Ayer’s Place in the History of Philosophy

Anthony Quinton
DOI:10.1093/acprof:oso/9780199694556.003.0014
Abstract and Keywords
This chapter examines A. J. Ayer's place in the history of philosophy. Ayer thought of himself as the main representative in his own time of the British empirical tradition, running from Hume through John Stuart Mill to Russell. It is a natural and intelligible point of view. His interpretation of the leading ideas of that tradition, in the sharpened form given to them by the Vienna Circle, undoubtedly gave it a new lease of life. In the first post-war decades, philosophy in Britain was dominated by Wittgenstein and Austin, one a genius, the other enormously gifted, but both execrable examples in many ways. Confronted by dire imitations of their respective styles of writing and reasoning Ayer stood out as a marvellous champion of the best traditions of rational discourse.

Keywords:   A. J. Ayer, philosophers, philosophy, empirical tradition, rational discourse

When A. J. Ayer arrived in Oxford in the autumn of 1929 he had no thought of becoming a professional philosopher. He intended to go to the Bar, but, in the manner of an Etonian, by way of Literae Humaniores rather than the study of law. He had read a couple of philosophical books. The first of them was Russell’s Sceptical Essays, which he bought on its first appearance in 1928. The other was Moore’s Principia Ethica (1903), to which he had been led by a reverent aside in Clive Bell’s Art (1914). These choices were significant. Ayer always thought of himself as Russell’s successor. He modelled his thought on that of Russell, both in its content and in its unguarded expression, and also, to some extent, his manner of life, both political and amorous. What he got from Moore is less obvious, although his respect for Moore is evident, as is shown by the preface to Language, Truth and Logic and by his devoting a book to a close examination of his ideas, along with those of Russell. An important likeness is that both Moore and Ayer were provoked to philosophize by the assertions of other philosophers, not by problems arising outside philosophy in mathematics or the sciences, in history or everyday life.

Ayer avoided the full rigour of an Oxford training in the classics, devoting one term only to the acquisition of a minimal requirement in them. So, at the beginning of 1930, he started on the formal study of philosophy. The philosophical scene in Oxford at that moment was, for the most part, drab and wintry. Adherents of Cook Wilson, who had died in 1915, exercised a fairly oppressive intellectual authority: H. A. Prichard from the chair of moral philosophy, H. W. B. Joseph as an overwhelmingly energetic college tutor. (For what it is worth, at least as an index of their subsequent reputations, neither is honoured with an article in Edwards’s Encyclopedia of Philosophy (1967) and Joseph does not even secure a mention.) British Hegelianism, in opposition to which the school of Cook Wilson originally defined itself, was represented by H. H. Joachim. He, like most Oxford philosophers of that time, had given himself over to the historical study of the subject. There was also the wayward and defiantly isolated R. G. Collingwood. The only notable books published by Oxford philosophers in the 1920s were Collingwood’s Speculum Mentis, of which nobody took any notice, in 1924 and, in 1926, Statement and Inference, the posthumous compilation derived from the lectures of Cook Wilson and used by his followers as a kind of textbook.

(p.138) By the 1920s orthodox idealism was more or less extinct outside Scotland. The Cook Wilsonians had turned their antiquated artillery on to what they saw as the even worse errors of the Cambridge philosophers, who had done so much more than they had to undermine it. Prichard had vehemently criticized Russell’s theory of knowledge in an article in Mind in 1915 and he and Joseph were persistently hostile to Russell’s logic. Moore, as an ethical consequentialist, was one of those who, in Prichard’s view, had committed a fundamental mistake in moral philosophy. The only book Prichard published in his lifetime, Kant’s Theory of Knowledge, had come out in 1909: Joseph’s main work, his Introduction to Logic, as long ago as 1906.

There were, however, signs of renewed life. H. H. Price, Prichard’s favourite pupil, in a disconcerting betrayal, had been converted to the theory of sense-data advanced by Russell and Moore. Ayer, as an undergraduate, attended the lectures that were to be published by Price in 1932 as Perception. The more influential of Ayer’s philosophy tutors was Gilbert Ryle. After an early interest in the phenomenological movement, revealed in a respectful, if finally suspicious, review of Heidegger’s Sein und Zeit (1929), Ryle was, by the beginning of the 1930s, showing marked Russellian tendencies. ‘Are there Propositions?’ of 1930 presents a reductionist account of propositions that was to be echoed in the second chapter of Ayer’s Foundations of Empirical Knowledge ten years later and in his London inaugural of 1946: Thinking and Meaning. Ryle then laid out a comprehensive programme of analysis on Russellian lines in 1932 in his ‘Systematically Misleading Expressions’.

Cambridge was unquestionably livelier as a philosophical centre, in spite of its comparatively minute population of philosophers. From 1922 it had Wittgenstein’s Tractatus to discuss and, until 1927, Ramsey to discuss it with. Also in 1922, Moore’s Philosophical Studies had brought together a number of influential essays. C. D. Broad published his Scientific Thought a year later and in 1925 The Mind and its Place in Nature. At a geographical distance, but not at a very large spiritual one, Russell brought out his Analysis of Mind in 1921 and Analysis of Matter in 1926.

