usefulness in practice’. Now, is the hypothesis which meets the requirements of the mad physicist’s prize true? If what is useful is true, then, although the hypothesis does not predict the future correctly, it is true, because it is useful in making its author win the prize. In this example, ‘usefulness’ can be identified either as ‘what makes me win the prize’ or as ‘what predicts the future in the lab’. Similarly, in the previous example, the ‘right’ calculation could be defined either as ‘the calculation which makes me win the lottery’, or as ‘the calculation which is correct according to the rules of mathematics’. ‘The rules of mathematics’ is not a synonym of ‘what makes me win the lottery’, and ‘prediction of the future’ is not a synonym of ‘what makes me win the prize’. It seems, then, that according to Wittgenstein the trouble with the pragmatist conception of truth is that it conflates these two aspects.

Once again, let us try to draw Wittgenstein’s words alongside Ramsey’s. Going back to the example of the chicken’s belief, let us suppose that the chicken thought the caterpillar poisonous, and therefore did not eat it, in spite of the fact that it was not poisonous. In this case, the chicken’s belief is not useful: indeed, the chicken’s belief is false. Ramsey imagines a similar situation in GPC (1929), within a discussion concerning variable hypotheticals and unfulfilled conditionals (which I will not reproduce in its entirety): ‘Suppose the human race for no reason always supposed strawberries would give them stomach-ache and so never ate them,’ he writes. ‘Then all their beliefs, strictly so-called, e.g. that if I ate strawberries I shall have a pain, would be true [because the antecedent of the conditional is false]; but would there not really be something wrong?’ (GPC: 161)

What Ramsey is concerned with here is whether one should consider the unfulfilled conditional, ‘if they had eaten the strawberry, they wouldn’t have had pain’, as a fact, or not, and he affirms that one should not, because, rather, ‘what is a fact is that I have eaten them and not had pain’ (GPC: 161). In order to make sense of our conviction that, in any case, the choice between two systems, both of which fit the facts (the system of the happy strawberry eaters and that of the unaware strawberry abstainers), is not arbitrary, Ramsey interestingly mentions Peirce, and his idea that a long enough investigation would lead to the truth. He explains: ‘What was wrong with our friends the strawberry abstainers was that they did not experiment’ (GPC: 161).

What may be wrong, then, in the pragmatist perspective, if formulated in a crude and not sophisticated way, is the lack of consideration for the ‘in the long run’ clause, or, more specifically, the lack of consideration for experiments: the failure to check the belief with what really is the case. The same kind of attitude emerges more explicitly in Ramsey’s writings on truth:

(p.38) The next theory of truth to be considered is Pragmatism, the view that by a true belief is meant one which is useful. By the inventor of Pragmatism, C. S. Peirce, this was, I think, given primarily as an answer to Question I [the nature of propositional reference], a belief that A is B, being roughly a belief leading to such actions as will be useful if A is B, but not otherwise. Although this is not the whole analysis of propositional reference, I think that like correspondence it gives one element in it, and is not simply to be mocked at.

(OT: 91)

This theory, Ramsey continues, is in harmony with his own account, because he too holds that a belief that A is B is useful if and only if A is B, that is, if and only if it is true: and so conversely the belief will be true if and only if it is useful. But he adds, referring to James’s application of this general idea to religious belief, that James ‘thought that the truth of the belief in hell depended not on whether hell in fact existed but on whether it was on the whole useful for men to think it existed’ (OT: 92). According to Ramsey, this position is flawed, because the belief in hell, resulting in a man’s avoiding doing certain things, ‘will be useful if it really saves him from hell[:] … if there is no such place it will be a mere waste of opportunities of enjoyment’ (OT: 91).

Ramsey’s harsh criticism of James is, in my view, only partly justified, for reasons I cannot detail here for obvious limits of space.20 In any case, what is relevant for our comparison with Wittgenstein is that the argument is drawn roughly along the same lines, using similar examples to show what goes wrong in a crude conception of truth as usefulness. In the case of the lottery (Wittgenstein), the winner claims his belief to be justified, in virtue of its usefulness and regardless of its actual adherence to the correct rules of mathematics; the mad physicist’s hypothesis (Wittgenstein again) is claimed to be the right one, in virtue of its usefulness and regardless of its adherence to reality; the belief in hell (Ramsey) is claimed to be true, in virtue of its usefulness and regardless of the effective existence of hell. But, using Ramsey’s own words to sum up the criticism, ‘whatever the complete definition of truth may be, it must preserve the evident connection between truth and reference that a belief “that p” is true if and only if p’, the disregard of which is ‘one of the main sources of the errors of the pragmatists’ (OT: 92).21

There is, here, I think, a convergence between the two thinkers. Nevertheless, in the next, concluding part of the chapter, I will argue for the existence of strong (p.39) divergences in their general attitude regarding not only the pragmatist theory of truth, but theories (of truth) on the whole.

