Instead of a motto

Owing something to pragmatism is not one of my obsessions.

Donald Davidson (in Borradori, 1994: 49)

WITTGENSTEIN WAS THE SINGLE MOST important inspiration behind Oxford conceptual analysis. This chapter explores the part that pragmatist ideas played in this legacy. I shall first dwell on the intellectual connections between Wittgenstein and Oxford philosophy. Next I turn to Wittgenstein’s relation to pragmatism. To this end I distinguish between Pragmatism, a specific historical tradition, and pragmatism, a loosely knit family of philosophical and meta-philosophical ideas that includes yet also extends beyond Pragmatism. Wittgenstein is clearly not part of Pragmatism, but he was aware—and suspicious—of the proximity between pragmatist ideas and some of his own later reflections. He shared the conviction that human practice is of paramount philosophical importance. But he was at odds with the pragmatist tendency to reject distinctions. In the main part of my text I compare and contrast Wittgenstein’s complex relation to both Pragmatism and pragmatism with that of leading Oxford philosophers like Ryle, Austin, Strawson, Grice, and Searle.

As regards the nature of philosophy, I concentrate on the shared tendency to avoid both epistemological naturalism and ontological supernaturalism, and the diverse stances towards meta-philosophical naturalism. As regards reference and (p.132) meaning, I explore the parallel transitions from meaning as reference to meaning as use, and the partly contrasting conceptions of linguistic use, mainly as regards the role of normativity. As regards truth, one finds a common rejection of the correspondence theory; but while Pragmatist theories define being true by reference to human beings, both Wittgenstein and the Oxonians confined themselves to the more moderate idea that the bearers of truth must be understood by reference to what human beings think and say, or could think and say.

Wittgenstein and Oxford

Wittgenstein spent most of the last year of his life—April 1950 to February 1951—at Elizabeth Anscombe’s place in Oxford. In spite of several invitations, he only addressed an Oxonian audience on a single occasion, the meeting of the Jowett Society in May 1947 (see Monk, 1990: 495–7, 564–75). Nevertheless his relation to Oxford is of interest on both biographical and philosophical grounds. In a letter to Malcolm (12 Jan. 1951), Wittgenstein described Oxford as a ‘philosophical desert’. In the same vein, Malcolm records: ‘It was reported to me that Wittgenstein had also referred to the philosophical circles of Oxford as the “influenza area”, a remark that gave offence to some Oxford dons’ (1984: 79).

Conversely, during his lifetime philosophers at Oxford might have been fascinated by Wittgenstein and his ideas, with which many of them had second-hand acquaintance; yet—with the exception of Anscombe—this was rarely without reservations. A leading representative of the old guard, Prichard, appears to have resented Wittgenstein. Even among those of the younger, analytic generation, many were suspicious of these siren songs hailing from those lugubrious fens.

Austin by all accounts disliked Wittgenstein (Hacker, 1996: 308 n. 46). Grice once heard him say: ‘Some like Witters, but Moore’s my man’ (1986: 51). He is also reported to have called Wittgenstein ‘a charlatan’. All the same, Wittgenstein’s Investigations were discussed in Austin’s legendary ‘Saturday Morning Meetings’, and he frequently referred to Wittgenstein in discussion (Warnock, 1973: 36; Pitcher 1973: 24). More importantly, Austin justified the appeal to ordinary language on grounds that are pragmatist in general and Wittgensteinian in particular: words are tools and, ‘as a minimum, we should use clean tools: we should know what we mean and what we do not, and we must forearm ourselves against the traps that language sets us’ (1956: 181–2; see Wittgenstein, 2009 [1953]: §§11–14, 23; Pitcher, 1973: 24; Glock and Kalhat, 2016). Ryle was a friend of Wittgenstein in the early 1930s, and their philosophies bear strong resemblances, especially concerning the nature of philosophy, language, and the mind. But even he balked at some of Wittgenstein’s contentions (see below).

Nevertheless Wittgenstein had a paramount influence on post-war Oxford philosophy, in particular on conceptual analysis (misleadingly known as Oxford (p.133) ordinary language philosophy). The Tractatus and Wittgenstein’s ideas during the 1930s—transmitted directly or indirectly—substantially influenced Ryle. The Investigations were even hailed as ‘our Bible’ by no less a figure than H. L. A. Hart (see Hacker, 1996: 163).