Not much was going on in other British philosophy departments. The most imposing product came from Manchester: Alexander’s Space, Time and Deity in 1920. Exported fruitfully to Australia by John Anderson, it was respectfully ignored in its country of origin. Kemp Smith’s Prolegomena to an Idealist Theory of Knowledge (1924) and some elegantly written works by Laird were all that Scotland had to offer. The United States supplied little in the way of exact or minute philosophy in the 1920s. It was an era of massive constructions: Whitehead’s Process and Reality (1929), Dewey’s Experience and Nature (1929), and The Quest for Certainty, the first volume of Santayana’s Realms of Being (1928).

The disparaging tone of these judgements of the philosophy of the English-speaking world in the 1920s calls for justification. The decade surely does look weak when it is compared with the effectively preceding period from 1900 to 1914. That roughly Edwardian epoch had contained the early and best works of Russell and Moore. Wittgenstein had come to Cambridge and electrified it. McTaggart’s Hegelian Cosmology (p.139) (1901) and Some Dogmas of Religion (1906) had provided exemplary displays of argumentative rigour. In Oxford idealism was still quite lively with Joachim’s The Nature of Truth in 1906 and Bradley’s Essays on Truth and Reality in 1914. Prichard and Joseph, as has been mentioned, published their most substantial work at that time. In the United States William James was copiously productive up to his death in 1910. The early, less amorphous, Dewey was at work. Santayana’s five-volume Life of Reason (1905) belongs to this period, as do Royce’s The World and the Individual (1900) and the collective volume The New Realism.

The large and altogether more adventurous philosophical activity of the Edwardian age as compared with the 1920s is to be explained—to the extent that it can be or needs to be explained—by the general spiritual devastation of the First World War. That suggestion is confirmed if one looks further afield to continental Europe. It was in the Edwardian era that the main and characteristic works of Croce and Bergson were published, as were the earlier and better works of Husserl, Vaihinger’s Als-Ob (1911), Cohen’s System der Philosophie (1915), and Cassirer’s Erkenntnisproblem (1906). By the 1920s Croce and Bergson had moved to the margins of the subject and the orthodox academic philosophers of importance were Husserl, Cassirer, and, perhaps, Nicolai Hartmann. Only near the end of the decade did Heidegger sound a new, arresting note.

From an Anglo-Saxon point of view a more promising publication than Sein und Zeit came a year later in 1928 with Carnap’s Logische Aufbau der Welt. In the United States a sign of a livelier period to come was C. I. Lewis’s Mind and the World-Order in 1929. These two books coincide approximately with the return of Wittgenstein to philosophy, and Cambridge, and the emergence of Price and Ryle in Oxford. Ayer’s philosophical career, then, began after a period of comparative sterility, but at a moment when, in Britain, the United States, and Europe, the first indications of renewal are to be discerned.

It might also be said that not only had the recent philosophical past been comparatively sterile, but the immediate philosophical present was, to vary the image, a bit of a vacuum. Russell’s best work was behind him and, so far as philosophy is concerned, he was a spent force. Readers of Raymond Chandler will remember Big Willie Magoon, head of the vice squad of the Bay City police. He failed to come across with something for which he had been paid and was seriously worked over by some heavies. As a result he was never much use afterwards. Something of the same kind happened to Russell when Wittgenstein demolished the ideas in his large and ambitious manuscript of 1913 on the theory of knowledge, in the course of a series of painful conversations in May and June of that year. ‘I saw’, Russell wrote to Ottoline Morrell three years later, ‘that I could not hope ever again to do fundamental work in philosophy.’ Moore had little new to say, however emphatically and persuasively he said it, as was shown by the publication in 1953 of lectures he had given as far back as 1910 and 1911. Broad confined himself to elaboration and criticism of the ideas of Russell, Moore, and, in the 1930s, McTaggart. Wittgenstein had been invisible for more than a decade. Ramsey was dead. The Vienna Circle had only just come into being. The field was wide open.

(p.140) Ayer’s account of his philosophical education in the first volume of his autobiography is rather sketchy. He recalls being introduced to the Tractatus by Ryle. He read Russell’s Our Knowledge of the External World (1914) and The Analysis of Mind (1921). The origins can be discerned in these three books of his later central doctrines about necessity, perception, and the mind. He also read Ramsey and Broad, James’s Pragmatism (1907), Poincaré and Nicod. He does not say what problems preoccupied him. Nor does he say anything much about the development of his thinking up to the point in 1936 of its first, strikingly finished and comprehensive expression.

There is little echo in his writings of any influence of the thought of Prichard and Joseph, apart from two negative reactions. The first of these is his repeated insistence that, as he puts it in The Problem of Knowledge (p. 15), ‘from the fact that someone is convinced that something is true, however firm his conviction may be, it never follows logically that it is true’. That is hardly consistent with his conclusion in the following chapter that statements in the present tense about one’s own immediate experience are incorrigible. For that is to say that, if I genuinely believe I am in pain or aware of a red patch, it follows that I am. The original thesis seems to be aimed at the Cook Wilsonian doctrine that knowledge is something unique and indefinable and such that we must know when we have it. Another echo is to be heard in Ayer’s treatment of the doctrine of internal relations. In The Foundations of Empirical Knowledge it is examined in connection with an argument quoted from Joseph (pp. 200–2).