Truth and its senses, theory and its nonsense

‘Whether an application has a use in practice’, Wittgenstein states in Quote I, ‘depends on the kind of life we lead.’ ‘The kind of life we lead’ is the anchorage of sense, understanding, and usefulness itself. So, in the case of the mad physicist and in the case of the pragmatist’s use of the word ‘useful’ as well, we do have an application of ‘usefulness’, but this application works for a singular context and is not anchored in the basic practices of our life. This is, in Wittgenstein’s view, what makes the pragmatist perspective, at best, limited: it may work in some circumstances, but that is all. Pragmatism’s claim to generality is what takes it, ultimately, off track. This is true, according to Wittgenstein, not only of the pragmatist conception, but of any theory of truth—in fact, in his view, it is true of any theory. It is here, I think, that the main divergence with Ramsey emerges. Quote IV helps to make it clear.

During a lecture or a conversation22 held at the beginning of the 1930s, Wittgenstein commented on some views expressed by C. D. Broad, who listed three ‘theories of truth’, the correspondence theory, the coherence theory, and the pragmatic theory, according to which ‘truth is what works’:

(IV) Philosophy is not a choice between different ‘theories’ … We can say that the word has at least three meanings; but it is mistaken to assume that any of these theories can give the whole grammar of how we use the word …

Pragmatism. The hypothesis that there are electrons is taken as being true because in practice you can work as if it were the case. So also Einstein’s theory of relativity is accepted because it works in practice. Thus Euclidean space is used for everyday purposes, and relativity for immeasurable and astronomical distances. To decide between them would need a great deal of empirical evidence, and this is certainly the sense of truth we apply to them.

But we do also use the word true of a ruler, which is a sense not ascribable to any of the three examples given.

Thus it is nonsense to try to find a theory of truth, because we can see that in everyday life we use the word quite clearly and definitely in these different senses.

(LEE: 75–6)

Wittgenstein connects the pragmatist approach to scientific theories: Ramsey makes a similar point in OT (33–4). But in these lines Wittgenstein also manifests his (p.40) general attitude about philosophy, which—he says—is not a choice between theories; accordingly, if one were to defend one theory of truth—be it the pragmatists’ or another—one would not be doing philosophy, as conceived by Wittgenstein.23

As a corollary, philosophy, being an activity aimed at investigating meanings by way of investigating modes of employing words, is not interested in causal relations or in the effects of signs. ‘Meaning is a stipulation, not experience,’ Wittgenstein states in BT, ‘and therefore it is not causality’ (BT: 35; orig. MSW 111: 111, 18 Aug. 1931; see also BT: 36).

Bearing in mind that Ramsey had it that pragmatism was correct in identifying in causes and effects the sources of meaning (FP: 51), this point marks a major divergence not only between Wittgenstein and pragmatism, but also between Wittgenstein and Ramsey.24 Causes and effects, according to Wittgenstein, belong to the province of science, not of philosophy. Ramsey once accused Wittgenstein of being ‘scholastic’:

The chief danger to our philosophy, apart from laziness and woolliness, is scholasticism, the essence of which is treating what is vague as if it were precise and trying to fit it into an exact logical category. A typical piece of scholasticism is Wittgenstein’s view that all our everyday propositions are completely in order and that it is impossible to think illogically.

(P: 7)

Clearly replying to this charge, though without lingering over scholasticism, Wittgenstein in later years remarked: ‘Ramsey was right in saying that in philosophy one should be neither “woolly” nor scholastic. But yet I don’t believe that he has seen how this should be done; for the solution is not: being scientific’ (MSW 163: 57v, 1941).25

(p.41) Concluding remarks

Let me sum up. Wittgenstein’s 1930 reflection about the pragmatist conception of truth is more comprehensible once read as being stimulated by the ‘innumerable conversations’ (PI: 4) he and Ramsey had in 1929. Issues concerning expectation and fulfilment, primary and secondary languages, generalizations, laws, and hypotheses were extensively debated between them. The idea of usefulness that emerges from their writings is linked to these issues. But the idea of usefulness that they both perceived in connection with pragmatism was a simplification which, in some cases, deserved criticism. For Wittgenstein, this is not all. The pragmatist theory of truth has another problem: it is a theory of truth, and, what is worse, it is a theory which risks assigning a primary role to causes and effects. ‘Truth’ and ‘agreement with reality’, according to Wittgenstein, are expressions belonging to our language and we use them in a variety of ways, which a theory, especially when scientific in spirit, is not able to encompass.