The flowers that blossomed as a result of Wittgenstein’s intellectual irrigation included important pragmatist ideas. But they also included ideas and attitudes potentially inimical to Pragmatism. That is what I shall argue in the following sections. First, however, we need to take a closer look at Wittgenstein and, respectively, Pragmatism and pragmatism.

Pragmatism and pragmatism

The demise of German idealism in the middle of the nineteenth century sparked off various intellectual trends that tried to overcome religious and metaphysical mystery-mongering by stressing the importance of human practice. The continental manifestations of this move from the Absolute to action include Marxism, Lebensphilosophie/existentialism, and hermeneutics. Pragmatism in the narrow historical sense is the Anglo-Saxon version of this development. Classical American Pragmatism was anticipated by Josiah Royce, founded by C. S. Peirce and Chauncey Wright, popularized by William James, and further developed by John Dewey, G. H. Mead, and C. I. Lewis (see Misak, 2013). Significant differences notwithstanding, these thinkers shared with their aforementioned continental cousins the fundamental conviction that philosophically important beliefs and concepts are rooted in intersubjective human activities. They also shared the Kantian claim that knowledge is not a passive mirroring of reality but requires an active contribution by the subject and Hegelian ideas about the dynamic and historical dimension of both mind and reality. Classical American Pragmatism differs from these cousins through its empiricist and utilitarian tendencies, and its association with natural science in general, and with Darwinism in particular.

Although it had followers elsewhere, e. g. F. C. S. Schiller in Oxford, Pragmatism is the only philosophical movement which is indigenous to the United States. More distant branches of the historical tradition include what I have labelled the ‘logical pragmatism’ of Quine and Davidson (see Glock, 2003), what one might call the ‘illogical pragmatism’ of Rorty and his acolytes, and various contemporary neo-Pragmatisms such as that of the later Putnam, Haack, and Brandom, as well as the ‘Cambridge pragmatists’ headed by Blackburn and Huw Price. The relation between Pragmatism and pragmatist ideas in Ramsey, the original inspiration behind Cambridge pragmatism, is a highly rewarding topic, yet one that I have to leave to more authoritative commentators (see, e.g., Hookway, 2005; various contributions to this volume).

(p.134) In addition to the narrow, historical notion of Pragmatism there is also a wider doxographic one—pragmatism. Here we are dealing with a family-resemblance concept denoting a ‘spectrum condition’ (see Bakhurst, Chapter 6, this volume). There are no ideas shared by all and only pragmatists. Instead, pragmatism is characterized by a set of related and overlapping philosophical tenets—topics, doctrines, methods, and aspirations. Nevertheless there is one fibre running through the whole family: the emphasis on human activity and the idea that philosophically contested or problematic concepts are to be clarified by looking at their role in our various practices, whether they be everyday or specialized, as in scientific investigations. In addition to these shared yet highly general leitmotivs, pragmatism tends to be associated with the following claims:

• • Anti-foundationalism: there are no indubitable starting points of inquiry. Our theories should not aspire to Cartesian certainty.

• • Instrumentalism: knowledge is not passive mirroring of a mind-independent reality, but the result of an active process. Our concepts and beliefs are instruments for the explanation and prediction of experience and thereby, indirectly, for guiding our actions.

• • Social touch: like language, knowledge is not a solitary phenomenon, but an inherently intersubjective achievement, contrary to the methodological individualism initiated by Descartes.

• • Semantic consequentialism: the meaning of an expression and the content of a concept or belief is determined (at least to a large degree) by certain consequences of using the expression/employing the concept/holding the belief. The meaning of a sign, in particular, is determined by dispositions or habits that its use induces in an interpreter. Semantic consequentialism comes in two varieties, depending on what consequences of sign use or beliefs it regards as semantically relevant. The more behaviourist variant—advanced by James and Mead—stresses dispositions for overt action or rational conduct. The more epistemic variant—propounded by Peirce—stresses the consequences for dispositions of thought and belief formation. The epistemic version is closely tied to verificationism.

• • Verificationism about meaning: the content of a concept/belief is determined by the experiential consequences that can be expected to hold if the concept applied or the belief were true. For example, to claim of the liquid in a flask that it is an acid is to claim, inter alia, that the action of placing a blue litmus paper in the flask would have a certain result, namely that of turning the litmus paper red. By this token, the meaning of a word like ‘acid’ consists of the ‘conceivable experimental phenomena’ implied by its affirmation or denial (CP 5. 412).