In November 1932 Ayer went to Vienna and was courteously invited to attend meetings of the Vienna Circle. It was what he learnt there and from the pages of Erkenntnis, which had started publication, under the editorship of Carnap and Reichenbach, in 1930, that he was inspired to the extraordinary tour de force of Language, Truth and Logic. A good way of bringing out the depth of Ayer’s involvement is to compare the impact of a similar visit by an American contemporary at much the same time.1 Ernest Nagel, a few years older than Ayer, spent the academic year 1934–5 in Europe, his principal stops being Cambridge, Vienna, and Warsaw. News of the existence of logical positivism had already been brought to American philosophers by an article about it in the Journal of Philosophy in 1931 by Albert Blumberg and Herbert Feigl. Nagel’s report of his intellectual expedition is, as one might expect, thorough, serious, and reliable. He became a recognized associate of the Circle, but more as a fellow-traveller than as a fully committed zealot. He had contributed a boiled-down version of his doctoral thesis on the logic of measurement to Erkenntnis three years earlier and also an article on reduction in the sciences around the time of his European trip. Perhaps he was already, in his mid-thirties, too much involved with American naturalism to undergo a major conversion.

Ayer, on the other hand, completely ingested the four main doctrines subscribed to by the Vienna Circle: the identification of meaning with verifiability; the reductionism (p.141) which that implies if subjective experience is taken to be the basis of knowledge; the theory that necessity is analytic, a matter of linguistic convention; and a radically non-cognitive account of judgements of value. In taking on these views he was not content simply to report them. He had even in his mid-twenties an established intellectual allegiance to the philosophical analysis of Russell, Moore, Wittgenstein, and Ramsey. The special achievement of Language, Truth and Logic was the remarkable, almost seamless unification of his initial philosophical inheritance with the whole range of the main ideas of the Vienna Circle. Where he differed with the orthodoxy of Carnap, Neurath, and Hempel, rejecting what he labelled as its ‘formalism’, it was with ideas developed by Schlick. Where Carnap and his allies denied that statements can be compared with extra-linguistic fact and so took basic statements to be adopted by convention, Schlick saw them as direct reports of the facts of experience.

A survey of the contents of Language, Truth and Logic will show how comprehensive Ayer’s reliance was on the doctrines of the Vienna Circle. The title of the first chapter is ‘the elimination of metaphysics’. That phrase is a translation of the first words of an article by Carnap in Erkenntnis: Überwindung der Metaphysik durch logische Analyse der Sprache (1932). There Carnap identifies the meaning of a sentence, first, following Wittgenstein, with its truth-conditions, and then with its method of verification or, in the case of a word, with the criteria of its application. He goes on to say ‘in the domain of metaphysics, including all philosophy of value and normative theory, logical analysis yields the negative result that the alleged statements in this domain are entirely meaningless’.2 That and a few other references to value in Carnap were developed, with the aid of some apparatus from Moore, into the notorious sixth chapter of Language, Truth and Logic: ‘Critique of Ethics and Theology’.

In his Überwindung Carnap admits that the problem of the given, of the empirical basis of knowledge or, as he would prefer to put it, of the nature of protocol statements, is not yet solved. He soon went on to opt for the conventionalist view that they are statements about what is observed that are agreed upon by that shadowy, but quite cosy-sounding group, the ‘scientists of our culture-circle’. In 1934 in ‘Über das Fundament der Erkenntnis’, Schlick effectively criticized that position.

Although Ayer agreed with Schlick that basic statements are to be verified by the experiences to which they demonstratively refer, he denied that they are absolutely certain and incorrigible. His reason was ‘that a sentence cannot merely name a situation, it must say something about it. And in describing a situation one is not merely “registering” a sense-content: one is classifying it in some way or other, and this means going beyond what is immediately given.’ He goes on to say that to apply a word such as ‘white’ to an element of one’s experience is to say that it is similar to other contents of experience: ‘those which I should call, or actually have called white’ (Language, Truth and Logic, 126, 128–9). He did not support this view for very long. In ‘Verification and Experience’, (p.142) within a year of Language, Truth and Logic, he was expressing doubts about it and in the second chapter of Foundations of Empirical Knowledge in 1940 and in the introduction to the second edition of Language, Truth and Logic in 1946 he explicitly rejected it.

In claiming that all necessity is analytic Ayer depends as much on the Tractatus as on the Vienna Circle. But like Carnap, and unlike Wittgenstein, he took the truths of mathematics as well as those of logic to be tautologies. He wisely did not follow Carnap in defining ‘tautology’ in a narrowly truth-tabular way. The idea that logic and mathematics are useful because we, as finite intelligences, cannot trace all the implications of our thoughts was, as a footnote shows, derived from Hans Hahn’s Logik, Mathematik and Naturerkennen of 1933. Ayer’s refutation of Mill’s account of mathematical truths as empirical generalizations of the widest possible scope was not dependent on Frege, whom, by 1936, he does not appear to have encountered. He unfortunately ascribes to Mill the view that the truths of logic, as well as those of mathematics, are empirical, fixing that mistake firmly in the heads of several generations of students. Mill, in fact, acknowledged what he called ‘propositions merely verbal’; held that all essential propositions are identical propositions; and that deduction gives no new knowledge, says no more in its conclusions than is already contained in its premisses.

Ayer’s fifth chapter, on truth and probability, follows Ramsey on truth. It advances the theory about the corrigibility of all empirical statements whatever which he soon came to doubt and eventually to drop altogether. In a few concluding pages he had some suggestive things to say about probability. He defines it as the degree of confidence with which it is rational to entertain a hypothesis. He does not specify criteria of rationality, but says they are conventional and vary through time. That is to embrace the kind of conventionalism he rejected in the case of basic statements. Our old friends, the scientists of our culture circle, whose practice determines for Ayer what rationality currently is, are called back from epistemological retirement. The conception of probability as an intrinsic property of a hypothesis and Keynes’s view of it as an unanalysable logical relation between hypothesis and the evidence for it are dismissed without argument. The frequency theory is not mentioned.

