encounter between the rationalism of Greek philosophy and the theology of omnipotence and inscrutability in Islam and was associated with Averroës, who, in his Deci- sive Treatise, tried to justify a double standard of truth for the masses and truth for the philosopher. This earned him the ire of Islamic and Christian theologians and led to the Paris Condemnations of Bishop Tempier of 1270 and 1277 in which Boethius was centrally involved, and to the attempts of St Thomas Aquinas to produce a coherent synthesis of pagan philosophy and Christian theology. The moral and intellectual privilege of the philosopher is a prominent theme in Spinoza (Ethics, proposition 41, Scholium). c.w. *subjective truth. G. F. Hourani, Averroës on the Harmony of Religion and Philosophy (London, 1961). doubt. When we doubt a proposition, we neither believe nor disbelieve it: rather, we suspend judgement, regard- ing it as an open question whether it is true. Doubt can thus be a sceptical attitude: one form of *scepticism holds that any cognitive attitude other than doubt is irrational or illegitimate—rationality requires a general suspension of judgement. The arguments employed by sceptics (for example, Pyrrhonists such as Sextus Empiricus) are thus designed to induce doubt, to shake our beliefs and certain- ties, and to force us to suspend judgement. Descartes made doubt the cornerstone of a philosoph- ical method: in order to place our knowledge on founda- tions which are genuinely secure, we should try to doubt all of our beliefs, retaining them only if they are absolutely indubitable. Ordinary empirical beliefs are threatened by the possibility that I am dreaming; as are even logical prin- ciples because I might be deceived by an evil demon. Unless I can eliminate these possibilities, I cannot escape the suspicion that all my beliefs are infected by unnoticed error. Few have been convinced by Descartes’s claims about when doubt is impossible, and many have ques- tioned his claims about the desirability of trying to extend doubt as far as possible. A problem emerges because Descartes’s arguments do not produce a genuine doubt: the possibility that I might be dreaming or deceived by a demon does not touch my everyday confidence that I will be supported when I sit down or my ordinary reliance upon elementary arith- metic. Descartes acknowledged that the doubt induced by hypothesizing an evil demon is ‘very slight, and so to speak metaphysical’: we can acknowledge the abstract possibility or appropriateness of doubt but we feel no live doubt. But many of his critics have claimed that he relied upon an inadequate, excessively ‘intellectual’ understand- ing of doubt and certainty. *Common-sense philosophers have questioned the apparent assumption that if we can conceive a possible situation incompatible with the truth of some everyday claim, then, unless we have independent grounds for ruling out that possibility, our everyday certainty is unwarranted. There are kinds of certainty (and indubitability) falling short of the absolute certainty criticized by sceptics. In 1675 John Wilkins defended our certainty that there was such a man as Henry VIII and that there are such places as Amer- ica and China. And John Tillotson insisted that ‘It is possible that the sun may not rise to Morrow morning; and yet, for all this, I suppose that no Man has the least Doubt but that it will.’ We do not hesitate to accept standards of rationality which underwrite such certainties; and it is unreasonable to follow sceptics in disregarding these standards. Doubt is made to appear a neurotic and unreasonable fear which leads us to doubt things because they cannot receive kinds of proofs which it is unreasonable to expect them to receive. They may not be beyond all possible doubt, but they are beyond all reasonable doubt. Similar arguments against the Cartesian use of the method of doubt are found in thinkers like Thomas Reid. Alongside this claim that sceptical doubts are unreason- able, we find the suggestion that they are unreal, that they are a pretence. The way in which I confidently trust that the chair will take my weight suggests that I entertain no real possibility that it is not there. Philosophers like Wittgenstein have insisted that these ‘practical certain- ties’, things we do not doubt ‘in deed’, form the true foun- dations of our knowledge: the Cartesian method of doubt misconstrues this distinctive kind of certainty as a form of intellectual assent. c.j.h. *certainty. R. Descartes, Meditations, in The Philosophical Writings of Descartes, tr. J. Cottingham et al. (Cambridge, 1985). M. J. Ferreira, Scepticism and Reasonable Doubt (Oxford, 1986). doxa. A Greek word signifying opinions, beliefs, conjec- tures, estimates. A very important notion in Aristotle’s philosophical methodology, where it means the ‘things that are said’ by the many or the wise regarding some problem or issue which any adequate philosophical assessment must take into account justly and properly. The *‘intuitions’ often appealed to in modern moral phil- osophy, or in John Rawls’s method of *‘reflective equilib- rium’, are all doxa, but it is not obvious that philosophical theorizing need be constrained by such things. n.j.h.d. A notable treatment is given by G. E. L. Owen in ‘Tithenai ta phainomena’, in Logic, Science, and Dialectic (Ithaca, NY, 1986). doxastic virtues: see virtues, doxastic. dread: see Angst. dreams. Hallucinations in sleep? Philosophers have con- cerned themselves with dreams in three ways. (1) Dream scepticism. The effectiveness of any self-applied waking-or-dreaming test presupposes that you did not merely dream you carried it out. Does it follow that you know neither that you are not dreaming nor any of those many things you think you know provided you are awake? 