The Sponge of Apelles

Alexander Nehamas

  • The Skeptical Tradition edited by Myles Burnyeat
    California, 434 pp, £36.75, June 1984, ISBN 0 520 03747 2
  • The Modes of Scepticism: Ancient Texts and Modern Interpretations by Julia Annas and Jonathan Barnes
    Cambridge, 204 pp, £20.00, May 1985, ISBN 0 521 25682 8
  • Skepticism and Naturalism: Some Varieties by P.F. Strawson
    Methuen, 98 pp, £10.95, March 1985, ISBN 0 416 39070 6
  • Hume’s Skepticism in the ‘Treatise of Human Nature’ by Robert Fogelin
    Routledge, 195 pp, £12.95, April 1985, ISBN 0 7102 0368 3
  • The Refutation of Scepticism by A.C. Grayling
    Duckworth, 150 pp, £18.00, May 1985, ISBN 0 7156 1922 5
  • The Significance of Philosophical Scepticism by Barry Stroud
    Oxford, 277 pp, £15.00, July 1985, ISBN 0 19 824730 3

Thales of Miletus, with whom histories of Western philosophy conventionally begin, was said to have been so concerned with the heavens that he fell into a well while he was gazing at the stars. Ever since then (ever since they have existed, that is) philosophers have been objects of amusement and subjects of satire. For one thing, philosophical views often seem intolerably abstruse. For another, their sheer number, the sheer multiplicity of different equally abstruse views on the nature of the world or on the form of the good life, can itself constitute the ground of suspicion and criticism. And so, of course, it has.

In Philosophies for Sale, for example, Lucian of Samosata (AD 115-200) imagines that representatives of major philosophical schools are being sold into slavery by the gods. The device allows him to parade, and to scoff at, Pythagoreans and Cynics, Heracliteans, Cyrenaics, and Platonists, Atomists, Epicureans, Stoics, and Sceptics. Coming at the end of this long procession, the Sceptic appears almost reasonable when he claims that, in contrast to all the others, he knows nothing. But Lucian quickly undermines this impression. Ridiculously, the Sceptic does not even know what he himself is. His aim in life, he says, is ignorance. He wants to be blind and deaf, ‘without judgment or sense: in a word, a worm’. He doubts whether he has been actually sold, and urges his buyer to suspend judgment about whether he owns him. But in the end he follows his new master: Sceptics seldom lead revolts.

Lucian seems aware that Scepticism bears a complicated relationship to other philosophical schools, though he presents it as a direct competitor to them. His practice thus raises the serious question of the philosophical status of Scepticism itself. This leads directly to the heart of ancient Scepticism and to the question of its differences from the Scepticism invoked or attacked in modern philosophy. Both issues are raised, individually and collectively, by the books reviewed here.

First some background. Ancient Scepticism begins with Pyrrho of Elis (BC 365-270), who, perhaps appropriately, wrote nothing and founded no school; the little we know about him all comes from other authors. Scepticism developed two main branches. The first, institutionalised in the Academy, the school which Plato had established a century earlier and from which this approach – Academic Scepticism – received its name, flourished from the third until the first century BC. Much of our information about it comes from Cicero’s Academica, written in the first century BC and after Aenesidemus, thinking that the Academy had betrayed Scepticism, had started its second branch. This came to be known as Pyrrhonism and it continued until about AD 200, when Sextus Empiricus composed his Outlines of Pyrrhonism and his tracts Against the Professors.

Ancient Scepticism underwent many changes during its five hundred-year history. Some of them are documented in The Skeptical Tradition, an admirable collection of essays on ancient and modern scepticism. Many of these essays, as well as Professor Burnyeat’s vision of the multiformity and elasticity of Scepticism, will have a permanent effect on how such issues are discussed. This is not a volume which simply records and codifies the great interest which Scepticism has recently provoked. In addition, it actually establishes a new subject for philosophers and historians to investigate.