The significance of his brisk aside about probability is that it sustains the only solution he admits to the problem of induction. It is rational, he says, to be guided by the past in forming expectations about the future because that is what rationality is, by definition. No doubt the current practice of those we should ordinarily describe as scientists is rational in that sense. But (as Ayer would surely have insisted) we can guarantee their rationality only by defining ‘scientist’ so that no one who deviates from being guided by the past counts as one.

His dealings with God and the immortal soul in chapter 6 derive fairly directly from Carnap’s Überwindung. But the insistence that neither metaphysical sentences nor their negations can be asserted since both are equally devoid of sense is Ayer’s own. It has, as he saw, the consequence that neither atheism nor even agnosticism is any better off than theism, a concession that was not much appreciated by its adherents.

(p.143) The main business of chapter 6 is the thesis that judgements of value are expressions of feeling and not statements of fact, natural or non-natural. Ayer relies heavily on Moore to dismiss naturalism, although he rephrases Moore’s argument. The intuitive non-contradictoriness of ‘this is pleasant but not good’ is appealed to instead of the intuitive substantiality of the question ‘is this pleasant thing good?’ The verification principle is invoked to deal with Moore’s positive view about the nature of goodness.

Ayer’s own positive account of the matter is a sharper edged version of Carnap’s Überwindung (1932) contention that metaphysics (including value judgements) ‘serves for the general expression of the attitude of a person towards life’.3 In his Philosophy and Logical Syntax of 1934 Carnap interprets moral affirmations as imperatives, a more plausible view than Ayer’s, as subsequent developments in moral philosophy have shown. It seems probable that the main source here was the celebrated distinction of scientific from emotive language in The Meaning of Meaning by Ogden and Richards of 1923. They say, in terms Ayer takes over almost word for word, ‘ “(this) is good” serves only as an emotive sign expressing our attitude to this and perhaps evoking similar attitudes in other persons, or inciting them to action of one kind or another’ (p. 125). Ayer’s account of moral disagreement is taken from W. H. F. Barnes’s ‘Suggestion about Value’ of 1933.

In the four chapters so far considered the main borrowings have been from the Vienna Circle. The balance tilts towards Ayer’s British, predominantly Cambridge predecessors, in chapters 2 and 3, where a conception of the nature of philosophy is set out, and in the two final chapters which are about mind, matter, and the self and three ‘outstanding philosophical disputes’.

The idea that philosophy should be a business of analysis was affirmed at some length in the last chapter of Russell’s Our Knowledge of the External World (1914) and it was consistently practised as analysis, with a slightly different inflection, in the writings of Moore. In his neutral monist period from about 1918 onwards, Russell tried to analyse material objects and minds into what he called ‘events’. He took material objects to be constructions out of sensations and sensibilia, these being gently reified equivalents of Mill’s ‘permanent possibilities of sensation’. Minds and their states were seen as constructions out of sensations and images, eked out with bodily behaviour where that seemed desirable.

Ayer drew on Price’s Perception (1950) in his firmly phenomenalist account of our knowledge of material objects in chapter 3. But he restates Price’s theory of objects as ‘families’ of actual and possible sense-data in linguistic form: sentences about objects are translatable into sets of sentences, categorical and hypothetical, about sense-data. No place is found for Price’s ‘physical occupant’, that ghostly vestige of the substratum of Locke.

(p.144) He follows Russell in holding that sense-data are neither physical nor mental, these being properties of logical constructions. But for the detail of his account of mind he draws, not on Russell, but on Carnap and Hume. In his Aufbau Carnap had distinguished the eigenpsychisch from the fremdpsychisch, the mental states I experience directly from those I ascribe to others on the basis of their bodily behaviour. Ayer was driven to this by his belief that the argument from analogy is unacceptably metaphysical. He was to question that four years later in Foundations of Empirical Knowledge and abandon it altogether in 1953. The argument from analogy was rehabilitated by the consideration that, since it is only a contingent fact that I have the experiences I do, there is no logical or necessary obstacle to the possibility that the experiences of others might have been mine. The way out he followed was suggested to him by Ryle’s article of 1936 ‘Unverifiability by me’.

Ayer’s Humean view of the self as a series of experiences contained one wholly original element. Hume had been baffled in his attempt to find a relation which could connect a series of experiences into a self. Ayer’s suggestion was that the experience of oneself must all ‘contain organic sense-contents which are elements of the same body’ (Language, Truth and Logic, 194). That, as he saw, rules out the possibility of the survival by the self of bodily dissolution. It also ensures that no experience can belong to more than oneself, which the resemblance and continuity which Hume was inclined to favour failed to do. It makes my identification of my body (under some other description, of course) prior to my identification of myself, which is not too bad. But it also requires that every total momentary experience of mine should contain a veridical organic sensation of my body. That is a very questionable proposal, but I shall not pursue the matter here. It may be noted that later theorists of personal identity—Bernard Williams and Sidney Shoemaker, for example—who have claimed that it entails bodily identity have not done so in Ayer’s way.

