Ayer, Anscombe and Empiricism

Locke, Berkeley and Hume were three very different philosophers with very different preoccupations, modes of argument and attitudes towards the world. But by the middle of the 19th century it had become the custom to view them as the successive representatives of a single empiricist tradition. It is the English rather than the British who excel in the invention of traditions. And although the presence of an Irish bishop and a Scottish sceptic in the empiricist trinity made it necessary to think of the tradition under the title of ‘British’ rather than ‘English’ empiricism, it was always as a very specifically English cultural tradition – like cricket, afternoon tea and Anglicanism – that empiricism flourished.

It has flourished as more than a philosophy, if by philosophy we understand the conventionally bounded academic discipline. For it has provided the intellectual basis for a distinctively secular and liberal view of the world. Its cultural power has derived above all from its ability to produce at intervals representatives who have both made large contributions to academic philosophy and been effective public spokesmen for the secular and liberal causes of their day. John Stuart Mill was one such, and Bertrand Russell another. In our own time, the latest and perhaps the last of this chain of great figures has been Sir Alfred Ayer. Ayer shares with Russell and with Mill not only an intellectual allegiance to empiricist doctrine, and a willingness to identify himself with a certain kind of public cause: he shares, too, a certain temper of mind, which blends acerbity with generosity, and an admirably elegant and lucid English prose style. It has indeed become difficult to imagine what a cogent statement of empiricism would be like in any language other than English.

This last claim may at first sight seem absurd to anyone who has read the sometimes witty and always rigorous exposition of logical positivist doctrine that appeared in Erkenntnis in its great days. And it is of course true that in the Thirties in Oxford the young A.J. Ayer was seen by most of his philosophical elders as expounding an alien and distinctively Germanic doctrine in Language, Truth and Logic. Ayer himself, although recognising how much he shared with the native analytical school of philosophers who had learnt their methods from G.E. Moore – Susan Stebbing and John Wisdom, for example – saw himself as above all in debt to and an adherent of the views and arguments of the Vienna Circle. But, in retrospect, what is remarkable is the extent to which he had already domesticated Austrian positivism and turned it into something much more different from the original than either he or the hostile Oxonians, such as H.L.A. Prichard and W.B. Joseph, suspected – into, in fact, a new and newly exciting version of British empiricism.

Empiricism, like every other major philosophical perspective, creates many of its problems out of its solutions. So the empiricist insistence that all genuine empirical knowledge is founded upon our sense-experiences, a conviction whereby the empiricist hopes to distinguish beliefs for which, on an empiricist view, we have adequate rational grounds – such as beliefs about cricket bats and tea-tables – from beliefs for which on an empiricist view there are no adequate grounds – such as the belief that there are unicorns or (for most, although not all, empiricists) the belief that God exists – itself immediately generates problems about our knowledge of other people, about the relationship of our experiences of seeing, hearing, touching, tasting and smelling to the physical objects which ostensibly cause them, about the nature of personal identity, and about how scientific generalisations which extend beyond our experiences could be warranted by them.

For if all that experience acquaints us with are the presentations of our senses, then how can we know anything of the thoughts and feelings of others, which are not so presented? And what can our conception of a physical body be, if not a construction out of a family of sense-experiences? Wherein can the unity of my experiences as mine or yours as yours consist, since we do not seem to have, over and above our particular experiences of perceiving, judging, wanting and the like, an additional experience of the unity of the self whose experiences they are? If the generalisations of science extend, as they do, to the unobserved and now unobservable past, how can our present observations provide grounds for rational belief in them? These are among the central internal problems of empiricism. They are problems which already preoccupied Ayer in Language, Truth and Logic, and they have continued to engage him in his later writings. Now 44 years later they provide the subject-matter for many of the contributors to the book of essays assembled to honour Ayer on the occasion of his retirement from the Wykeham Chair of Logic at Oxford.

Consider, for example, three closely related questions: that of the nature of the relationship between those immediate perceptual experiences which on an empiricist view provide the grounds for all our legitimate beliefs and our everyday pre-philosophical beliefs about the world of physical objects, of tables, cricket bats, grass and lamp-posts; that of the nature of the relationship between those same immediate sense-experiences and our scientific beliefs about what the most up-to-date theories tell us are the constituents of the world – protons, neurons, light-waves and fields of force; and the consequent question of the nature of the relationship between these two sets of beliefs. How, to pose the kind of naive question which would instantly have sprung to the mind of one of Macaulay’s imaginary schoolboys, could one and the same set of sense-experiences warrant two such very different sets of beliefs? And are the two sets of beliefs compatible with each other? If the world is really composed of colourless, silent particles, some of which have such remarkable properties as that of spin without mass, how can it also be really composed of green grass and round tables?