Bibliography references:

Boncompagni, A. (2016). Wittgenstein and Pragmatism: On Certainty in the Light of Peirce and James (London, Palgrave Macmillan).

Boncompagni, A. (forthcoming). ‘Wittgenstein and Pragmatism: A Neglected Remark in Manuscript 107’, in M. Baghramian and S. Marchetti (eds.), Pragmatism and the European Traditions: Encounters with Analytic Philosophy and Phenomenology before the Great Divide (London, Routledge).

Campbell, N. (1920). Physics: The Elements (Cambridge, Cambridge University Press).

Carnap, R. (1924). ‘Dreidimensionalität des Raumes und Kausalität: Eine Untersuchung über den logischen Zusammenhang zweier Fiktionen’, Annalen der Philosophie und philosophischen Kritik, 4: 105–30.

Engelmann, M. (2013). Wittgenstein’s Philosophical Development (London, Palgrave Macmillan).

Glock, H.-J. (2005). ‘Ramsey and Wittgenstein: Mutual Influences’, in M. J. Frapolli (ed.), F. P. Ramsey: Critical Reassessments (London and New York, Continuum), 41–68.

Glock, H.-J. (2013). ‘Judgment and Truth in the Early Wittgenstein’, in M. Textor (ed.), Judgment and Truth in Early Analytic Philosophy (London, Palgrave Macmillan), 242–70.

Majer, U. (1991). ‘Ramsey’s Theory of Truth and the Truth of Theories: A Synthesis of Pragmatism and Intuitionism in Ramsey’s Last Philosophy’, Theoria, 57 (3): 162–95.

Marchetti, S. (2015). Ethics and Philosophical Critique in William James (London, Palgrave Macmillan).

Marion, M. (1998). Wittgenstein, Finitism, and the Foundations of Mathematics (Oxford, Oxford University Press).

Marion, M. (2012). ‘Wittgenstein, Ramsey and British Pragmatism’, European Journal of Pragmatism and American Philosophy, 4 (2): 54–80.

(p.42) Methven, S. J. (2015). Frank Ramsey and the Realistic Spirit (London, Palgrave Macmillan).

Methven, S. J. (2016). ‘Whistling in 1929: Ramsey and Wittgenstein on the Infinite’, European Journal of Philosophy, 24 (3): 651–69.

Misak, C. (2016). Cambridge Pragmatism: From Peirce and James to Ramsey and Wittgenstein (Oxford, Oxford University Press).

Moore, G. E. (1993). ‘Wittgenstein’s Lectures in 1930–33’, in J. Klagge and A. Nordmann (eds.), Ludwig Wittgenstein: Philosophical Occasions, 1912–1951 (Indianapolis, Hackett), 46–114.

Peirce, C. S. (1923). Chance, Love and Logic (London, Kegan Paul, Trench, Trubner).

Peirce, C. S. (1982–). The Writings of Charles S. Peirce: A Chronological Edition (W), gen. ed. E. Moore, 7 vols. to date (Bloomington, Indiana University Press).

Price, H. (2011). ‘Ramsey on Saying and Whistling: A Discordant Note’, in Naturalism without Mirrors (Oxford, Oxford University Press), 148–62.

Putnam, R. A. and Putnam, H. (1990). ‘William James’ Ideas’, in H. Putnam, Realism with a Human Face (Cambridge, MA, Harvard University Press), 217–31.

Ramsey, F. P. (1925). ‘Epilogue’ (E), in Ramsey (1990), 245–50.

Ramsey, F. P. (1927). ‘Facts and Propositions’ (FP), in Ramsey (1990), 34–51.

Ramsey, F. P. (1929). ‘Causal Qualities’ (CQ), in Ramsey (1990), 137–9.

Ramsey, F. P. (1929). ‘General Propositions and Causality’ (GPC), in Ramsey (1990), 145–63.

Ramsey, F. P. (1929). ‘Philosophy’ (P), in Ramsey (1990), 1–7.

Ramsey, F. P. (1929). ‘Theories’ (T), in Ramsey (1990), 112–36.

Ramsey, F. P. (1931). The Foundations of Mathematics and Other Logical Essays, ed. by R. B. Braithwaite (London, Routledge and Kegan Paul).