220 double truth (2) The interpretation of dreams. Freud said dreams are the (disguised) fulfilment of a (repressed) wish. Are such explanations causal, purposive, or something else? And what would vindicate or refute them? Or is the point to change dreamers rather than to understand them? (3) The concept of dreams. Given that most of what is reported as dreamt belongs to Cloud-cuckoo-land, is this remembering at all? If so, of what? And what would count as misremembering? Do dreams occur during sleep or are our waking impressions memory illusions? j.e.r.s. *psychoanalysis, philosophical problems of. C. E. M. Dunlop (ed.), Philosophical Essays on Dreaming (Ithaca, NY, 1977). Dretske, Fred I. (1932– ). American philosopher, who has made significant contributions to epistemology, meta- physics, and the philosophy of mind. In the philosophy of perception he defended the idea that there is a ‘non- epistemic’ variety of visual experience—the sense of seeing an object that is attributable in purely extensional sen- tences. (*intentionality.) In epistemology he was one of the pioneers of the ‘relevant alternatives’ approach to *know- ledge. In recent work, Dretske has offered a reductive account of the intentionality of mental states in terms of the notion of information—reliable lawlike correlation between types of phenomena. Clouds are reliably correlated with rain. There is a sense therefore in which clouds carry infor- mation about the presence of rain: they are ‘reliable indica- tors’ of rain. Dretske argues that intentionality can ultimately be reduced to such reliable indication. t.c. *perception; experience. Fred I. Dretske, Knowledge and the Flow of Information (Oxford, 1981). —— Naturalizing the Mind (Cambridge, Mass., 1995). dualism. The theory that mind and matter are two dis- tinct things. Its most famous defender is Descartes, who argues that as a subject of conscious thought and experi- ence, he cannot consist simply of spatially extended mat- ter. His essential nature must be non-material, even if in fact he (his soul) is intimately connected with his body. The main argument for dualism is that facts about the objective external world of particles and fields of force, as revealed by modern physical science, are not facts about how things appear from any particular point of view, whereas facts about subjective experience are precisely about how things are from the point of view of individual conscious subjects. They have to be described in the first person as well as in the third person. Descartes argued that the separate existence of mind and body is conceivable; therefore it is possible; but if it is possible for two things to exist separately, they cannot be identical. A modern form of this argument has been pre- sented by Saul Kripke, against recent forms of scientific materialism which claim that the relation of mental states to brain states is like the relation of water to H 2 O. What happens in the mind clearly depends on what happens in the brain, but facts about the physical operation of the brain don’t seem to be capable of adding up to subjective experiences in the way that hydrogen and oxygen atoms can add up to water. Theoretical identifications of which both terms are physical and objective don’t provide a model for identifications where one term is physical and the other is mental and subjective. However, while there are problems with the identification of mind and brain, it is not clear what other kind of entity could have subjective states and a point of view, either. Substance dualism holds that the mind or soul is a separ- ate, non-physical entity, but there is also *double aspect theory or property dualism, according to which there is no soul distinct from the body, but only one thing, the person, that has two irreducibly different types of properties, men- tal and physical. Substance dualism leaves room for the possibility that the soul might be able to exist apart from the body, either before birth or after death; property dual- ism does not. Property dualism allows for the compatibil- ity of mental and physical causation, since the cause of an action might under one aspect be describable as a physical event in the brain and under another aspect as a desire, emotion, or thought; substance dualism usually requires causal interaction between the soul and the body. Dualis- tic theories at least acknowledge the serious difficulty of locating consciousness in a modern scientific conception of the physical world, but they really give metaphysical expression to the problem rather than solving it. The desire to avoid dualism has been the driving motive behind much contemporary work on the mind– body problem. Gilbert Ryle made fun of it as the theory of ‘the *ghost in the machine’, and various forms of *behav- iourism and *materialism are designed to show that a place can be found for thoughts, sensations, feelings, and other mental phenomena in a purely physical world. But these theories have trouble accounting for *consciousness and its subjective *qualia. Neither dualism nor material- ism seems likely to be true, but it isn’t clear what the alter- natives are. t.n. *identity theory; mind–body problem. René Descartes, Meditations. S. Kripke, Naming and Necessity (Cambridge, Mass., 1980). C. McGinn, The Character of Mind (Oxford, 1982). Ducasse, Curt John (1881–1969). A French-born Ameri- can philosopher who taught at Brown University and was an advocate and practitioner of *analytical philosophy before it became the dominant mode in the English- speaking world. In contrast to most analytical philoso- phers, however, Ducasse had a comprehensive philosoph- ical system. ‘When any philosophically pure-minded person sees a brick strike a window and the window break’, Ducasse said in his attack on Hume on *causation, ‘he judges that the impact of the brick was the cause of the breaking, because he believes that impact to have been the only change which took place then in the immediate environment of Ducasse, Curt John 221 the window.’ According to his adverbial view of sensing (influential on the epistemology of his student Chisholm), when we sense a red colour, the red colour is not a sub- stantive but refers to a way of sensing—‘I see redly’. Ducasse was also celebrated for his lifelong fascination with psychical research. p.h.h. P. H. Hare and Edward H. Madden, Causing, Perceiving and Believ- ing: An Examination of the Philosophy of C. J. Ducasse (Dordrecht, 1975). duck-rabbit. A visually ambiguous drawing, introduced by J. Jastrow. It can be perceived either as a duck or as a rabbit, but not both simultaneously. It constitutes the starting-point for Wittgenstein’s study, in Philosophical Investigations, ii. ix, of aspect perception. It exemplifies the concept-laden character of some forms of *perception, and provides a connecting link to examination of the per- ception of speech and writing. p.m.s.h. *illusion. S. Mulhall, On Being in the World: Wittgenstein and Heidegger on See- ing Aspects (London, 1990), 1–52. Duhem, Pierre (1861–1916). French physicist, philoso- pher, and historian of science most famous for Quine’s use of his thesis that theories cannot enjoy empirical conse- quences of their own, but only in complexes. Duhem’s stated position varied from the thoroughgoing *instru- mentalism of To Save the Phenomena to a *conventionalism tinged with what has been interpreted as structural real- ism in Aim and Structure. Whether any properties of the world can be inferred from the success of a physical the- ory, the power responsible for these successes was, for Duhem, the mathematical structure beloved of ‘French’ minds, rather than the strings and pulleys of English atom- ism. Thus Duhem was enamoured of—and contributed to—phenomenological thermodynamics as expressed in abstract differential equations. n.c. r.f.h. *holism. P. M. M. Duhem, To Save the Phenomena, tr. E. Doland and C. Maschler (Chicago, 1969). ——The Aim and Structure of Physical Theory, tr. P. Wiener (Prince- ton, NJ, 1982). Dühring, Eugen (1833–71). A prominent socialist intellec- tual, Dühring was originally trained as a lawyer and came to teach philosophy and economics at the University of Berlin. Building on Feuerbach’s materialism, Dühring developed an atheistic optimism allied to a fairly mech- anistic sort of *positivism. From this he constructed pro- posals for the reform of society which were distinctly utopian. In 1875 he became the object of a polemic by Engels, entitled Anti-Dühring, in which Engels counter- posed his own, and allegedly Marx’s, *dialectical material- ism to the supposedly cruder *materialism of Dühring. d.m cl. Dummett, Michael (1925– ). British philosopher of lan- guage, logic, and mathematics, noted for his exposition of Frege’s philosophy and defence of *anti-realism. Dum- mett characterizes anti-realism in terms of a denial of the principle of *bivalence—the principle requiring that any assertoric sentence is either true or false. To hold that this principle fails for sentences concerning a given domain of discourse—such as past events, other minds, or mathe- matics—is to be, in Dummettian terms, an anti-realist with respect to that domain. Dummett’s anti-realism stems from his approach to the theory of *meaning, and has affinities with verificationism. Like Davidson, Dummett believes that a learnable language must have a compositional semantics, but rather than associate sen- tence-meaning with realist truth-conditions Dummett associates it with assertibility-conditions, because where- as a child can be taught to recognize circumstances in which evidence suffices to justify the assertion of a sen- tence, it cannot be taught to grasp circumstances in which a sentence would be true independently of any evidence that might bear upon its truth. Consequently, if a sentence about (say) the past is such that neither it nor its negation is justifiably assertible, we can have, it seems, no genuine grasp of what it would be for that sentence to be true and its negation false, or vice versa. Dummett’s views on anti-realism and the theory of meaning are intimately related to his work in logic and the philosophy of mathematics, especially his sympathetic treatment of *intuitionism. Yet at the same time he has perhaps done more than any other commentator to pro- mote interest in Frege’s philosophy of language and math- ematics and to elevate Frege ahead of Russell, Moore, and Whitehead as founder of modern analytic philosophy—all this despite Frege’s strong Platonist leanings, which run quite counter to intuitionist precepts. e.j.l. *normalization; Tarot. M. Dummett, Frege: Philosophy of Language, 2nd edn. (London, 1981). —— Truth and Other Enigmas (London, 1978). —— The Seas of Language (Oxford, 1993). —— Truth and the Past (New York, 2004). Duns Scotus, John (c.1266–1308). Scholastic philosopher, the ‘Subtle Doctor’, the original ‘dunce’, and, for Gerard Manley Hopkins, ‘Of realty the rarest-veined unraveller’, who was one of the great Christian *medieval philoso- phers. His critical attitude to Aquinas led to Ockham’s more radical criticism. The details of Scotus’ life are uncertain. He was born in Scotland and became a Franciscan. He did not live to revise his writings, and they are only now being properly edited and disentangled from spurious works. His genius lies not only in the novelty of his doctrines but also in his meticulous exposition and dissection of arguments, even