The Skeptical Tradition also shows that, despite their many differences, the ancient Sceptics shared a basic similarity of outlook. This outlook, in turn, distinguishes them sharply from their modern counterparts, and the clarity with which this contrast emerges is one of this collection’s main achievements. Modern Scepticism may in fact be descended from Arcesilaus, Carneades, Aenisedemus and Sextus. But we shall see that the ancients would almost certainly not have recognised it as their legitimate heir.

The opponents of ancient Scepticism, the ‘Dogmatists’, were those who accepted positive doctrines or ‘dogmas’. Unlike those we characterise by that term today, classical Dogmatists offered reasons and arguments for their views. The Sceptics regarded the Stoics as paradigm Dogmatists. But the background of the dispute included all the schools ridiculed by Lucian, and then some – notably Aristotelianism. The fact that there were so many different philosophers with so many different reasons for and against so many different answers to each of their questions was the main weapon in the sceptical arsenal. This weapon, whose name is ‘equipollence’, seems to align the Sceptics with Lucian and against the philosophers, with whom he prefers to group them.

At least in the ancient world, people turned to philosophy in the hope that learning the truth would lead to happiness. To some, however, it appeared that every plausible philosophical view is counterbalanced by another, equally convincing opinion. Consistently confronted with this situation, and eventually becoming adept at producing equipollent views themselves, some enquirers became incapable of endorsing any opinion about the way things are. They found themselves suspending judgment. They became Sceptics in a more technical sense. They withheld assent, and acted only on the basis of how things appeared to them, with no regard for their real nature. And, surprisingly, this refusal to accept anything as true, this desertion of philosophy and of much of common sense, led to tranquillity, to the very goal for which philosophy was originally undertaken.

It is crucial to realise that Sceptics are not reasoned into these positions and that they do not try to persuade others to follow them. This, as we shall see, would be to endorse the view that Scepticism is true and to invite the charge of self-refutation. Sextus thinks of his Outlines as a ‘narrative’, not as a demonstration, of Pyrrhonism. The Sceptic, he writes in a famous passage, is like the painter Apelles. Frustrated at his failure to depict the foaming mouth of a horse, Apelles flung his sponge at his painting and, quite by accident, secured just the effect he had despaired of getting. Tranquillity is the equally unanticipated product of suspending judgment.

The precise extent to which the ancient Sceptics avoided belief is, in fact, quite unclear. And so is the answer to the question whether a life without belief, either tranquil or agitated, is at all possible: whether human beings can truly live as the ancient Sceptics claimed they did. These very complicated questions are raised in some of the most interesting recent work on Scepticism – in the essays by Myles Burnyeat and Jonathan Barnes in The Skeptical Tradition and elsewhere, as well as in papers by Michael Frede. But what is also needed is a more elementary account of Scepticism, a presentation of the Sceptics’ main approaches when undermining their opponents’ views and reaching their own position.

This need is addressed by The Modes of Scepticism, a lively and accessible introduction by Julia Annas and Jonathan Barnes. The book, which begins with some general remarks, contains mostly a detailed examination of the ‘modes’ of Scepticism, traditionally attributed to Aenesidemus. These ‘patterns or schemata which constitute ways of inducing scepticism’ all have roughly the same form. Each argues that objects appear different in different situations, that there is no rational reason to prefer one situation to another, and that we can therefore neither affirm nor deny that any given object is really one way rather than really another. The equipollence of such ‘oppositions’ leads to suspension of judgment.

The modes differ in that each concerns a different type of ‘situation’: one considers how objects appear to different animals, another discusses different human observers, a third involves the various customs which can be found in the world. They also, as Gisela Striker has argued in The Skeptical Tradition, differ in their strategy. One set leads to the view that the real nature of things, whatever it is, is for ever inaccessible to us. The second implies that, since features are relative to particular situations, there is no such thing as the real nature of things at all. The modes, finally, differ in respect of how persuasive they are; in this respect they cover a broad spectrum.