Ramsey, F. P. (1990). Philosophical Papers, ed. by D. H. Mellor (Cambridge, Cambridge University Press).

Ramsey, F. P. (1991a). Notes on Philosophy, Probability and Mathematics (NPPM), edited by M. C. Galavotti (Naples, Bibliopolis).

Ramsey, F. P. (1991b). On Truth. Original Manuscript Materials (1927–1929) from the Ramsey Collection at the University of Pittsburgh (OT), edited by N. Rescher and U. Majer (Dordrecht, Kluwer).

Rosso, M. (1999). ‘Introduzione’, in L. Wittgenstein, Osservazioni filosofiche (Turin, Einaudi), vii–cxxxvi.

Rumfitt, I. (2011). ‘Ramsey on Truth and Meaning’, in B. Morison and K. Ierodiakonou (eds.), Episteme, etc.: Essays in Honour of Jonathan Barnes (Oxford, Oxford University Press), 213–45.

Sahlin, N.-E. (1990). The Philosophy of F. P. Ramsey (Cambridge, Cambridge University Press).

Sahlin, N.-E. (1995). ‘On the Philosophical Relations between Ramsey and Wittgenstein’, in J. Hintikka and K. Puhl (eds.), The British Tradition in 20th Century Philosophy. Proceedings of the XVII Wittgenstein Symposium (Vienna, Hölder–Pichler–Tempsky), 150–63.

Schulte, J. (1999). ‘Wittgenstein—auch ein Pragmatist?’, in R. Raatzsch (ed.), Philosophieren über Philosophie (Leipzig, Leipziger Universitätsverlag), 103–20.

Stern, D. (1995). Wittgenstein on Mind and Language (Oxford, Oxford University Press).

Sullivan, P. M. (2005). ‘What is Squiggle? Ramsey on Wittgenstein’s Theory of Judgment’, in H. Lillehammer and D. H. Mellor (eds.), Ramsey’s Legacy (Oxford, Oxford University Press), 57–71.

(p.43) von Wright, G. H. (1993). ‘The Wittgenstein Papers’, in J. Klagge and A. Nordmann (eds.), Ludwig Wittgenstein: Philosophical Occasions, 1912–1951 (Indianapolis, Hackett), 480–510.

Weyl, H. (1921). ‘Über die neue Grundlagenkrise der Mathematik’, Mathematische Zeitschrift, 10 (1/2): 39–79.

Wittgenstein, L. (1922) Tractatus Logico-Philosophicus (TLP) (London, Routledge and Kegan Paul, 1961).

Wittgenstein, L. (1969). The Blue and Brown Books (BBB) (Oxford, Blackwell).

Wittgenstein, L. (1974). Philosophical Grammar (PG), edited by R. Rhees (Oxford, Blackwell).

Wittgenstein, L. (1975). Philosophical Remarks (PR), edited by R. Rhees (Oxford, Blackwell).

Wittgenstein, L. (1980). Wittgenstein’s Lectures Cambridge, 1930–1932 (LEE), edited by D. Lee (Oxford, Blackwell).

Wittgenstein, L. (1982). Wittgenstein’s Lectures Cambridge, 1932–1935 (LEA), ed. by A. Ambrose (Oxford, Blackwell).

Wittgenstein, L. (2000). Nachlass: The Bergen Electronic Edition (BEE) (Oxford, Oxford University Press).

Wittgenstein, L. (2005). The Big Typescript TS 213 (BT), edited by C. G. Luckhardt and M. A. E. Aue (Oxford, Wiley-Blackwell).

Wittgenstein, L. (2009). Philosophical Investigations (PI), edited by P. M. S. Hacker and J. Schulte (Oxford, Blackwell). (p.44)


(1) I am indebted to the participants of the British Academy Dawes Hicks Symposium, and particularly to Cheryl Misak and Hans-Johann Glock, for their remarks on the first version of this chapter. I am also indebted to David Stern for reading and commenting on the chapter, offering new suggestions with which I deal more extensively in Boncompagni (2016: ch. 1).

(2) Manuscript numbers are as catalogued in von Wright (1993) and cited as MSW n, where n is the manuscript number.

(3) On the phenomenological phase in Wittgenstein’s work, see Stern (1995: ch. 5) and Engelmann (2013: ch. 1).