when he accepted their conclusions. He believed, for example, in the immortality of the soul but regarded none of the arguments for it as conclusive; and, in discussing the proofs of God’s existence, he took pains to distinguish 222 Ducasse, Curt John those cases in which an infinite regress of causes is vicious (and thus needs to be curtailed by the postulation of an uncaused entity) from those in which it is not. In his discussions of theological questions, he elab- orated several doctrines that diverge from *Thomism. (He rarely mentions Aquinas, however, but attacks less eminent opponents such as Henry of Ghent.) He rejects negative theology, since ‘negation is only known through affirmation’. Being (the subject of metaphysics), and other terms predicated of both God and creatures, are univocal. I may be certain that something, e.g. God, is, or is wise, but uncertain whether he is finite or infinite, created or uncre- ated; but my concept of being or of wisdom will be the same whichever of these alternatives is true. (This argu- ment is open to an objection: If I overhear someone saying ‘That’s too hard’, I may be certain that something is too hard, and that the speaker believes it to be so, while being uncertain whether what is referred to is a chair or a ques- tion; but it does not follow that ‘hard’ is univocal as applied to chairs and to questions. Analogously, the fact that we can believe God to be, while remaining uncertain of his categorial status, does not demonstrate the (other- wise plausible) conclusion that ‘being’ applies univocally to entities in different categories.) Moreover, concepts are derived from our acquaintance with creatures. If the con- cepts applied to God are not the same concepts, we can nei- ther give a sense to them nor validly argue from premisses about creatures to truths about God. He criticized a position close to Ockham’s nominalism, arguing that things have ‘common natures’, e.g. the humanity common to Socrates and Plato. But he rejected the Aristotelian–Thomist view that individual things are distinguished by their (designated) matter and are thus not truly intelligible. Tweedledum is distinct from Tweedledee in virtue of his haecceitas or thisness, a formal feature intelligible to God and angels if not to fallen humanity: ‘the ultimate specific difference is simply to be different from everything else’. The distinction between an entity’s common nature, its *haecceity, and its existence is intermediate between a real and a conceptual distinction, namely, an ‘objective formal distinction’ (distinctio formalis a parte rei). This type of distinction also obtains between the divine attributes, the powers of the soul, etc. The will, both of God and of man, is distinct from the intellect and not deter- mined by it. The will does not necessarily choose the summum bonum even when it discerns it intellectually. Will, not intellect, plays the main part in our free ascent from worldly perfection to beatitude, the love of God. God too is free, and the world does not emanate from him by intel- ligible necessity, but results from his freely given love. By freely willing the moral law God makes it binding on us: ‘To command pertains only to the appetite or will’. But the content of the primary precepts, e.g. that one should not worship other gods, is not determined by God’s will; God wills them because they are intrinsically self-evident, and we cannot be dispensed from them. There are, how- ever, secondary precepts which, though in harmony with the primary, are neither derivable from them nor self- evident, and their content as well as their obligatoriness depends on God’s will; from these he can dispense us. Scotus is half-way between Thomism and Ockham’s view that all law stems from the will of God alone. He was less fond than Aquinas of Aristotle’s proof of God’s existence from the occurrence of motion, not because the proof is invalid, but because God transcends the physical realm: ‘it is a more perfect and immediate knowledge of the first being to know it as first or necessary being than to know it as first mover’. Proofs of God’s exist- ence must be a posteriori. But Anselm’s *ontological argument is a ‘probable persuasion’, if not a demonstra- tive proof, as long it is appropriately ‘coloured’. That is, Scotus (like Leibniz) added the premiss that the most per- fect being is possible, i.e. can be ‘thought without contra- diction’, but held that we cannot prove that it contains no contradiction from our inability to detect one. His immense influence extends to Peirce and Heideg- ger as well as to his medieval followers. m.j.i. F. Copleston, A History of Philosophy, ii: Mediaeval Philosophy, pt. 2: Albert the Great to Duns Scotus (Westminster, Md., 1950). R. Cross, Duns Scotus (New York, 1999). E. Gilson, Jean Duns Scot: Introduction à ses positions fondamentales (Paris, 1952). T. Williams (ed.), The Cambridge Companion to Duns Scotus (Cam- bridge, 2003). A. B. Wolter, The Philosophical Theology of John Duns Scotus (Ithaca, NY, 1990). Durkheim, Émile (1858–1917). From a French rabbinical family, he started his career teaching secondary-school philosophy, then sociology at the Universities of Bor- deaux and Paris. He claimed that societies are irreducible entities, the laws governing which could not be derived from biology or psychology. ‘Collective representations’ of a society, such as social traits, customs, legal systems, languages, and ‘group emotions’ are said to ‘exist outside the individual consciousness’, on which they have an effect greater than the mere sum of the effects of other individuals. So sociology is a distinctive science with a dis- tinctive subject-matter (which happily prevents sociolo- gists being redundant). What is ‘normal’ is relative to particular stages of society. Lack of social norms, or conflict between them, produces ‘anomie’, a moral lawlessness. Durkheim attempted functional explanations of the division of labour, primitive religions, etc. in terms of societies’ (not individuals’) needs. a.j.l. *anomie; society. Émile Durkheim, The Rules of Sociological Method, tr. W. D. Halls, ed. Stephen Lukes (London, 1982). Dutch book. A Dutch book has been made against you if you accept odds and make bets in such a way that you lose regardless of the outcome. For example, suppose you bet $4 at 5–2 that the Canadiens will win the Stanley Cup, and $4 at 5–2 that the Nordiques will win. Hedging, you then Dutch book 223 224 Dutch book bet $7 at even odds that neither will win. Whoever wins, you lose: Dutch book! j.j.m. B. Skyrms, Choice and Chance, 3rd edn. (Belmont, Calif., 1986). Dutch philosophy: see Netherlands philosophy. duty. Along with the concepts of *‘ought’ and *‘obliga- tion’, the concept of duty expresses moral action as demanded or required. ‘The moral law’, wrote Kant, is, for us, ‘a law of duty, of moral constraint.’ How is this cluster of concepts related to the contrasting cluster centring upon ‘good’ and the realization of value? For some moral- ists (including Kant again), ‘Duty and obligation are the only names’ for ‘our relation to the moral law’ (Critique of Practical Reason). For others, our duties, though not reducible to different terms, make sense only as regulating human life so as best to achieve good ends and to respect rational and sentient beings. Certain performances, such as promise-making, gener- ate duties to act in quite specific ways: other duties result from special relationships—parent to child, doctor to patient: others again are owed to living beings simply on the ground of their sentience or their rational, personal status. r.w.h. *supererogation. D. P. Gautier, Practical Reasoning (Oxford, 1963). I. Kant, Critique of Practical Reason (1788); tr. L. W. Beck (Chicago, 1949). O. O’Neill, Constructions of Reason (Cambridge, 1989). Dworkin, Ronald (1931– ). American Professor of Jurisprudence at Oxford then London, whose explicitly liberal theory of law radically extends Hart’s ‘internal viewpoint’ by treating philosophy of law as a primarily normative contribution to political, particularly judicial, deliberation. Moral, political, and legal theory should be not goal or duty-based but *rights-based, upholding principles (rights) over policies (collective goals), so as to respect everyone’s right to equality of concern and respect (Taking Rights Seriously (1977)). This fundamental right requires that governments be neutral about worthwhile or worthless forms of life, and support even suicidal individual self-determination (Life’s Dominion (1993)). Such principles are already part of the law; ‘creative interpretation’ (Law’s Empire (1986)), seeking to make the law the best it can be, legally authorizes substantial transformations of ‘settled law’ by judges duty-bound to apply only law. j.m.f. *law and morals; law, philosophy of; moral scepticism. Ronald Dworkin, A Matter of Principle (Cambridge, Mass., 1985). Stephen Guest, Ronald Dworkin (Edinburgh, 1992). dyadic: see relations. earliest known philosophical term: see apeiron. Earman, John (1942– ). American philosopher, a member of the History and Philosophy of Science department at the University of Pittsburgh. He is best known for his work in the history and philosophy of modern physics. An early proponent of casting philosophical problems about space and time as conjectures within the mathematical language of space-time structure, he has probed, in the words of an early paper, the ‘thicket of problems growing out of the intersection of mathematics, physics, and metaphysics’. He has explored this thicket in books on determinism, absolute and relational theories of space and time, and acausal space-time structure. Alongside books on Bayesian inference, and more recently on Hume’s argument against miracles, he has maintained his tech- nical expertise with work on inflationary cosmologies and quantum field theories. e.r.j. John Earman, A Primer on Determinism (Dordrecht, 1986). —— World Enough and Spacetime (Cambridge, Mass., 1989). ecological morality: see environmental ethics. economic determinism: see base and superstructure. economics, philosophy of. The philosophy of economics may be taken as the *philosophy of social science run with economic examples, in which case there is not much to be said specifically about it. Or it may be taken to designate a more or less distinct area of inquiry: one which overlaps with the philosophy of social science, as it is bound to do, but which is motivated by distinctively economic con- cerns. I shall take it here in the latter sense. Economics is a highly distinctive approach to social theory, and the philo- sophical problems which it raises are cast in nice perspect- ive by contrasting it with other social disciplines of thought. I will consider the contrasts it displays with socio- logy, psychology, and politics. The most striking contrast between economics and sociology—economics in the dominant neoclassical sense and sociology in the traditional, Durkheimian mould—is that economics is individualistic, sociology not. (*Social philosophy.) The individualist thinks that none of the aggregate patterns and pressures revealed in social science—or revealed otherwise—gives the lie to our gen- eral sense of the individual human agent, while the non- individualist denies this. The individualist holds that human agents conform to our commonplace folk psy- chology, being more or less rational in the attitudes they form and the choices they make. The non-individualist believes