Annas and Barnes highlight the differences among the modes, but they still show how all can be plausibly seen as expressions of a unified general attitude. They connect the modes with contemporary epistemological concerns, but they still preserve the distinction between ancient Scepticism and its modern namesake.

This distinction is at least partly due to the fact that the ancient Sceptics, as we said, cannot argue directly for their position. To do so would be to assent to it in a way that would make Dogmatists out of them. What, then, is the function of the modes? How are they supposed to ‘induce’ Scepticism? According to Annas and Barnes, the Sceptics cannot argue that suspension ought rationally to follow from opposed appearances, since the Sceptics cannot endorse the dogmatic standards of argument and rationality by means of which this ‘ought’ would have to be supported. Remarking that Sextus insists that suspension is merely something which ‘happens’ to us, they write that ‘suspension is the actual – perhaps the inevitable – result of our recognising the force of the premises.’ But as they themselves note, suspension is not inevitable: people persist in making judgments in the face of opposed appearances. The price of making it inevitable, I am afraid, is to build all the relevant normative standards into ‘recognition’ in the first place (‘If you really understand the premises ...’). This is to think that recognition involves, from the Sceptic’s own point of view, the correct appreciation of the implications of opposed appearances. But how is such a point of view different from that of the Dogmatist?

According to Annas and Barnes, the Sceptics ‘do not concern themselves with the soundness of their arguments but with their efficacy. A “good” argument ... is an argument which works,’ which actually leads to suspension. This view, which Sextus himself suggests, is not unattractive. Yet the notion of ‘argument’ cannot be so easily separated from the notion of ‘soundness’. If soundness is in no way part of your concern, are you still interested in argument? If the Sceptics only care about efficacy, why do they keep arguing in order to win over their opponents? Why not also try, say, invective, bumps on the head, bribes?

The answer is, of course, that such methods cannot (or should not) have an effect on the Dogmatists. And this is part of the solution to our difficulty. The Sceptics can rely on argument and can appeal to the normative notions of soundness and rationality because these are all notions their opponents already accept. Ancient Scepticism is polemical through and through. If, by the Dogmatists’ own criteria, a particular application of a mode constitutes a sound argument, then the Dogmatists are obliged, they ought, to suspend judgment. The Sceptics do not so much try to persuade the Dogmatists to become Sceptics as invite them to be as consistent Dogmatists as they can. If the Dogmatists succeed, they will see that their own standards force them to abandon them. If they do not, they may still realise that, despite their claims, their lives are not ruled by their standards: they have already left those standards behind. In either case, Dogmatism undermines itself.

According to the Sceptics, the Dogmatists, who do not withhold assent, should by their own standards do so. But the Sceptics, who do, are under no such obligation since those are not standards they are willing or, for that matter, unwilling to endorse. There is no neutral point of view from which it can be claimed objectively that rational standards ought to be abandoned. Any such ‘ought’ presupposes that some rational standards are being followed. The Sceptic can therefore say only that suspension, if it occurs, is merely something that happens to us. Hence the standard image of the purge that purges everything, including itself.

Concerned to abolish all belief, then, ancient Scepticism was very radical. Aiming at a life which, based only on how things appear, led to happiness, it was primarily practical. Being a philosophy directed at renouncing philosophy, it was irreducibly equivocal. But whether it was only another philosophical view or a revolt against philosophy, it actually existed: the ancient world contained real Sceptics, and they seemed to themselves to be living well.

The ancient Sceptics, however, died out; their writings disappeared. Though Cicero’s Academica was not unknown during the Middle Ages, Scepticism emerged again only when the Renaissance rediscovered Sextus. The price of resurrection was, as always, transformation.