(4) Except for the parts which already appear in published texts, translations from the manuscripts are mine, but I am indebted to Joachim Schulte for his very precious suggestions. As is well known, there is a debate among Wittgensteinian scholars on the translation of Satz as ‘proposition’ or ‘sentence’. In this context, I will usually translate it as ‘sentence’, as Wittgenstein is usually referring to ordinary language as opposed to logic and to phenomenological language. Nevertheless, one should bear in mind that Wittgenstein used the same word for both aspects.

(5) A reason for this is perhaps a little inexactitude in the existing Nachlass transcriptions, which reported ‘pragmatischen’ instead of ‘pragmatistischen’. For a wider analysis, see Boncompagni (2016), and Boncompagni (forthcoming) for a contextualization in the main philosophical perspectives of the times.

(6) Cf. PR: §50, as noticed by Rosso (1999: lxv). Marion (1998: 128–9) traces this back to Heinrich Hertz’s definition of dynamical models, which Wittgenstein mentions already in TLP: 4.04.

(7) In Ramsey (1931; 1990). See also NPPM: 34, 55, 61, 228–36, 253–6.

(8) For a general reconstruction, see Sahlin (1990: ch. 5).

(9) Carnap (1924) distinguished between primary and secondary worlds. Campbell (1920) introduced the concept of a ‘dictionary’, cf. Sahlin (1990: 133–5); according to Rescher and Majer, Campbell also used the expression ‘secondary systems’ for theories; see their introduction to Ramsey (1991b: xiv). As already mentioned, the distinction between ‘first’ and ‘second’ systems is also present in Hertz (see Marion, 1998: 128–9).

(10) See also LEE: 16, 83, 110.

(11) See also CQ: 137 and NPPM: 235.

(13) It is clear that it was the reflection on the Tractatus which ultimately gave rise to Ramsey’s elaboration of the notion of the variable hypothetical; indeed, it is within a discussion about this concept, in GPC, that Ramsey famously remarked that ‘what we can’t say we can’t say, and we cannot whistle it either’ (GPC: 146). For a discussion of this theme in relation to ineffability and the resolute reading of the Tractatus, see Methven (2016).

(15) Ramsey too used the image of the map when he said that beliefs are ‘maps by which we steer’ (GPC: 146).

(16) That this is the meaning Wittgenstein had in mind is explained in MSW 110: 229.

(17) See also BT: 432 (orig. in MSW 114: 19r, 3 June 32; later in PI §578), and PR §12c (orig. MSW 107: 229, 10 Jan. 1930).

(18) For the differences between Wittgenstein’s and the pragmatists’ ideas of use and usefulness, see Schulte (1999).

(19) These passages also appear in PI: §§466–70.

(20) But contrast this criticism with the attitude Ramsey himself expresses in the last paragraph of his 1925 ‘Epilogue’, perfectly Jamesian in spirit (E: 249–50). For a classical defence of the Jamesian position, see Putnam and Putnam (1990); for a contemporary perspective, see Marchetti (2015: ch. 4).

(21) Not surprisingly, Peirce expresses the same attitude in The Fixation of Belief: ‘The true conclusion would remain true if we had no impulse to accept it; and the false one would remain false, though we could not resist the tendency to believe in it’ (W 3: 244). This article was also published in Peirce (1923), widely read by the Cambridge philosophical community.

(22) This part of the Lectures 1930–32 belongs to a section which is unclear about the source of the text: it may be some notes of Wittgenstein’s lectures, but it may also be some notes of conversations which Maurice Drury had with Wittgenstein (see LEE: xiii n. 3).

(23) According to Misak (2016: 279), to some extent it is possible to say that Wittgenstein is here putting forward a theory of truth: one according to which in order to understand the meaning of ‘is true’ we need to account for all the different uses of it; and this is precisely the pragmatist theory of truth. In my view, the suggestion to look at the different uses of the word ‘true’ does not amount to offering a theory of truth, in the same sense in which the suggestion to look at the different uses of any expression in order to grasp its meaning does not amount to offering a theory of meaning. This is actually one of the most controversial issues in the Wittgensteinian literature, and, of course, expanding on it is beyond the scope of this chapter.

(24) According to Marion (2012: 63), while this critique is justified in the case of Russell, it is not in the case of Ramsey. By contrast, Glock (2005: 59–62) defends Wittgenstein’s criticism as addressed to Ramsey as well.

(25) See also MSW 152: 94–5 and MSW 164: 67. Wittgenstein surely had a copy of Ramsey’s 1931 volume Foundations of Mathematics edited by Braithwaite: in MSW 162: 9v (1940), in a remark concerning vagueness and exactitude, after writing the name of Ramsey he added, in English and in brackets, ‘Last Papers’, which is the title of a section of the book.