• • Alethic anti-realism: the general idea is that a belief is true to the extent to which it ‘works’. Again that idea comes in two variations (see Russell, 2014 [1908]). These are aligned, respectively, with the behaviourist and epistemic (p.135) positions on meaning/content mentioned above, through the underlying (mostly tacit) assumption that the content of a sentence or belief depends on the conditions under which it is true. According to the utilitarian version, a belief works if it motivates us to engage in successful activity; according to the cognitivist version, it works to the extent to which its consequences are epistemically acceptable. One way of being acceptable is coherence with or entailment of well-corroborated other beliefs. Another way of being in good epistemic standing is captured by Peirce’s famous claim that truth is ‘the opinion which is fated to be ultimately agreed to by all who investigate’ (W 3: 273). The most prominent formulation of the utilitarian version is James’s notorious dictum: ‘“the true,” to put it very briefly, is only the expedient in the way of our thinking, just as “the right” is only the expedient in the way of our behaving’ (1975: 106). On both versions, however, truth depends not exclusively on how things are, irrespective of how we think they are, but at least partly on our beliefs and cognitive or affective requirements.

Wittgenstein and pragmatism

Wittgenstein stands outside the historical tradition of Pragmatism. To be sure, he was influenced by James. However, this influence was exerted by James’s philosophy of mind and philosophy of religion rather than by his propagation of Pragmatism. Furthermore, as regards philosophy of mind and introspective psychology it was negative rather than positive.

James ist eine Fundgrube der Psychologie des Philosophen. [James is a rich source, not for the philosophy of psychology but for the psychology of philosophers.]

(2015: MS 165: 182)

Seine Bewegungen sind lauter/ so viele/ Versuche sich vom Spinnennetz der Metaphysik, in dem er gefangen ist, zu befreien. He cannot walk or fly at all, he only wriggles. Nicht, dass das nicht interessant ist. Es ist nur nicht eine Wissenschaftliche Tätigkeit. [His movements are nothing but/so many attempts to free himself from the spider’s web of metaphysics, in which he is trapped. He cannot walk or fly at all, he only wriggles. Not as if that weren’t interesting. It just is not a Scientific Activity.]

(2015: MS 165: 151)

At the same time, Wittgenstein’s acquaintance with James explains why he had a (very rough) idea of Pragmatism, even though he treated it as a doxastic position rather than a historical movement. A second source of acquaintance was Russell. But Wittgenstein’s image of Pragmatism, such as it was, seems to have been shaped more by Russell’s sharp criticisms of Pragmatist theories of truth—in particular that of James—than by his adoption of a behaviourist-cum-pragmatist (p.136) account of meaning in Analysis of Mind. Wittgenstein also recognized Pragmatist themes in Ramsey’s later work. He seems to have regarded them as erroneous, but also as posing a serious challenge (see 1978: pt VI §23).

Finally, Wittgenstein was aware of the proximity of some of his ideas to some tenets associated with Pragmatism. Yet this was something he was suspicious of. In particular, he was wary of pragmatism as a kind of Weltanschauung. ‘So I am trying to say something that sounds like pragmatism. Here I am being thwarted by a kind of Weltanschauung’ (1975: §422).

This quote occurs in the context of deliberating whether and how hitherto unheard of events might upset his everyday Moorean certainties, e. g. being in England and never having been to Asia Minor. The idea that smacks of pragmatism appears to be the following: There is no practical point in contemplating the possibility of such outlandish possibilities. In my view, however, there is a wider implication to the way in which Wittgenstein vents his qualm. Wittgenstein regarded proper philosophy as a critical activity and attitude rather than a set of doctrines. Pragmatism as a self-proclaimed system or movement struck him as doctrinaire and ideological, albeit no more so than other self-proclaimed movements like logical positivism.

What about Wittgenstein’s relation to pragmatism? Leaving aside details, there is no gainsaying that Wittgenstein was, at various stages of his career, close to several pragmatist claims listed above. More important, however, is his endorsement of two more basic pragmatist tenets. He shared the emphasis on the ‘primacy of practice’ (Putnam), or an ‘anthropological approach’ (Hacker, 2013: ch. 5) to philosophy. Thus Putnam (1995: 52) observes that though the later Wittgenstein may not have been a pragmatist ‘in the strict sense’ (i.e. a Pragmatist in my parlance), he ‘shares a central—perhaps the central—emphasis with pragmatism: the emphasis on the primacy of practice’.