The differences between ancient and modern Scepticism are obvious from the opening pages of Skepticism and Naturalism: Some Varieties, Sir Peter Strawson’s urbane, elegant, and, alas, short new book: ‘The Sceptic is, strictly, not one who denies the validity of certain types of belief, but one who questions, if only initially and for methodological reasons, the adequacy of our grounds for holding them. He puts forward his doubts by way of a challenge – sometimes a challenge to himself – to show that the doubts are unjustified, that the beliefs in question are justified.’ Though Strawson discusses a broad range of beliefs to which sceptical doubts may be appropriate, his list falls far short of the broad domain of ancient Scepticism. More importantly, Scepticism now attacks directly the grounds of belief and not belief itself. The position no longer urges us to stop believing, but only to cease claiming knowledge of certain topics. Both the intensity, as it were, and the range of the view have decreased. In addition, Scepticism has now lost its practical implications. Strawson’s sceptic doubts only for ‘methodological’ reasons, while Barry Stroud agrees with Descartes that Scepticism is appropriate only ‘in the philosophical investigation of human knowledge and not in everyday life’. Irrelevant to life and, except perhaps for a few Early Modern thinkers, not actually followed by anyone, Scepticism has become, in an entirely new sense, academic. Which is not, by itself, to say that it is unimportant.

In fact, Scepticism has always provided the background for Professor Strawson’s own metaphysical scheme, itself one of the important achievements in recent philosophy. Strawson’s ‘descriptive’ metaphysics has aimed to characterise, without ‘revising’, our actual modes of thought and speech. His goal was to show that unless the world is in general as we think and speak of it, we would not be able to think or speak at all. Needless to say, this project was as controversial as it was ambitious. Some, Professor Stroud among them, argued that such Kantian or ‘transcendental’ arguments will still allow the world to differ radically from our description unless they are supplemented by a ‘verificationist’ theory of meaning. This is the now widely rejected view that our words and concepts are meaningful just to the extent that they are verified (or falsified) in experience. Only this would ensure that the world would correspond to our way of characterising it. Others, by contrast, claimed that our characterisation of the world, our conceptual framework, is only one among many equally legitimate such systems. Since different frameworks may well envisage completely different worlds, no conclusions about the world’s real nature can be drawn from our specific conceptual scheme. Such claims are attacked in Dr Grayling’s book.

In his own book, Strawson does not try to answer such criticisms, but neither does he give in to them. In a bold manner, he turns away from Kant and towards Hume, whose ‘naturalistic’ way with Scepticism he now adopts and adapts. Dissatisfied with direct common-sense, scientific or theological arguments against Scepticism as well as with indirect transcendental refutations, Strawson argues that Scepticism cannot, and need not, be refuted. The reason is just that ‘whatever arguments may be produced on one side or the other of the question, we simply cannot help believing in the existence of body.’ Sceptical doubts are in this sense idle, ‘powerless against the force of nature, of our naturally implanted disposition to belief’; and so are arguments against them. The way to deal with Scepticism about the external world (the ancients, we should note, seem to have lacked this concept altogether) is simply to turn our back to it. This is, in fact, according to Strawson, what we do in any case, though, unfortunately, not always with as clear a conscience as we should.

Those who, in view of the use which Strawson makes of his naturalism, may want to return to Hume will find Robert Fogelin’s Hume’s Skepticism in the Treatise of Human Nature a useful guide. Fogelin offers a detailed, though not systematically structured, account of the various sceptical arguments in the Treatise. In contrast to many recent interpreters, Fogelin tries to restore these arguments to the central position he believes they occupy in Hume’s thought. Fogelin is an astute and sympathetic reader, and is himself attracted to Hume’s sceptical attitude. Sometimes, indeed, he focuses too narrowly on Hume’s arguments, and exposes their weakness, without offering a satisfactory general account of why Hume was drawn to Scepticism in the first place. Nevertheless, Fogelin’s central contention, that ‘Hume’s naturalism and skepticism are mutually supportive,’ which coheres well with Strawson’s reading, is important and to my mind correct. Fogelin concludes: ‘Hume’s skeptical arguments are intended to show that the logical (or epistemological) issues that philosophers have discussed admit of no solution. He can then turn, with perfect justification, to the factual question of how humans are able to form beliefs despite the skeptical arguments that can be brought against them.’