A final thesis of Language, Truth and Logic which was soon dropped was also borrowed. From C. I. Lewis’s Mind and the World-Order Ayer took the thesis that statements about the past are really hypotheses about future experience, on which we shall have to rely on in order to verify them. It implies, I suppose, that our memories are really thoughts about the future. It can be circumvented by the consideration that it is a contingent fact that my experiences occur at the time that they do.

Language, Truth and Logic, then, is almost wholly composed of pre-existing material. It was brilliantly presented, with astounding concision, and its content fitted very well together, despite the geographical remoteness of its two main sources. That is not surprising when one reflects that the philosophers of Cambridge and Vienna continuously interacted. Russell and Wittgenstein were, in a turbulent way, teachers and pupils of each other. The Vienna Circle studied them both and as intently as they studied anyone.

A special virtue of the book, over and above its excellences of style, its force, lucidity, and euphony, is its intellectual order. The elements may be borrowed but they are admirably arranged. After Our Knowledge of the External World Russell’s books became (p.145) increasingly loose and casual in construction. That can presumably be attributed to the loss of self-confidence caused by Wittgenstein’s ruthless criticism. Moore’s laborious repetitiveness and his confinement to a minute range of topics, however strategically important, was unsatisfying in a different way. The Tractatus, with pretty well all the argument left out, hovers about on either side of the frontier of intelligibility. Ramsey’s small, brilliant Nachlass was, in its more philosophically interesting parts, largely rough notes. Broad and Price were admirably lucid and thorough and Price’s writing had a particular kind of charm. But they were not, as Ayer unquestionably was, exciting.

Presenting the ideas of Vienna Circle in British costume he extricated them from the lavish use of symbolism, the off-putting technical terms, and the computer-like detachment of Carnap’s writing. More in the manner of Schlick he conceived the philosopher to be, however doctrinally subversive, a citizen of the republic of letters. Also like Schlick, who wrote an article on the Vienna Circle and traditional philosophy, he drew attention to the many anticipations in the philosophy of the past of the ideas he was expounding. Berkeley was invoked for his phenomenalism, so long as its theistic attachments were removed. Ayer was also in sympathy with his slogan: ‘Mem, To be eternally … recalling men to Common Sense’ (Life and Letters of George Berkeley, Clarendon, 1871, p. 455). The task of philosophy is not to justify our beliefs but to analyse them. In fact, Ayer probably absorbed that idea from Moore. He saw Hume as right in what he said about meaning, the two kinds of significant proposition, causation and, broadly speaking, the self, even if his thoughts were couched in too psychologistic an idiom. But Ayer’s attention to the philosophy of the past faltered when he came to consider examples of metaphysical illusion. Carnap in his Überwindung had illustrated his thesis with a short passage from Heidegger and a single sentence from Hegel. Ayer made do with a single sentence from Bradley.

It was the fact that Ayer’s revolutionary message was expressed in the literary form of a discursive essay and not in the manner of a scientific textbook, together with its elegance and brevity, that gave it such an impact. It was addressed to intelligent readers in general, from the then radical publishing house of Gollancz, and in an attractively up-to-date physical form: dumpy in shape with wide margins and a wholly non-Victorian typeface: Aufbau by Bauhaus. (The move to Macmillan for his next book in 1940—they were to be his publishers for the next thirty-two years—was a turn towards academic respectability and exclusiveness.) Many of the intelligent readers aimed at took it up. No British philosophical book since Russell’s Problem of Philosophy (1912) a quarter of a century earlier can have been as widely read.

Ayer’s later books—Foundations of Empirical Knowledge, Problems of Philosophy, and Central Questions of Philosophy—and his five essay collections are, by and large, confined to the topics treated in Language, Truth and Logic. That is as much a tribute to the superb economy with which they were treated there as a criticism. The book’s philosophical agenda covered a wide enough range for anyone’s philosophical career. And there are new thoughts presented and new fields explored in the later works. But he did not move far from his starting point.

(p.146) The main novelty of Foundations of Empirical Knowledge, apart from the limited changes of mind about incorrigibility and the argument from analogy already mentioned, is the contention that theories of perception are not competing substantive doctrines but, rather, proposals of ‘alternative languages’. It is not very clear what that amounts to. Does someone who rejects the sense-datum language have to say that some material objects exist only for one person, or intermittently, or have different properties for different people? The arguments used to show that only sense-data are directly perceived could then be mobilized to show that all direct perception is of private material objects and that public material objects, spatiotemporally and causally interconnected in a common world, must somehow be constructed out of them. In that case the alternative offered is of the most trivial, merely typographical kind. What is clear is that nobody seems to have paid much attention to it until J. L. Austin criticized it in Sense and Sensibilia (1962). By that time Ayer had explicitly, if quietly, dropped the whole idea. In his essay of 1945 on the terminology of sense-data he accepts the criticism of it by H. H. Price.4

Thinking and Meaning, Ayer’s London inaugural of 1947, is very properly dedicated to Ryle. For it is a thoroughgoing exercise in dereification, in which the thinking self, the instrument with which it thinks, the process of its thought, conceived as a sequence of mental acts, and the object of thought, its reference or meaning, are all resolved into the expression of thought in significant sentences. Particularly Rylean is the insistence that what have usually been conceived to be mental acts, such as believing and doubting, are really dispositions to come out with certain sentences or to act in certain ways. There is also a distinct echo of Ryle’s ‘systematically misleading expressions’ in the assertion that for an expression to have a meaning there does not have to exist a thing which is its meaning.

