The philosophy of memory has been largely dominated by what could be called ‘the representative theory of memory’. In trying to give an account of ‘what goes on in one's mind’ when one remembers something, or of what ‘the mental content of remembering’ consists, philosophers have usually insisted that there must be some sort of mental image, picture, or copy of what is remembered. Aristotle said that there must be ‘something like a picture or impression’; William James thought that there must be in the mind 'an image or copy’ of the original event; Russell said that ‘Memory demands an image’. In addition to the image or copy a variety of other mental phenomena have been thought to be necessary. In order for a memory image to be distinguished from an expectation image, the former must be accompanied by ‘a feeling of pastness’. One has confidence that the image is of something that actually occurred because the image is attended by ‘a feeling of familiarity’. And in order that you may be sure that the past event not merely occurred but that you witnessed it, your image of the event must be presented to you with a feeling of ‘warmth and intimacy’. When all the required phenomena are put together, the mental content of remembering turns out to be, as William James says, ‘a very complex representation’.

While thinking philosophically we see problems in places where there are none. It is for philosophy to show that there are no problems.

Those of us who are not colour blind have a happy command of colour concepts. We say of trees that they are green in spring, that they are the same colour as grass and a different colour from the sky. If we shine a torch with a red bulb upon a white surface, we say that the surface looks pink although it is white. And if we suffer a bout of jaundice we (allegedly) claim that white things look yellowish to us, although they are not yellow, nor do they (publicly) look yellow. We employ this tripartite distinction unworriedly and unthinkingly. But when, in doing philosophy, we are called upon to elucidate colour concepts it becomes evident that these elementary concepts present intricate problems to the philosophical understanding. It is extraordinarily difficult to obtain a proper surview of colour grammar, and the temptations of philosophical illusion are legion. We go wrong before the first step is even taken, and hence do not notice our errors, for they are implicit in every move we make. We multiply impossibilities seriatim, getting better, like the White Queen, with practice. We then either slide into scepticism, or alternatively exclude it on empirical grounds - appealing, as is so popular in American philosophical circles, to the wonders of science, in particular physics and neurophysiology, to keep the malin genie from the door.

It is well known that Wittgenstein's reading of the philosophical classics was patchy. He left unread a large part of the literature which most philosophers would regard as essential to a knowledge of their subject. Wittgenstein gave an interesting reason for his non-reading of Hume. He said that he could not sit down and read Hume, because he knew far too much about the subject of Hume's writings to find this anything but a torture. In a recent commentary, Peter Hacker has taken this to show that ‘Wittgenstein seems to have despised Hume’. Hume, he adds, ‘made almost every epistemological and metaphysical mistake Wittgenstein could think of.

What could an empirical theory of the Mind be? Surely one which demonstrated that questions about the existence of minds were empirical questions – to be decided by observation, by the senses. This in turn would require an explanation of the meaning of statements about minds or mental states in terms referring to observable events, states and objects.

In the present state of philosophy in the English-speaking world, to choose to talk about sense data may seem perverse. What could be more boring for one's audience than to attempt variations on so threadbare a theme? And worse, what could be more unfashionable in the aftermath of Wittgenstein and Austin? My reasons for selecting this unpromising topic are twofold. First, the general theme of this series of lectures is empiricism. And whatever meanings we put upon that ambiguous word, it is clear that as a matter of history the problems of perception have been important problems for nearly all those philosophers who would consider themselves to be empiricists. And however unsatisfactory sense datum theories of perception may now be held to be, such theories have been central to the empiricist tradition. Secondly, it is important not to be too much impressed by the fact that a particular philosophical opinion is fashionable or unfashionable. The former certainly does not guarantee its truth nor the latter its falsity. It has often been remarked that philosophical opinions are very rarely refuted. Instead they fall out of vogue only to return some years later in another guise. It is perhaps time to take another look at the notion of sense data. The most ingenious and persistent attacks on analyses of perception in terms of sense data have been at best indecisive, as Professor Ayer showed in his reply to Austin's Sense and Sensibilia.

In those twenty or so pages of section xi of Part Two of the Philosophical Investigations in which Wittgenstein discusses the concept of noticing an aspect and its place among the concepts of experience, there are three passages which are explicitly concerned with the relations between seeing and interpreting in the experience of noticing an aspect.

To be an Empiricist with respect to knowledge of the natural world, is to insist that all knowledge of that world is rooted in perceptual experience. All claims which go beyond the deliverances of the senses must, in the end, be justified by, and understood in terms of, relations holding between those claims and sensory data. Crucial to the Empiricist case, therefore, is an account of how perception can be a source of knowledge. How can sensory experiences provide, for the owner of those experiences, information about objects and events which exist independently of the experiences themselves?

The following essay scavenges in contemporary sources to arrive at a fresh Empiricist account of perceptual knowledge. There are sufficient parallels with earlier doctrines to call the outcome ‘New Phenomenalism’, but the label is not important. The materials for the thesis have been gathered (and probably twisted) from several current writers, most notably P.F. Strawson and Jonathan Bennett, but no one of these writers is a proponent of the expounded thesis as a whole. As with a composite photograph, no face completely fits.