Five of the 12 essayists – Michael Dummett, P.F. Strawson, David Pears, D.M. Armstrong and Charles Taylor – are concerned with these or with closely related questions. Collectively – taken together with those earlier writings of Ayer on which these essays are a commentary and with Ayer’s reply at the end of this book – they constitute an impressive body of philosophical writing which will certainly play an important part in the education of future graduate students in philosophy. David Pears provides a characteristically thorough and illuminating examination of the ways in which Russell, Austin and Ayer have argued over how far and in what ways I can be mistaken about my own immediate sense-experiences. Charles Taylor argues trenchantly on Kantian grounds that there can be no such items as sense-data. And D.M. Armstrong, in one of the best essays in the book, presents the case in favour of the view that what we perceive directly just are physical objects. And Dummett and Strawson both develop cogent and impressive arguments. But when all these papers are read together, certain characteristics of the discussion obtrude on our notice which are not so obvious in any single paper.

The first is very simply – in spite of Taylor’s trenchant thesis – how little has genuinely been settled in this area. The Ayer of Language, Truth and Logic was shocked by, and wanted us to be shocked by, what he took to be the inability of traditional metaphysicians to solve their problems with any finality, and he ended his book with a chapter bravely entitled ‘Solutions of Outstanding Philosophical Disputes’. But the inability of empiricism to solve its problems is surely, on the evidence of this book, quite as great as that which Ayer ascribed to the metaphysicians.

Secondly, the plain non-philosophical reader will, I think, be surprised at the extent to which different philosophers disagree both about the nature of his or her experiences and that of the beliefs and judgments which are presupposed by them and based upon them. J.L. Mackie, for example, who wrote about these problems in his masterly Problems from Locke – in the present book Mackie’s essay is on the problem of the justification of inductive inferences – ascribes to the everyday beliefs of common sense a view according to which our immediate perceptual experiences are construed as representations of actual physical objects, so that the plain pre-philosophical person, for example, takes ‘colours-as-we-see-them to be resemblances of qualities actually in the things’. Strawson strongly denies this. Ayer treats the common-sense view of perception as having the status of a theory or interpretation of what is actually immediately experienced. This, too, Strawson denies. How is such a dispute to be settled?

Dummett takes us part of the way towards answering this last question when he first points out that ‘a common-sense view is a conception on which most of those who belong to [a particular] culture [at a particular time] habitually rely in their everyday thinking,’ and then argues that the common-sense point of view cannot therefore be a single and unchanging accompaniment of ‘being a sane adult human being. There is no such single unchanging theory. It is, for instance, a mistake to think of anyone, however little educated, in our own society as entirely ignorant of science or as uninfluenced by it in his common modes of thought.’ But Dummett does not go quite far enough. For he still seems to treat philosophers as having projected onto the plain person philosophical theories which the plain person just does not hold explicitly or implicitly; and he may indeed be right in his particular accusations. Yet he does not point out adequately how often and how deeply the common sense of the plain person is informed, not only by scientific, but also by philosophical beliefs. Strawson thinks that Mackie misconstrues the plain person by representing him or her as too much like a philosopher who has read John Locke. But Locke’s ideas, in at least two senses, have been around for a long time now in the English-speaking world. And I should not be at all surprised to discover – the question is, after all, one of empirical sociology – that English plain persons, sometimes at least, are just the kind of Lockeans that Mackie takes them to be. Common sense is often enough a graveyard for dead philosophical theories.

What is the importance of this point? Classical empiricism requires a view of sense-experience as not only characterisable prior to and independently of all theory and interpretation, but characterisable prior to and independently of all theory and interpretation, but characterisable at this level in such a way as to provide an ultimate ground for any rationally justifiable belief with empirical content. Hence, if it were the case that our sense-experience itself already is such as to presuppose a set of philosophical and scientific interpretations of the physical world, then the whole empiricist project undergoes shipwreck. This is why the characterisation of the plain person’s experience is not a side-issue for empiricism. On the evidence of the arguments in these essays, it is still perhaps a little too early to start arranging memorial services for the victims: but it is clear that the ship is very low in the water and no lifeboats are clearly visible.

Ayer’s replies to his critics are at their strongest when he is dealing with central empiricist themes, and at their weakest when he approaches subjects which have traditionally posed difficulties for empiricism. So he is uncharacteristically, even if apologetically, brusque in his treatment of Richard Wollheim’s quite extraordinarily interesting piece on the way in which memory partially but very importantly constitutes our personal being. And he is perhaps least convincing of all in his reply to David Wiggins’s excellent essay in defence of the view that we cannot rationally discard a qualified Leibnizian conception of essences without violating a condition which must be satisfied if the physical sciences are to go about their work of explanation. Wiggins contends that ‘there are predicates which stand for essential properties of a thing and register the condition the satisfaction of which is a prerequisite of the very thing’s being articulated at all from the rest of reality ... These predicates are not in the business of explaining because they are presupposed to there being anything to explain.’ And to these he adds a second class of predicate, itself the product of scientific explanation, which enables us to identify those properties which constitute something as a such-and-such in the light of a particular scientific theory. Science, in other words, if Wiggins is right, cannot do without notions of necessity and possibility of a kind to which empiricists have always found it difficult to give hospitality. And Ayer proves to be no exception. His reply to Wiggins simply never encounters his central contentions head on.