What has gone unnoticed, so far, is an equally fundamental and important point of dissent. Thinkers of a pragmatist bent tend to indulge in the ‘debunking of dualisms’ (see Rorty, 1986: 333, 339). This inclination is already evident in the classical American Pragmatists. Because of their naturalism and their behaviourism, they rejected the Cartesian dualism of mind and matter. Because of their Hegelian heritage and their general dislike for academic compartmentalizations, they revolted against Kantian dichotomies, especially the differentiation of concepts and intuition, of theoretical and practical reason, and of philosophy and empirical science.

It is with regard to this second tendency that a rift with Wittgenstein opens up. The latter’s work is fundamentally at odds with the pragmatist tendency to reject distinctions and to ignore or play down conceptual differences. Against the ‘debunking of dualisms’ he was prone to set his ‘I’ll teach you differences’.²

(p.137) Philosophical method and style

Some of these meta-philosophical differences between pragmatism/Pragmatism on the one hand and Wittgenstein on the other carry over to Oxford conceptual analysis, mutatis mutandis and subject to case. Among the great three—Ryle, Austin, and Strawson—Ryle was most keenly interested in the nature and method of philosophy, largely inspired by Wittgenstein (Ryle, 2009 [1953a]: 266; 2009 [1971]: xxiii). In many respects his account is also close to Wittgenstein’s as regards its content. Like Wittgenstein, he held that philosophy should not simply address metaphysical problems by solving them, but should instead dissolve them by laying bare the conceptual confusion—notably ‘category mistakes’ to which these problems (in the eyes of Wittgenstein and Ryle) owe their existence. But whereas Wittgenstein denied that there could be non-trivial theses, let alone theories, in philosophy (e.g. 2009 [1953]: §§126–7), Ryle was more accommodating in this regard. He vehemently rejected the therapeutic conception of philosophy occasionally mooted by Wittgenstein (2009 [1953]: §255, see also §§254, 233). He took issue, in particular, with any suggestion that philosophy does not offer arguments, refutations, rectifications, or even ‘theories’—the last not understood, however, as giving ‘new information’ (2009 [1949]: lix; 1979: 131–2; see also Ryle, 2009 [1945]; Glock and Kalhat, 2016).³ Ryle’s strategy of dissolving problems by argument rather than therapy closely resembles one of James, insofar as the latter also regards certain questions and disputes as vacuous on account of his consequentialism about meaning, notably in his justly famous example of the squirrel going round a tree (James, 1975: 27–8).

At the same time Ryle shared Wittgenstein’s aversion to official doctrines.

There is no place for ‘isms’ in philosophy. The alleged party issues are never the important philosophic questions, and to be affiliated to a recognizable party is to be the slave of a non-philosophic prejudice in favour of a (usually non-philosophic) article of belief. To be a ‘so-and-so ist’ is to be philosophically frail. And while I am ready to confess or to be accused of such a frailty, I ought no more to boast of it than to boast of astigmatism or mal de mer.

(Ryle, 2009 [1937]: 161)

By this token, James’s campaign for pragmatism is misguided; and Peirce’s rebranding of his own position as ‘pragmaticism’ adds stylistic insult to meta-philosophical injury. The proper reaction, according to Ryle, is to forsake the comfort of labels and party alignments. (For a discussion of the strengths and weaknesses of such taxonomic austerity, see Glock, 2008: 4–9.)

(p.138) Wittgenstein neglected the history of philosophy, and perhaps even held it in disdain. This is one aspect of his thought that hardly influenced Oxford philosophy at all. Thus Ryle writes that Wittgenstein

gave the impressions, first, that he himself was proud not to have studied other philosophers—which he had done, though not much—and second, that he thought that people who did study them were academic and therefore unauthentic philosophers, which was often but not always true. This contempt for thoughts other than Wittgenstein’s seemed to me pedagogically disastrous for the students and unhealthy for Wittgenstein himself.

(Ryle, 1970: 11; see also Glock, 2006)

Ryle resolved not to be a ‘philosophical monoglot’, and indeed became a Plato scholar of note. Strawson, Hare, and Warnock, for their part, were Kant scholars (the first a truly important one), Hampshire was a Spinoza scholar, Pears a Hume and a Wittgenstein scholar, and so on.