that individuals take second place, in a manner inconsistent with common sense, to the sorts of social regu- larities that social science is well equipped to reveal. For example, he may say that there are social regularities that are predetermined or predestined to obtain, in such a way that individuals are bound to act as the regularities require: they are bound to have the attitudes that lead by ordinary psychology to suitable actions; or they are bound, at whatever cost to their attitudinal coherence— they may ‘go on the blink’—to display the behaviour involved. The debate between *individualism and non- individualism does not have much prominence in the philosophy of economics, because individualism reigns almost unquestioned; unlike certain sorts of sociology, economics has never suggested that it has a new and iconoclastic image of the human being to offer. But a related, methodological question does often figure in current debates. Assuming that economics is individual- istic in the more or less ontological sense explicated, does this mean that it must also be methodologically individu- alistic? Does it mean that economics must deny validity and interest to *explanations that relate events or patterns to aggregate antecedents: say, that it must reject as ill-con- ceived the sort of explanation that traces a rise in crime to a rise in unemployment or a decrease in religious practice to growing urbanization? This question is of particular relevance, because many so-called macro-economic explanations appear to be aggregate in character. Whether or not economics can countenance such aggregate-level explanations, it is associated in practice with a style of individual-level explanation that marks a contrast with the explanations preferred in traditional sociology. This style of explanation involves two elem- ents, one psychological, the other institutional. The psychological element suggests that given a certain circumstance or change or whatever, it is unsurprising that people should generally—or at least in significant E numbers—come to behave in a certain way. The institu- tional element then goes on to show that given this shift in overall behaviour, there are bound to be certain conse- quences—in all likelihood, unintended consequences— that make for an aggregate change. If the consequences are thought of as beneficial, then the pattern identified in the explanation is traditionally described, in a phrase of Adam Smith’s, as an invisible hand; if they are thought of as harmful, it is sometimes described as an invisible back- hand or an invisible foot. Invisible hand and invisible backhand explanations are the very stuff of economic theorizing in the received, neoclassical mould. So much for the contrast between economics and soci- ology. A second contrast that points us towards distinctive features of economics is that with psychology. Here there are a number of things to notice. Economics has tradition- ally been more or less behaviouristic in orientation, pre- ferring to build a picture of the human subject out of actions displayed rather than on a reflective or introspect- ive basis. Again, economics has traditionally been not just behaviouristic, but also rationalistic. It has assumed that decision theory is on the right tracks in seeking to explain human behaviour by reference to the maximization of expected utility: the maximization of expected prefer- ence-satisfaction. And, finally, economics has tended, at least in practice, to be egocentrically reductionistic, assuming that the preferences which human agents seek to satisfy are, on the whole, self-concerned or egoistic desires. These features of economics put it in contrast with many traditions of psychological theorizing and they even create tensions with our commonplace psychology. Economic psychology may not reject commonplace psych- ology in the manner of non-individualistic theories, but it gives a controversial gloss to much that that psychology contains. (Michael Bacharach and Susan Hurley (eds.), Essays in the Foundations of Decision Theory.) The status of the psychological assumptions that eco- nomics makes is a matter at the core of the philosophy of economics: there is much discussion both of the necessity for such assumptions and of the plausibility of the assump- tions made. The questions raised have been greatly sharp- ened with the increasing application of economic method, not just in the explanation of market and related behav- iour, but also in the explanation of behaviour outside the market: for example, in the explanation of social inter- action, as in so-called exchange theory, and in the explan- ation of political behaviour, as in what is known as the theory of public choice. Is it really reasonable to treat social and political agents as hard-headed and hard- hearted calculators of the kind that economics projects into the market-place? Some have thought that it is, on the grounds that human beings are always unconsciously of this cast of mind. Others have sought less dramatic means of vindicating the contribution that economics can make in non-economic domains. (Philip Pettit, The Common Mind, ch. 5.) The final contrast that I want to mention is between economics and politics. Traditional political thinking, especially normative political thinking, is characterized by two features. First, a willingness to contemplate exogen- ous ideals—say, ideals of equality or liberty or solidarity— in the assessment of social and political institutions. And second, a tendency to assume that the main task in nor- mative thought is just to argue for the ideals introduced and to provide a sense of what their institutionalization would involve. Economics breaks with both of these dis- positions, being