Ayer’s remarks about meaning in this lecture indicate a much closer attention on his part to philosophical logic. In his Philosophical Essays of 1954 it is the topic of the first three items: on individuals, the identity of indiscernibles, and negation. Then four epistemological items rehearse, with minor modifications, the familiar topics of sense-data, basic propositions, phenomenalism, and our knowledge of other minds. Finally Quine’s ontology is considered, Ayer’s ethical theory is presented with greater suavity but without substantial alteration, Bentham’s principle of utility is sympathetically explored, and the view that freedom and determinism are compatible, affirmed in pellet form in the footnote of Language, Truth and Logic, is defended and modified. In its 1946 version, as reprinted in Philosophical Essays, freedom is defined as absence of constraint, not as having no cause at all. A group of factors is brought together as constraining causes and thus as cancelling responsibility, but no account is given of what principle links them together and distinguishes them from other, non-exculpating causes of action.

The essay on individuals tentatively puts forward arguments for the eliminability of singular terms and the possibility of a purely predicative language implied by Quine’s (p.147) extension of the theory of descriptions to all referring expressions. The issue is taken up again in an essay on names and descriptions in The Concept of a Person, but now what was earlier suggested is categorically affirmed.

Quine’s use of his regimented language, in which all reference is carried out by quantifiers, to reinstate ontology in a new, logically respectable form, set Ayer thinking generally about what there is. In his study of William James ( in The Origins of Pragmatism) and again in an essay ‘What there Must Be’ in his Metaphysics and Common Sense of 1969, Ayer turns to the question. Against Quine and Russell he concludes that reducibility does not imply some kind of inferior existential status, as ‘mere constructions’ or fictions, in the items to which it applies as contrasted with the unassailably real existence of the elements from which they are constructed. Analysis may eliminate singular terms. It does not fictionalize material things. Here again I think there is a distinct echo of the Vienna Circle, particularly of Carnap’s Scheinprobleme (1928) and Schlick’s ‘Positivismus und Realismus’ (1933).

They had maintained that the reducibility of material things to sense-experiences casts no shadow whatever on the reality of the material world. It simply makes clear what the empirical reality of material things consists in. For such things to exist the course of sense-experience must exhibit certain kinds of regularity. If it does there is no further question about the real existence of the material things defined in terms of these regularities. To infer that, as Schlick puts it, ‘only the given exists’ (Es gibt nur das Gegebene) is to fall into unacceptable metaphysics.

In his positive views about ontology Ayer takes an exactly opposite position to Russell. Sense-experiences are not ontologically prior to material things since their occurrence is absorbed into a comprehensive theory of a material world. There they occupy a very modest place, as parts of the histories of sentient organic bodies. Priority for knowledge is one thing: priority of existence another. Ayer takes ordinary material things and the theoretical entities of sciences as competing for the status of being the real components of the material world. He vacillates between the two claimants—intellect urging electrons, sentiment pressing for tables—and concludes that it is a matter for choice and not very important anyway.

The Problem of Knowledge of 1956 has an excellent chapter on memory. It dispels the influence of a radically misguided chapter of Russell’s Analysis of Mind (1921) on the subject. Memory is having beliefs about the past that do not depend on inference or testimony. It is not images accompanied by feelings of familiarity and pastness, whatever they may be. Some interesting points are raised in the final chapter on myself and others. If a person is simply a contingently related series of experiences, why should an experience not occur on its own, unrelated in a person-constituting way to other experiences? Why, in other words, should there not be unowned experiences? Personal identity is addressed once more, in a guarded, hesitant way. Perhaps Hume’s accusation of a circularity in the definition of personal identity in terms of memory can be circumvented. Survival of bodily dissolution seems conceivable. Yet purely psychological criteria of identity seem insufficient.

(p.148) There are three substantial novelties in the book. First, there is the neat account of what is needed to turn true belief into knowledge as the ‘right to be sure’. The question has not been pressed as to why Ayer does not take the ethics of belief to be as non-cognitive as the ethics of conduct, as he did by implication in his account of rationality in Language, Truth and Logic.

Secondly, there is the helpful schematization of problems in the theory of knowledge, about perception, other minds, the past, induction and so on as arising from the scepticism generated by a gap between certain kinds of belief—about material things, past events, the minds of others, the laws of nature—and the only evidence that is available for them—sense-data, memories and traces, behaviour and observed regularities. That was not new. It was present in the pre-war writings of John Wisdom and set out clearly in an article about Wisdom by Douglas Gasking in 1954. Something along the same lines had also been worked out by Friedrich Waismann. These sources make one wonder if it was not something casually let fall in conversation by Wittgenstein.

Ayer, like the other exponents of the scheme, recognized four standard ways of dealing with the gaps. First, intuition, which claimed direct access to the other side of it. Secondly, the theory of a general principle which could serve as a bridge across it. Thirdly, reductionism, which closed the gap by defining the things concluded to in terms of the evidence for them. Fourthly, the melancholy option of scepticism. Ayer conjured up a fifth contender, grandly called ‘the method of descriptive analysis’, which seemed to be a matter of admitting the gap and then blithely doing nothing about it.