During the twentieth century, interest in the history of philosophy tended to be more pronounced along the Isis than along the Cam. It is unsurprising, therefore, that Cambridge pragmatists tended to neglect historical issues more than their Oxford counterparts. Yet it also remains tempting to think that pragmatism in general is riveted by the idea of practice, and hence forward rather than backward looking. Nevertheless some pragmatists were keenly interested in historical issues. This holds in particular of the movement’s greatest representative, Peirce, but also of more recent specimens like Rorty and Brandom.

On average, the pragmatist community would appear to share the attitude of Austin. While not being as bashful as Wittgenstein, they take a business-like stance. Austin trained as a classics scholar, and wrote (but did not publish) a couple of papers on Plato and Aristotle early on in his career. Later on, however, he came to the view that engagement with the history of philosophy could hinder progress in philosophy. In his classic paper ‘A Plea for Excuses’, for example, he justifies his choice of topic partly on the grounds that it can be discussed

without remembering what Kant thought, and so progress by degrees even to discussing deliberation without for once remembering Aristotle or self-control without Plato … Here at least we should be able to unfreeze, to loosen up and get going on agreeing about discoveries, however small, and on agreeing about how to reach agreement.

(1956: 183)

A more persistent contrast between Oxford conceptual analysis on the one side and pragmatism and Pragmatism on the other exists with respect to the drawing of distinctions—a contrast partly linked to Wittgenstein’s influence. Asked at a party what he actually did, an Oxford philosopher replied: ‘You clarify a few concepts. You make a few distinctions. It’s a living’ (Swift, 2001: 42). An ‘Oxford living’, though of a different kind to that in the eighteenth century. Oxford conceptual analysts are (rightly) famous for their distinctions. Here is a very selective list:

• (p.139) • Ryle: falsehood vs. nonsense; language vs. speech; use vs. usage; knowing how vs. knowing that.

• • Austin: performative/constative; locutionary/illocutionary/perlocutionary; accidentally vs. unintentionally.

• • Strawson: meaning/reference/truth; reactive vs. objective attitude; analytic/synthetic.

• • Grice: analytic/synthetic; semantics/pragmatics; implications vs. implicatures.

• • Urmson: act vs. rule-utilitarianism.

The issue of naturalism

Let us now turn to another, more challenging parameter of comparison, the issue of naturalism. To this end, we must first distinguish different types of naturalism in philosophy.

• • Meta-philosophical naturalism claims that philosophy is a branch of or continuous with natural science;

• • Epistemological naturalism is nothing other than scientism: it insists that there is no genuine knowledge outside natural science;

• • Ontological naturalism denies that there is any realm other than the natural world of matter, energy, and spatio-temporal objects or events.⁴

One Wittgensteinian idea accepted by Oxford conceptual analysts (with the partial exception of Austin) is the rejection of meta-philosophical naturalism. Unlike science or common sense, philosophy does not itself describe or explain objects of any specific kind. It is not an empirical investigation of the material world. Nor is it based on a priori intuitions of the abstract entities or essences postulated by Platonism and Aristotelianism. Instead, it is a second-order discipline which reflects on the conceptual scheme that science and common sense employ in their empirical descriptions and explanations of empirical reality.

More specifically, conceptual analysts regarded philosophical problems as conceptual and concepts as embodied in language. They tried to resolve philosophical problems not through substituting artificial terms and constructions for (p.140) the idioms of natural languages, but through clarifying the latter. They described the ordinary uses of philosophically troublesome terms and contrasted them with their uses in philosophical theorizing. If philosophical problems originate in our actual conceptual framework, as ideal language philosophers granted, the introduction of a novel scheme will merely sweep these problems under the carpet, unless its relation to the old scheme is properly understood. Once we have elucidated ordinary language, conceptual analysts like Strawson continued to reason, we no longer require an artificial one. For the problems arise not out of ordinary language as such, but out of its distortion and misunderstanding in philosophical theories (Strawson, 1963; see Rorty, 1967: 15–19). Like Wittgenstein, Oxonians also repudiated epistemological naturalism. For instance, they were sympathetic to the idea that the way we understand human action and speech in everyday life, and, more systematically, in the historical and social sciences, does not revolve around deductive-nomological theories built on causal explanations. On scientism, Strawson comments, ‘From such philistinism as this we can only avert our eyes’ (1997: 35). Like Wittgenstein, Ryle and Strawson abhorred Cartesian dualism. At the same time they (gradually) became weary of nominalism and extensionalism.