What this suggests is that major philosophic perspectives differ from each other, not only over what theories are held to be true or false, but also over what problems are taken to be central rather than peripheral, over which concepts are paid the homage of extended attention and which treated merely as sources of little local difficulties, and over which texts from the past are accorded canonical status. It is one of the signs of the difficulties of contemporary empiricism that Leibniz’s writings are being treated now with a quite new kind of respect. Wiggins is by contemporary standards a very moderate Leibnizian indeed; and the Leibniz who is now venerated as the progenitor of modern modal semantics is a very different Leibniz from the one written about approvingly by Russell. Hegel’s thesis that how we do philosophy and how we write the history of philosophy are inseparably connected is once more vindicated.

Leibniz is not, of course, the only philosopher with whom empiricism has found it difficult to come to terms. Unsurprisingly, in the index to the book of essays in honour of Ayer the number of references to Hume is four times larger than the number of references to Aristotle. If the volume of essays that have been written by pupils and colleagues in honour of Professor G.E.M. Anscombe had had an index, the proportion would have been very different. But this would have been by reason of the attention justly paid to Aristotle, and not at all because of any neglect of Hume.

The criticism of Humean empiricism, as Cora Diamond remarks in her preface, has always been one of Elizabeth Anscombe’s major preoccupations. It has indeed been one of her most remarkable talents to use the criticism of major philosophers with whom she is in strong disagreement to open up whole new ranges of philosophical inquiry. And in so doing she has very often put to work Wittgensteinian insights in a way which has illuminated Wittgenstein as well as the authors whom she criticises. It was, for example, from Wittgenstein that she drew the resources to identify a major and striking mistake which pervades much of the history of modern philosophy and which, like many deeply-rooted philosophical mistakes, is apt to command our untutored assent quite unhesitatingly. The mistake in question is that of supposing ‘I’ to be a referring or denoting expression, so that the only philosophical question is that of what it is that ‘I’ refers to. Descartes answers that ‘I’ (or rather ‘je’ or ‘moi’) refers to that thinking thing which is the self; Hume supposes that it refers to that bundle of sensations which is the self; and other philosophers have given other answers. But all the answers are in error, for the question itself is asked only in error. Descartes saw rightly that once we had learnt how to use ‘je’ or ‘moi’ correctly we could not misapply those expressions. Anscombe commented: ‘Getting hold of the wrong object is excluded, and that makes us think that getting hold of the right object is guaranteed. But the reason is that there is no getting hold of an object at all. With names, or denoting expressions (in Russell’s sense) there are two things to grasp: the kind of use, and what to apply them to from time to time. With ‘I’ there is only the use.’ These pithy and condensed few sentences mark the opening up of an entirely new set of discussions – new, that is, relative to the empiricist and rationalist philosophical traditions – and five of the 17 essays in honour of Professor Anscombe are devoted to discussing the importance of Anscombe’s Wittgensteinian thesis.

The distance between Ayer’s philosophical focus and Anscombe’s is emphasised still further by the attention paid in some of the other essays to moral and theological questions. In his reply to Professor Stephan Korner’s essay, Ayer points out that, although in Language, Truth and Logic he dismissed the assertion of a transcendent deity as ‘literally nonsensical’, he did allow in a later book, The Central Questions of Philosophy, for ‘the possibility that the positing of a superior intelligence as the author of nature should serve as an explanatory hypothesis, though I also maintained that since it did not in fact fulfil this function there was no justification for it.’ Robert Hambourger, however, uses some extremely interesting discussions by Anscombe on causality and chance to defend the possibility of rehabilitating one version of the traditional argument from design for the existence of God. Hence it is the Hume of the Dialogues Concerning Natural Religion who is challenged by making use of an Anscombian insight; and even if Hambourger’s conclusions are necessarily tentative, they succeed in putting in question a certain amount of empiricist anti-religious dogmatism.

Others of the essays in honour of Anscombe discuss issues in the philosophy of action, the form of practical reasoning, a large family of problems connected with truth and modality, and Anscombe’s notable contribution to the discussion of the morality of war and sexual conduct. All of these essays are well worth reading. R.C. Jeffrey remarks in his essay how when he first read a particular passage by Anscombe it ‘struck me as gibberish’, but that a good deal later he came to see that the passage which had baffled him contained a key insight. I am sure that Cora Diamond is right in suggesting that a large number of those now working on philosophical problems must have had similar experiences. Ayer’s place in the history of philosophy is by now fairly clear. It is one of which he can be very proud. But Anscombe’s is very far from clear as yet. She has been Wittgenstein’s best interpreter and a major contributor in her own right on a wide range of issues. This book, with its bibliography of Anscombe’s writings, gives us the beginnings of a sense that as time passes she is going to appear to have been an even more remarkable transformer of the philosophical scene than she does now.