It is a pleasure to read Hume, and to watch him explore recalcitrant problems with agility of mind and grace of style. Ironically these twin abilities have worked against each other from the beginning, in the first place because in the matter of writing Hume was an innovator — nobody before him had so successfully albeit unwittingly adapted French syntax to the writing of English-and-Scottish - and in the second place because on the grace of his style subtleties of thought flow past his readers, who then accuse him of obscurity. So abstruse were his writings to his contemporaries that he failed to achieve the literary recognition for which he craved; and even today, long after the elegance of his style has been received, it is said by Passmore that Hume in contrast to Berkeley ‘was a philosophical puppy-dog, picking up and worrying one problem after another, always leaving his teeth-marks in it, but casting it aside when it threatened to become wearisome.’ Similarly Selby-Bigge says in his introduction to the Enquiries:

His pages, especially those of the Treatise, are so full of matter, he says so many things in so many different ways and different connexions, and with so much indifference to what he has said before, that it is very hard to say positively that he taught, or did not teach, this or that particular doctrine. He applies the same principles to such a great variety of subjects that it is not surprising that many verbal, and some real inconsistencies can be found in his statements. He is ambitious rather than shy of saying the same thing in different ways, and at the same time he is often slovenly and indifferent about his words and formulae. This makes it easy to find all philosophies in Hume, or, by setting up one statement against another, none at all.

As my title implies, I think the verifiability criterion is indeed a criterion of something. I do not intend, therefore, merely to commemorate it. On the other hand I am not sure that those who put it forward in its more liberal forms as a criterion of ‘factual significance’ or ‘literal meaningfulness’ were right in what they identified as the consequence of a sentence's failing to satisfy it. What I want to argue for, in a somewhat reductionist spirit, is a resurrected version of the ‘weak’ verifiability criterion. My resurrected version will certainly appear more rarefied, in so far as it is independent of (and does not therefore require to be embodied in) empiricism. It will, I hope, also be purified of some of the mortal blemishes from which the criterion, as construed by members of the Vienna Circle, seems not to have recovered.

The term ‘Empiricism’ has had at least two different, though not unconnected, applications in modern thought, one to scientific method and the other to philosophical theory. My intention in this lecture is to try to show that, while these two applications of the term have a common source, their actual referents are widely divergent and in large measure even mutually incompatible.

In the past empiricist philosophy has urged one or other or both of two interconnected, and sometimes interconfused, theses. The first has been a thesis about the causal origins of certain beliefs, the second a thesis about the proper criteria for appraising these beliefs. The causal thesis is that all beliefs about the structure and contents of the natural world are the end-product of a process that originates wholly in individual experiences of seeing, hearing, smelling, tasting, or touching. The criterial thesis is that all these beliefs are ultimately to be appraised for their truth, soundness or acceptability in terms of the data afforded by such perceptual acts. Of recent years the causal version of empiricism has been much attacked, primarily in regard to its implications about language-learning. The language in terms of which our beliefs are constructed is heavily conditioned, we are told, by certain congenital features of the human brain. But, whenever Chomsky and his followers have assailed the causal version of empiricism, they have always been careful to claim for their doctrines the warrant of empirical evidence. They have never questioned the correctness of the criterial version of empiricism.

Besides the observable properties it exhibits and the actual processes it undergoes, a thing is full of threats and promises. The dispositions or capacities of a thing — its flexibility, its inflammability, its solubility — are no less important to us than its overt behaviour, but they strike us by comparison as rather ethereal. And so we are moved to inquire whether we can bring them down to earth; whether, that is, we can explain disposition terms without any reference to occult powers.

Determinism is a spectre that has haunted our scientifically-oriented culture from the beginning. I happen to think that it is literally a ‘spectre’, a trick of the vision, an appearance with an internal cause only, and that it is no more than the ghost of our own conceptual determinations projected outward into a world in which it has no place and no proper being. From one point of view it is no more than an alienated fantasy involving a number of incoherent assumptions. Of these, one of the most important, and one of the most deeply eroded by much contemporary work, is the assumption that science and scientific understanding is a potentially completable system. From another point of view, however, the deterministic picture seems an inevitable product of scientific activity.

The purpose of this essay is to exhibit certain crucial shortcomings of some representative empiricist and anti-empiricist ethical theories and to sketch an empiricist ethics which is not exposed to these objections and adequate to our cognitive and practical position in the world. The discussion falls into two parts. Part I, which is mainly critical, begins with a general distinction between empiricist and anti-empiricist ethical theories and surveys the assumptions which are permissible to the former in the sphere of factual beliefs, practical attitudes and logical inference (section 1). It then examines the central theses of three kinds of empiricist ethical theory, namely utilitarian, contractual and emotivist theories (section 2) and of three corresponding kinds of anti-empiricist theory which can be viewed as attempts at correcting the faults of their empiricist counterparts (section 3). Part II, which is mainly constructive, contains a sketch of a new empiricist ethics (section 4) and in its light a brief discussion of various types of moral system (section 5) of the limits of moral pluralism and the nature of moral argument (section 6).