For us the crucial point is that this repudiation of naturalism does not constitute a pervasive radical disagreement with either Pragmatism or pragmatism. Admittedly, the current employment of the label in English goes back to Dewey among others (see Krikorian, 1944; Kim, 2003). Moreover, many pragmatists are sympathetic to meta-philosophical naturalism. In this spirit Peirce wrote in a letter to James, ‘philosophy is either a science, or it is balderdash’ (quoted in Misak, 2013: 43).

At the same time, official professions notwithstanding, many pragmatists tend to avoid epistemological naturalism and meta-philosophical naturalism at least in practice. Thus Peirce rejected psychologism and genealogical fallacies about logic as much as Frege, Wittgenstein, and the Oxonians (CP 8. 190). If anything, he was even more hostile to nominalism than Strawson (see Mounce, 1997: 29–31). And a leading contemporary pragmatist, Brandom, has strongly resisted the influential naturalistic programme of reducing semantic and mental notions like meaning and content to more basic, non-normative, phenomena (1994).

Without being able to substantiate this in detail, I detect an underlying ambition that unites Wittgenstein and Oxford conceptual analysis with many, though not all, pragmatists. The aim is to steer a middle course between the Scylla of epistemological naturalism and the Charybdis of ontological supernaturalism. Human beings are special not because they are connected to a reality beyond the physical world of space, time, and matter (a Platonist third realm or Cartesian soul substances, for example), but because they can only be adequately understood from a normative perspective, one that is alien to the natural sciences. There is knowledge outside of natural science, knowledge of language, logic, and mathematics, for example. Yet the special status of such knowledge does not derive from a special subject (p.141) matter—supernatural entities beyond space or time; it must instead be explained by reference to normative practices (speaking, reasoning, calculating). These practices in turn presuppose agents with distinctively human capacities. But while these capacities cannot be adequately characterized in physical terms, they do not transcend the natural world. They are perfectly intelligible features of animals of a unique kind (intelligent, social, and co-operative); and their causal prerequisites and evolutionary emergence can be explained by science.

Wittgenstein epitomized this ‘third way’ by comparing language to a game like chess. On the one hand, a chess-piece is a piece of wood that can be described by physics. On the other hand, one cannot explain what a chess-piece or what the game of chess is in purely physical terms. But the difference between a chess-piece and a simple piece of wood is not that the former is associated with an abstract entity or with a process in a separate mental realm. It is rather that the chess-piece has a role in a rule-guided practice (2009 [1953]: §108).

Reference and meaning

On this cue, we now turn to the philosophy of language. The early Wittgenstein diverged from a referential semantics in contrasting referring (names) and saying (propositions), denying that logical constants refer, stressing the role of use in turning mere ‘signs’ into meaningful ‘symbols’, and his contextualism. Yet he retained the idea that elementary propositions consist of names and that the meaning of a name is the object it stands for (see Glock, 1996).

Whereas Wittgenstein’s philosophy of language moved gradually from reference to use, Oxford conceptual analysis could take the second perspective for granted from the start, largely on account of his influence. Thus Ryle expanded Wittgenstein’s later criticism of referential semantics. Not all meaningful expressions refer. And even for those that do, their meaning is not identical with their referent. To suppose otherwise is a ‘category-mistake’.

In the same vein, Strawson stressed the difference between meaning and reference. Referring is something speakers do rather than a feature of type-expressions. Strawson also favoured Grice’s communication-intention approach to meaning over the truth-conditional alternative.

In short, Oxford conceptual analysts propounded or presupposed approaches to meaning revolving around use and communication. This constitutes an obvious and highly important common ground with pragmatism. After all, the idea that meaning is ‘a property of behavior’ and determined by use was propounded explicitly by Dewey (1925: 179, see 170–85). It is implicit in the verificationism that is part and parcel of many pragmatist accounts; for the way in which the fulfilment or non-fulfilment of an expression is established is a central aspect of its use. The link between the meaning of a sign and the effect of its communicative (p.142) use is even stronger according to G. H. Mead’s social interactionism (1934). Finally, a move from reference to use also underlies causal theories of meaning of a pragmatist bent as propounded by Ogden and Richards (1923) and by Russell (1921), as well as the success semantics suggested by Ramsey.