We need only add to this that the distinction between the ‘logical’ and the ‘factual’ (or causal) need not be exhaustive. The end of Book One of the Treatise already suggests that the logical and isolated individual who has ended up a sceptic at this stage of the argument is beginning to see himself as a rational and committed social being. To such a being, more mechanisms of belief-formation than he had earlier envisaged might be available. Belief, that is, may have a social basis. The contrast between the logical and the causal may involve an intermediate term. We may be able to soften Strawson’s sharp distinction between Hume’s simple and Wittgenstein’s more complex or ‘social’ naturalism.

Strawson, we have seen, believes that Scepticism about the external world cannot ever be serious; this is what is meant in saying that it is idle. In other cases, however, the situation is more complicated. Visual objects, for example, are smooth and coloured. Yet science says that they are collections of minute colourless particles separated by immense relative distances from one another. In a different context, we usually have direct moral reactions to other people. We cannot help seeing them as endowed with moral features and characteristics. Yet sometimes we take towards others an ‘objective’ attitude, devoid of the involvement which characterises our more common interactions. As Strawson has written in his earlier essay, ‘Freedom and Resentment’, we sometimes do, and must, look at another human being ‘as an object of social policy; as a subject for treatment; as something ... to be managed or handled or cured or trained’. Which of these opposed attitudes is correct in each case? Are there perceptual and moral properties, or is their existence, as a sceptical position would urge, merely an illusion? Furthermore, are there mental features (beliefs, say, or desires) which account for our behaviour, or is all human action to be explained in neurophysiological terms? Do speech and thought involve the understanding of meanings and universals, or are they simply behaviour conforming to a community’s practice?

The view Strawson describes as ‘hard’ naturalism replies that there really are only atoms and the void, natural non-moral creatures, neurophysiological facts and social conformity. Such scepticism about morality or mind would surely count as Dogmatism for Sextus: but this is not Strawson’s concern. Instead, he supplies a ‘soft’ naturalist answer to these questions: in none of these cases, he would like to say, does one alternative exclude the other. Strawson supports his answer by providing a general philosophical approach about which he is openly and disarmingly tentative. He is not sure that it applies in the same way to each case, and he doubts whether it applies at all to the last.

The approach is to claim that ‘error lies not on one side or the other of these two contrasting positions, but in the attempt to force the choice between them.’ Strawson’s central philosophical thesis is that ‘the appearance of contradiction arises only if we assume the existence of some metaphysically absolute standpoint from which we can judge between the two standpoints ... But there is no such superior standpoint.’ Our ‘conception of the real’ involves an irreducible relativity. Relative to our normal standpoint, human action really possesses moral features; relative to our less usual, though sometimes necessary, objective standpoint, it really has only natural properties. Relative to our perceptual standpoint, physical objects really have visual properties; relative to our scientific standpoint, they really have only physical characteristics. And so on.

This programmatically-described general approach is promising and exciting – work in progress in the best sense. It ties together many of the issues with which Strawson has been occupied over the years, and the direction of its future development is already charted. A specific issue worth considering further may be whether the apparent contradiction regarding perception can be resolved without appealing to relative standpoints. Instead, we might argue that the microstructure of physical objects, supplemented by a theory of perception, accounts for their phenomenal properties. But since such an explanatory relation between natural and moral features can hardly be said to exist, the parallel between the two cases may appear even weaker than Strawson suggests.

A more general issue concerns the fate of the ‘real’ once, as Strawson proposes, it is irreducibly relativised to different standpoints. One of the intuitive hallmarks of reality is just the fact that it transcends all points of view, that it exists by itself. Strawson’s ‘relativising move’ may be itself perfectly acceptable. But does it show that physical objects really are coloured (from the standpoint of ordinary perception) and really are colourless (from that of science)? Or does it, much more radically and in accordance with the second type of the ten modes, undermine the very idea that things are really anything, that there are real properties? Does not Strawson’s ‘soft’ naturalism, designed as it has been to dispute with the modern sceptic, unexpectedly appear to be close to ancient (dare we say ‘real’) Scepticism after all?