associated with quite a different sense of how normative thinking should go. (Geoffrey Brennan, ‘The Economic Contribution’.) Perhaps the main assumption of normative thinking in traditional economic circles—an assumption now fre- quently questioned (Amartya Sen, Choice, Welfare and Measurement; John Broome, Weighing Goods)—is that it is inappropriate to judge the institutions of a society except by reference to the preferences of the people they affect. This assumption is broadly utilitarian in character—unsur- prisingly, since the history of economic thought was closely tied up with the utilitarian movement in the last century—but economics has given its own distinctive twist to the utilitarian thought. Arguing that we cannot compare preference-satisfaction across individuals and therefore cannot compute the level of total preference- satisfaction in a society—the exercise is not epistemolog- ically feasible—economics has explored other ways of developing the preference-based idea. A development that gained momentum in the 1930s yielded the Pareto- criterion, according to which one arrangement is better than another if and only if it satisfies the preferences of some and does not frustrate the preferences of any. This criterion has been at the centre of what came to be known as welfare economics. Another, more recent development, and one which has made for a connection with philosoph- ical discussions, suggests that one arrangement is better than another if and only if it would be preferred by suitable parties in a suitable collective choice. This contractarian development is of great contemporary importance. But not only is economics distinguished by the role it gives to preferences in normative thinking, it is also marked off from traditional normative thought by the emphasis it places on feasibility. It is not enough, so eco- nomics suggests, to be able to identify a plausible ideal and to describe what it would institutionally require. What is also necessary is to be able to show that the institutional- ization in question represents a feasible way of realizing the ideal: one that is currently accessible and one that would remain in place, if once established. There are many products of this concern with feasibility, among them the minimalist approach of F. A. Hayek, which argues that the information that good government would require in a more-than-minimal state is never going to be reliably available. Public choice theory and social choice theory are also products of this concern and they have had a major influence on contemporary political philosophy. (Iain McLean, Public Choice.) The public choice theorist argues that it is silly to prescribe a form of government, or to allocate certain 226 economics, philosophy of responsibilities to those in government, unless one has reason to believe that the arrangement is institutionally robust: at the least, it won’t lead to worst results than would otherwise ensue. Public choice theory is meant to enable us to deal with the problem raised: to give us an idea of what to expect from those in government under this or that institutional amendment. The social choice theorist, on the other hand, is concerned with more abstract matters. He argues that there are many ways of aggregating individual preferences into a social prefer- ence-ordering—many ways of moving from what you want and what I want to what we should prefer as a group—that cannot simultaneously satisfy various attract- ive conditions; the most famous result in the area is Kenneth Arrow’s impossibility theorem. Social choice the- ory castigates traditional theory for being over-relaxed about such matters and tries to explore questions about the aggregation of preferences in a systematic manner. p.p. *psychology and philosophy; capitalism; rational choice theory. Michael Bacharach and Susan Hurley (eds.), Essays in the Founda- tions of Decision Theory (Oxford, 1991). Geoffrey Brennan, ‘The Economic Contribution’, in R. E. Goodin and Philip Pettit (eds.), A Companion to Contemporary Political Philosophy (Oxford, 1993). John Broome, Weighing Goods (Oxford, 1991). Iain McLean, Public Choice (Oxford, 1987). Philip Pettit, The Common Mind (New York, 1993). Amartya Sen, Choice, Welfare and Measurement (Oxford, 1982). economics and morality. All societies must make moral decisions on how their resources are to be distributed. But for every resource allocation that brings benefits, cond- itions of scarcity mean that there are associated burdens, or costs. The discipline of economics offers a method of analysis for comparing alternative distributions by clarify- ing what is at stake, in terms of the burdens associated with the benefits. However, the relationship between eco- nomics and morality is not straightforward. Neo-classical (Welfare) Economics (NCE), by far the most influential modern school, is based on a number of theoretical pre- suppositions that raise moral questions. Though economists may seek a value-free analytic technique, specific value commitments underpin the way they conceptualize benefits and burdens. They employ a metric of comparison for alternative distributions, within which burdens can be counted against benefits. Yet, in order to do this in a ‘value-neutral’ framework, they make benefit gains relative to each person’s own conception of value. This they do predominantly by adopting the metric of individual welfare, understood as the maximisation of preference satisfaction (although some alternatives have been offered). To this individualized notion of welfare, economists add the principle that distributions which optimize wel- fare gains across individuals, according to the Pareto improvement and optimality theorems, are economically