Therein, however, also lies a point of contrast, albeit only with some pragmatists. Both a causal theory and success semantics are incompatible with Wittgenstein’s perspective, and he explicitly criticized the former (see Glock 2006). Mutatis mutandis for the Oxonians. They preferred normative conceptions, according to which meaning is determined by semantic rules. This holds most obviously of Ryle, who explicitly linked meaning to rules of use (2009 [1953b]). But it also holds for the constitutive rules of speech acts described by Austin (1970) and more closely characterized by Searle (1969). Rules are equally essential to Grice’s famous theory of conversational implicatures (1989), yet these principles of co-operative discourse are not specific to particular expressions and are assigned to pragmatics rather than semantics.

Once again, however, this difference is only partial. There are normativists among Pragmatists and even among paid-up pragmatists as well. These tend to be those antipathetic to ontological and epistemic naturalism, such as Peirce and Brandom. Furthermore, even pragmatists who are hostile or indifferent to the idea of semantic rules describe meaning in terms of use. For both pragmatists on the one side, and Wittgenstein and the Oxonians on the other, language is first and foremost not a formal logical system or an individual mental faculty, but a form of communication and an intersubjective practice embedded in other, non-linguistic activities. At a methodological level this means that concepts are elucidated by describing their roles in human life rather than through logical analysis. This is in line with Oxford conceptual analysis. To possess a concept is to know the meaning of certain expressions; by the same token, concepts are neither mental occurrences nor entities beyond space and time, but abstractions from our use of words. What survives the Wittgenstein-induced demise of logical and reductive analysis is conceptual analysis and linguistic paraphrase. Philosophical problems are resolved by explaining expressions and by establishing the status and inferential powers of the statements in which they occur. There is just one minor fault line here; yet it divides the Oxonians from both pragmatism and Wittgenstein. For better or worse, the former tended to be less expansive in the aspects of human practice they allowed as semantically relevant, a point which is most evident in Grice’s strict separation of semantics and pragmatics.

… and truth

We have seen that Peirce and James link meaning to truth, albeit more or less tacitly. In the wake of Frege this link became explicit not just in many formal (p.143) semanticists, but also in the Oxonians, though in a significantly different fashion. Mindful of the fact, also recognized by Frege, Wittgenstein, and Ramsey, that linguistic signs are unsuitable as truth-bearers, Strawson formulates the connection between meaning and truth as follows: ‘the meaning of a sentence is determined by the rules which determine what statement is made by one who, in uttering the sentence in given conditions, makes a statement’ (Strawson, 1971: 15). The bearers of truth and falsehood are neither sentences (type or token), nor speech acts of uttering sentences (assertings), nor mental episodes or states of believing. Instead, they are what is said/asserted/believed.

This commits Strawson to the reality of propositions (and concepts), i.e. of entities which are not just abstract but also intensional. He regarded this as the philosophical lesson of his career (2011: 194). But he does not endorse a Platonist reification. Propositions are simply what people say or believe, or could say or believe—they are nothing but sayables and thinkables: ‘the use of “true” always glances backwards or forwards to the actual or envisaged making of a statement by someone’ (1971: 192). The ‘idea of truth … leads straight to the idea of what is said, the content of what is said, when utterances are made; and that in turn to the question of what is being done when utterances are made’ (1971: 183–4). The same commitment arises for Austin and Searle on account of their distinction between illocutionary force/mood and propositional content.

Roughly speaking, therefore, there is a rapprochement with many pragmatist positions concerning the bearers of truth and falsity. When it comes to the account of what it is for such a bearer to be, respectively, true or false, however, significant differences loom. Wittgenstein and the Oxonians commit to alethic realism:

1. 1) ~ (It is true that p ⇒ it is believed/stated that p)

2. 2) ~ (It is believed/stated that p ⇒ it is true that p).

That a proposition is true neither entails nor is entailed by the proposition being stated or believed to be true by someone, whether now or at a projected terminus of inquiry. By the same token, that a proposition is true neither entails nor is entailed by believing the proposition to be true benefiting someone, either practically or cognitively. Instead, whether a proposition is true quite simply depends on how things are. ‘[A] sentence (Satz) is true if things are as we say they are by using it’ (Wittgenstein, 1961 [1922]: 4.062). ‘What he says is true = Things are as he says’ (Wittgenstein, 1974: 123). Similarly for Strawson. A speaker makes a true assertion if and only if things are as he asserts them to be. This idea also influenced two other Oxford philosophers not part of the conceptual analysis mainstream, namely Prior and Mackie.