Strawson’s undogmatic theme is that Scepticism is to be sidestepped and not confronted. But what he skirts, Anthony Grayling attempts to engage directly. He, too, believes that Scepticism is ultimately to be met with ‘the Humean shrug’, but he tries to provide good objective reasons for it. His short and terse essay contains a sustained frontal argument against Scepticism. In broadest outline, the argument holds, first, that our perceptual beliefs about the world presuppose other beliefs to the effect that the objects of our experience exist independently of our perception, that there is an external world. These, Grayling claims, are transcendental beliefs; they are basic to our conceptual scheme. Secondly, relying on the views of Donald Davidson, Grayling argues that there is only one conceptual scheme: anything we recognise as such must include similar beliefs. Finally, Grayling accepts Michael Dummett’s ‘anti-realist’ (weak verificationist) view that our beliefs and statements receive their sense from the conditions under which they can be properly asserted, and not from those under which they are true (and which we might never come to know). If all this is so, Scepticism might be refuted by showing that our belief that things exist which are independent of our perception is necessary for us, not by showing that it is actually true. Grayling’s book is mainly devoted to the details of this complicated argument.

G.E. Moore thought that all sceptical arguments must depend on some premise less certain than the common-sense belief in external objects which they attack; we might say that Grayling’s argument against Scepticism is itself based on some highly controversial philosophical views. Its success is conditional upon their truth. And, the merits of the argument notwithstanding, it is difficult to say whether this condition has been met. The Refutation of Scepticism is the newest addition to a long line of proofs, from the Stoics to St Augustine, from Descartes to Kant, from Moore to Carnap. As such, it inevitably brings to mind Nietzsche’s view that ‘it is certainly not the least charm of a theory that it is refutable; it is precisely thereby that it attracts subtler minds.’

But can Scepticism be refuted? That it cannot is the central, uncompromising idea in Barry Stroud’s engaging and, in some ways, disturbing The Significance of Philosophical Scepticism. Contrary to many recent authors, Stroud takes philosophical Scepticism very seriously. He draws a sharp distinction between two questions. The first is the ordinary question ‘How do you know?’, which we ask on particular occasions for particular reasons and which receives different answers depending on the circumstances. The second is its philosophical counterpart, which is expressed in the very same words, but which concerns all knowledge in general. This second question, Stroud claims, can only be answered by the stark sceptical thesis: ‘No one knows anything about the world around us.’

The philosophical question is asked by Descartes. Seated comfortably before his fire and having no specific reason to doubt that this is the case, Descartes still asks whether this is something he knows to be true. After all, he claims, he cannot immediately exclude the possibility that he is dreaming. And if he is, no matter what else is true, he does not know that he is seated comfortably before his fire.

Stroud thinks that this case is typical of all our claims to have any knowledge whatever. He also thinks that Descartes is right to insist that in order to know anything we must know that we are not dreaming. If that is right, then indeed we can never know that we are not dreaming, since whatever tests we use to determine that we are not may always simply be part of our dream. We must therefore conclude that we know nothing of the world. Our experience can be just as it is, and yet nothing might exist outside us.

Stroud’s oblique attitude towards the issues he discusses is part of the reason why this book is attractive. Though he does not advocate Scepticism, Stroud does not believe that it has been effectively disarmed. Much of his text discusses what he finds unsatisfactory in the anti-sceptical positions of Austin, Moore, Kant, Carnap and Quine, as well as in the diagnoses of its sources in Stanley Cavell and Thompson Clark. One could easily take issue with a number of Stroud’s specific claims. But the book is engaging because it is unusual to find a philosopher who is not so concerned with offering the final word on a problem, and who is trying to keep a sense of the importance of that problem alive. There is something philosophical just in that.