In fact, once more confronting the question of our knowledge of material objects, he does something rather different, and offers a kind of watered down reductionism. Our beliefs about the material world are not translatable into statements about sense-data. They should be understood as a theory which draws all its support from the evidence of sense-data. There is an air here of trying to settle an overdraft with a cheque drawn on the same account. It shifts the problem to that of the relation of a theory to the evidence for it: is it like the detective’s theory that the injury to the deceased was caused by a blunt instrument (later found in the shrubbery) or like Newton’s theory that the fall of apples and the orbiting of the sun by the planets are manifestations of the same gravitational force? For better or worse, Ayer stuck to this position until his final treatment of the topic in Central Questions of Philosophy in 1973.

By carrying out the ungracious task of tracing most of Ayer’s original ideas, and many of those he came to express later, to external sources, I am not suggesting that there is no more to his achievement than the communication of these borrowings. He wrote some lively pieces of philosophical polemic: defending the sense-datum theory against Austin’s often captious objections, setting ferociously about Malcolm’s view that dreams are stories we are disposed to tell when we wake up, and arguing against Wittgenstein’s view that a private language is impossible. He expounded Peirce and James, Russell and Moore, with thoroughness and distinction. His special kind of clear-headedness proved valuable in his work on probability, a subject often liable to be swamped by fruitless technicalities.

(p.149) An unstated assumption behind this excavation of Ayer’s intellectual borrowings is that a thinker’s place in the history of philosophy is determined by the originality of the interesting or influential things that he says. I do, indeed, believe that assumption to be broadly correct. No doubt the more one learns of the history of philosophy the more one realizes that what one had supposed to be the proprietary great thoughts of those commonly reckoned to be great philosophers had often been thought by someone else first. Descartes’s cogito ergo sum, one learns, is an echo of Augustine’s si fallor sum. Locke drew heavily on Descartes, Gassendi, and Boyle. Nevertheless Descartes put his cognito to uses of which Augustine had no conception and, although there are identifiable fragments of other men’s ideas in Locke’s encyclopedic Essay (1690), the plan and basis of the whole great undertaking were his own. The special greatness of Kant is surely for the most part due to the novelty of his ideas.

Rather than considering the question in general terms it will be better to try to fix Ayer’s place in the scheme of things by comparing his achievements with those of his contemporaries who qualify for such a comparison. Broad and Price were lucid and persevering exponents of ideas put into circulation in the first years of the century by Russell and Broad. Broad had the advantage of knowing a good deal about natural science and about the history of philosophy. Price’s great merit was his cautious deliberation, his extremely high standard of what the clear understanding of a problem and its proposed solutions consists in. Their main service was to domesticate the inchoate innovations of others. That is something that Ayer accomplished with greater force and comparable distinction.

Two more obviously appropriate comparisons are with Ryle, ten years his senior, and Austin, his near-contemporary. Ryle, too, was an extensive borrower. His philosophical logic derives from Russell and the Tractatus, his philosophy of mind from intimations that had come to him of the thinking of the later Wittgenstein. But he put his own stamp very markedly on what he took. That is in part a matter of his breezy, epigrammatic, resolutely non-technical style. That style was in tune with a mode of proceeding that relied heavily on analogy and, in general, avoided the laborious protection of these from all possible objections. Oddly, in the light of their respective ways of life—collegiate in Ryle’s case, metropolitan in Ayer’s—it is Ryle who is the less academically respectable and conformist. He steered clear of all the standard etiquette of philosophical writing: the identification of aims, the footnotes, the polite references to Professor This and Doctor That. In the end I think I should accord him a larger place on the philosophical map than Broad, Price, or Ayer, but I admit that it is a close-run thing.

The main reason for it is that, whether by accident or design, his own versions of late Wittgensteinian doctrine are really very different from the original. The distinction between knowing-how and knowing-that was an insight of great value, however easy it is to take it for granted once it has been pointed out. The boldness with which he dismissed Cartesian dualism, by interpreting all mental life in terms of dispositions to behaviour, has the merit of providing something definite to discuss, rather than endlessly puzzle over.

(p.150) Austin was the critic who got furthest under Ayer’s skin as is clear from the indignant tone of Ayer’s response to his criticism of the sense-datum theory. Like Prichard and Moore he was a rigorously minute philosopher whose scrupulous pursuit of exactness gave him an extraordinary personal command over his contemporaries. He went further in that direction than his acknowledged master, Prichard, because of his greater intelligence and general mental agility. But his reputation has not survived him. A later generation finds it difficult to understand. Apart from ‘speech acts’ his innovations in the philosophical vocabulary are seldom heard.

The two philosophers with whom Ayer is most appropriately to be compared are Quine and Popper. Both, like Ayer, are intellectual by-products of the Vienna Circle; in Quine’s case gratefully (Word and Object (1960), after all, is dedicated to Carnap), in Popper’s with a constant determination to emphasize his differences from it. I think it is fair to say that there is no new idea in Ayer that deserves to receive or has received the kind of attention that has been given to the innovations of Quine and Popper.

Quine’s multifarious assaults on the notion of meaning as the chief instrument of philosophical explanation, on the analytic–synthetic distinction, on modal and intentional discourse, on assumptions about translatability, his revival of ontology, and his naturalistic account of the theory of knowledge, add up to a major change of direction in philosophy. Not all that many have chosen to follow him very far down the path he has pointed out but his ideas have proved to be inexhaustibly discussable.