At the same time, it stands in sharp contrast to the Pragmatist theories of truth, at least as they are standardly understood. This is already evident in Wittgenstein’s implicit and explicit criticism of James’s theory of truth. In the diaries he wrote (p.144) during the First World War, Wittgenstein declared Christianity to be ‘the only safe way to happiness’, while at the same time recognizing that someone like Nietzsche could coherently resist this Pascalian siren song, albeit at the price of misery (1991: 49–50). During the 1930s he stated: ‘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)’ (1974: 185). And in the 1940s he queries: ‘But aren’t you a pragmatist? No. For I am not saying that a proposition is true if it is useful’ (1980: §266).

Nevertheless, for Wittgenstein there remains an essential anthropocentric dimension to truth that is also acknowledged by pragmatism and, at least implicitly, condoned in Strawson’s account of propositions. It is the connection between truth and ‘what human beings say’:

‘So you are saying that human agreement decides what is true and what is false?’ It is what human beings say that is true and false; and they agree in the language they use.

(Wittgenstein, 2009 [1953]: §241)

In the same spirit a logical pragmatist like Davidson writes: ‘Nothing in the world, no object or event, would be true or false, if there were not thinking creatures’ (Davidson, 1990: 279). For, without such creatures, nothing would count as a sentence, and hence the concept of truth would have no application.

Combining alethic realism with anthropocentrism—i.e. a kind of use theory—about truth-bearers yields something like the following view. Whether a statement is true does not depend on human consensus or human practice. But:

1. 1. The concept of truth would not exist without people capable of classifying statements and beliefs into true and false ones.

2. 2. The bearers of both truth-values are things people say/think, or could say/think, which means that they can only be explained by reference to what people can say or believe.

3. 3. We identify truths (propositions) by classifying actual or potential token-sentences according to what they say.

(2) and (3) imply that there is a conceptual connection between truth and human thought/speech, although this connection does not form part of the definition of truth (see Glock, 1997).

Realism about truth and anthropocentrism about truth-bearers and hence about propositions and sentence meanings: that is the common denominator between Wittgenstein and Strawson. Arguably, the same combination prevails in Peirce (see Misak, 2013: 35–41), and it is evident in Ramsey (1990: 38–9). The latter’s deflationary theory respects alethic realism. At the same time Ramsey is adamant about a point that is often overlooked: a deflationary theory of what it is (p.145) for a proposition to be true must rest on a substantive theory about ‘propositions’, which explains what it is for something to be a truth-value. On the one hand, ‘there is really no separate problem of truth but merely a linguistic muddle’. On the other hand, ‘the problem is not as to the nature of truth and falsehood, but as to the nature of judgement or assertion’; consequently we need to analyse the nature of propositions to solve the problem of truth definitively. In true pragmatist spirit, Ramsey, Wittgenstein, and Strawson explain propositions by reference to judgement and assertion, and hence by the way humans think and speak.

My own favoured version of this anthropological and demystifying approach to propositions runs as follows (for a fuller version, see Glock, 2011). Human beings have highly evolved perceptual and cognitive capacities; these allow them, among other things, to form complex beliefs. This capacity is intertwined—both factually and conceptually—with their capacity to express and communicate these beliefs linguistically, through assertoric utterances, roughly, the utterance of declarative sentences. Both their behaviour and their utterances allow them to ascribe beliefs to others through oratio obliqua, prefixing a ‘that’ to a declarative sentence. Furthermore, the move to oratio obliqua removes the restriction to a single language that oratio recta is generally presumed to suffer from. As a result of the apparatus of oratio obliqua, we can speak not just about linguistic relations between the sayings of people, but also of logical/conceptual relations between what they say and what they believe. It is at this juncture that talk of propositions and their conceptual components comes into play, as an expedient abstraction for describing and explaining our linguistic practices and the abilities that underlie them. Both the practices and the abilities are unique and astonishing; and to explain their phylo- and ontogenesis is a formidable scientific challenge; but they do not constitute any metaphysical mysteries.