Some philosophers may consider this is a wasted effort, a desperate rearguard action begun after all has already been lost. But even if traditional Scepticism about the external world does sometimes seem to have had its day, Stroud’s view still needs to be taken seriously. The reasons are the three assumptions which underlie Scepticism.

He claims that Scepticism springs from an assumption so natural that it is hard to imagine what it would be like not to make it. This is ‘the idea that the way things objectively are is completely independent of us and our language, and that we seek knowledge of those independent facts.’ The content of our experience may therefore fail completely to reflect the facts as they are. Stroud might have offered a stronger defence of this ‘realist’ assumption (which is attacked by Grayling). To say, as Stroud does, that it is deeply natural is a weak defence, especially given his own distinction between the everyday and the philosophical. Realism may be acceptable only in the former sense, as Scepticism is reasonable only in the latter. Stroud must show that such philosophical or metaphysical realism is a plausible view. Though some discussion of this can be found in the chapter on Kant, this aspect of Stroud’s case is far from complete.

Scepticism also depends on the idea of ‘the epistemic priority of ideas or appearances or perceptions over external objects’. What this says, in effect, is that what we can be certain of knowing is the content of our own experience. The rest of our knowledge, if any, is inferred from this experience. Few philosophers today accept the priority of experience in any of its forms, yet no genuine alternatives to it have yet been produced. Stroud’s argument can be taken as a call to proceed with that project. And what would prevent Stroud from joining in it himself?

Scepticism, which denies that knowledge is possible, finally depends, perhaps paradoxically, on the assumption that ‘knowledge’ designates a single relation, that at least a unitary conception of knowledge exists. The force of this view, which Stroud does not discuss, is that all our beliefs are ultimately justified (if they are justified at all) in the same way. Justification is in all cases the same, whether it concerns a belief that is ordinary or scientific, moral or aesthetic, theological or practical, individual or social. Without such a view we could not even ask whether Descartes seated by the fire constitutes a typical case of ‘the human cognitive situation’ in the first place.

If, by contrast, justification always depends on pragmatic factors, on specific practices in specific contexts, if there is no single practice called ‘justification’, then the pressure of Scepticism may appear less urgent. One may even fail to feel the pressure of doing philosophy in order to arrive at a unitary account of justification – at a ‘theory of knowledge’.

These three asumptions indicate the distance between modern and ancient Scepticism. ‘Philosophical scepticism,’ Stroud writes, ‘is a “benefactor of human reason” ’ because it leads to ‘a more accurate understanding of ... our familiar everyday knowledge’ and perhaps to ‘a general account ... “a theory of knowledge” ’ as well. Modern Scepticism is internal to philosophy. It is used by philosophers in order to reach philosophical conclusions. It is also narrow in range and theoretical in intent. It questions, often provisionally, only the notion of knowledge ‘no matter what convictions, beliefs or opinions we continue, perhaps inevitably, to hold.’ And, in an ironic reversal, it is a dogmatist view. Modern Scepticism begins with the idea that we at least have knowledge of our own experience and often ends with the hope that the rest of our beliefs amount to knowledge as well. Sextus would have shrugged.

The books reviewed here show that Scepticism continues to attract subtle minds. And whether directly philosophical or essentially equivocal, as it is in the writings of Jacques Derrida or Richard Rorty, Scepticism often leaves things just the way it found them. The Sceptics, according to Sextus, end up leading their lives as nature, need, law and custom, art and science, dictate. Thales, who along with Pythagoras and Democritus (not to mention Plato and Aristotle) is responsible at least for what the Sceptics find in law and custom, art and science, would have been pleased. The bottoms of wells, quite by accident, turned out to be very good vantage-points for gazing at the stars. And gazing at the stars, in turn, may have been a good preparation not only for building better wells but also for learning how to share their water.