Popper’s fundamental idea that science progresses by adventurous conjectures, controlled by attempts to overthrow them, is not without forerunners. Something like it is crucial in the philosophy of science of Whewell and it plays a part, if not such a commanding one as for Popper, in the brilliant chaos of the writings of Peirce. But there is no reason to suppose that Popper had the least familiarity with either of them at the time in the early 1930s when he was writing his Logik der Forschung (1935). Whewell had been obliterated by Mill in the English-speaking world and was quite unknown anywhere else. As for Peirce, M. R. Cohen’s pioneering selection, Chance, Love and Logic had come out in 1923 but it did not contain any of Peirce’s more Popperian pieces and was anyhow almost certainly unknown to him.

Secondly, and more to the point, Popper’s very various developments of his fundamental thought go in directions altogether unexplored by Whewell and only partially by Peirce. Peirce, like Popper, adopts a comprehensively fallibilistic theory of knowledge and rejects determinism. Of Popper’s own invention are his attack on the view that there are general laws of historical change, his account of probability as objective propensity, his defence of science as an account of the real nature of things against instrumentalism and essentialism, his application of evolutionary ideas to the theory of knowledge, and the connected doctrine of ‘World 3’, over and above the realms of matter and consciousness.

Quine and Popper arrived at their main new ideas early on: Quine by 1936 in ‘Truth by Convention’ and 1939 in ‘Designation and Existence’, Popper by the date of his first book, 1935. Holding firmly on to them ever since, they have established their fertility by (p.151) the substantial and interesting body of extensions and applications they have made of them. Ayer, on the other hand, tended to circle round a number of topics, varying his attitude towards them, sometimes opting for one view or another, sometimes concluding that it is a matter for decision, sometimes simply confessing doubt as to what the right answer is. The one thing he never questioned is the primacy of sense-experience, even if, with his later replacement of sense-data by ‘qualia’, he tried to mitigate the subjectivity of his preferred foundations for knowledge. For all the bold iconoclasm of Language, Truth and Logic the abiding impression is one of doctrinal indefiniteness. (It is interesting that he never responded to Quine’s rejection of analyticity except in a couple of non-combative pages of Central Questions of Philosophy.)

It is that, I think, that must explain the fact that his influence has been so much less than that of the two major contemporaries with whom I have been comparing him. I speak here of influence on philosophers, since Quine is unknown to the part of the general reading public which thinks it would like to know something about philosophy. As every teacher of the subject knows, Ayer has enormously influenced beginning students of philosophy. Since most of them do not go far with the subject, that influence must be, with many, a persisting one. As a paradigm of a philosopher for educated people he is surely exemplary for his unswerving dedication to argument, to explicit reasoning concisely expressed in the clearest language.

It is noticeable that his influence has been comparatively small in the United States. In the post-war years when this country was a net importer of visiting philosophers they crossed the Atlantic to learn about Wittgenstein, Ryle, and Austin. That must be partly due to the fact that, from the late 1930s on, the Americans did not need an intermediary to convey the message of the Vienna Circle, since most of its leading exponents were coming to be installed there in university posts. Furthermore Ayer’s ideas about necessary truth and phenomenalism had been anticipated in the United States, seven years before Language, Truth and Logic, in C. I. Lewis’s more substantial, if less exciting, Mind and the World-Order.

Another possible factor is that philosophers in the United States, as in Cambridge in its greatest days, typically know more mathematics and natural science than philosophers in Oxford. Ayer was always an enthusiast for science, but the enthusiasm was of a generalized, rather ideological kind. I believe it was essentially secondary to his animus against religion. The sensationalism to which he always remained faithful was something he shared with the great generation of late nineteenth-century unbelieving philosophers of science: Mach, W. K. Clifford, Karl Pearson. Present-day philosophers of science—Popper, J. J. C. Smart, the Reichenbach of Experience and Prediction (1938), even Carnap after the mid-1930s—take for granted the objectivity of the basis of empirical knowledge. Ayer’s distaste for religion seems to be the central vision and object of emotional commitment which it is the office of the brilliantly deployed dialectical manoeuvres to protect, rather as a conviction of the immortality of human souls, related by spiritual love, is the soft centre in need of fascinating ingenuities of argumentative defence in the case of McTaggart.

(p.152) Ayer thought of himself as the main representative in his own time of the British empirical tradition, running from Hume through John Stuart Mill to Russell. It is a natural and intelligible point of view. His interpretation of the leading ideas of that tradition, in the sharpened form given to them by the Vienna Circle, undoubtedly gave it a new lease of life. But in its most vigorous form at the present time that tradition has moved beyond the doctrinal constraints he accepted from it about necessity, perception, and the nature of science.

Although I have come to think his place in the history of philosophy is not as large as I had thought before settling down to write this essay, I think I understand why I should have thought I did. In the first post-war decades philosophy in Britain was dominated by Wittgenstein and Austin, one a genius, the other enormously gifted, but both execrable examples in many ways. Confronted by dire imitations of their respective styles of writing and reasoning Ayer stood out as a marvellous champion of the best traditions of rational discourse.

Notes:
(1) Ernest Nagel, ‘Impressions and Appraisals of European Philosophy’ (first published 1936), in Nagel, Logic Without Metaphysics (1956).

(2) Translated in Ayer, Logical Positivism, 60–1.

(3) Carnap in Ayer, Logical Positivism, 78.

(4) A. J. Ayer, ‘The Terminology of Sense-Data’, in Philosophical Essays, 103–4.