superior. Accordingly, the institution of the market is cen- tral to NCE. Under ideal conditions, a market is ‘efficient’ in its allocation of resources because, free, preference- based exchanges optimally maximize welfare (preference satisfaction). These basic principles of individual welfare maximiza- tion, aggregation, the imperative to optimize gains, and market efficiency are not morally neutral; nor is it clear that they are compatible with important moral values. Trivially, distributions which satisfy these conditions may fail to respect rights or uphold communally important values, such as equality and fairness, unless one falsely identifies optimizing welfare allocations with fairness, or assumes that they respect rights. Less trivially, the idea that well-being can be understood in terms of maximizing preference satisfaction is controversial, as is claiming superiority for set-ups which optimally maximize it across individuals. Some goods, such as health or personal free- dom, for example, may not be commensurable with other preference-satisfiers which can be exchanged on a market. So even asking people to consider goods, like the envir- onment or human relationships, solely in terms of prefer- ences, especially those which can be expressed in market exchanges, seems morally distorting. Nor is it obviously ethically acceptable to treat persons in terms of one ‘snap- shot’ of their preferences. These can change over time, and, more importantly, they can be responsive to reasons. Yet the value of political participation—influen- cing communal decisions through argument and discus- sion—is not itself expressible in terms of preferences in the market-place. Nevertheless, might not NCE help us with choices where all other morally relevant factors are taken into account? Whilst there is nothing wrong with redu- cing burdens, a moral problem remains for NCE. What counts as a burden, that which is costly, and what costs it is appropriate to associate with a choice, are questions themselves sensitive to moral judgement. It seems wrong, for example, to see the discharge of a moral duty as some- thing for which one could be compensated. NCE also employs a theory of rationality for individuals (rational choice theory) and for collectives (social choice theory). This aims both to explain actual states of affairs and to evaluate choices. The theory has a normative core which characterizes rational agent motivations as maxi- mizing preference satisfaction and supplies axioms for ordering preferences. As a comprehensive description of rationality, then, the view places conditions on moral rea- soning. For some philosophers, it even forms the basis of an account of morality. One response is that a plausible notion of good and bad reasons exists which is independ- ent of actual preferences. It is not obviously rational for someone to act on their strong preferences for harming others, even if they can do so with impunity. The conclu- sions of rational choice theory, then, may sometimes com- pete with moral reasoning. The productive successes of market economies, and paucity of viable alternatives to NCE, pose a different moral problem. Can we retain market benefits, without economics and morality 227 NCE’s theoretical presuppositions? It seems possible to do this by retaining the market, within a morally deter- mined regulatory framework. However, this leaves us in need of a principle for determining to what degree personal choice (preference), versus moral duties to others, can legitimately play a role in setting economic priorities. s. m-g. *economics and philosophy; justice. E. Anderson, Value in Ethics and Economics (Cambridge, Mass., 1993). D. M. Hausman and M. S. McPherson, Economic Analysis and Moral Philosophy (Cambridge, 1996). A. Sen, On Ethics and Economics (Oxford, 1989). education, history of the philosophy of. A problem con- fronting anyone writing on the history of the philosophy of education is that many of the names mentioned in stand- ard treatments of the topic, such as Pestalozzi, Froebel, and Hebart, are unlikely to be mentioned in standard his- tories of philosophy. Conversely, many if not most of the great philosophers have had little directly to say on the subject of education, and sometimes, as with Locke, when they do say something directly it is of little consequence. Nevertheless, despite this mismatch between philosophy and educational thinking, the topic of education does raise important philosophical issues. In concentrating on phil- osophers of importance and topics central to education, my account may be found somewhat revisionary. The starting-point, though, must, as always, be Plato, whoseRepublic, though not exclusively that, is the first and greatest work in the philosophy of education. In the Republic, Plato is concerned with educating people in such a way that a just society is the outcome. Many would find this an extraordinary overestimation of the powers of edu- cation, and indeed it would be were education conceived in terms acceptable to a liberal democracy and did not involve, as Plato advocates, a form of child-farming. Plato is not concerned with the liberal ideal of individuals pur- suing their own tastes and interests. He rejects the family and private property, at least for the rulers. For Plato, the good life is characterized rather by a general turning towards what is good and true outside of us and independ- ently of us. Although this external good has left its traces deep within us, it is hidden and needs to be recovered by a process of externally directed discipline and thought. Each individual is born destined to play a particular type of role in a society which aims at the good, and will be happy when his own powers are so arranged as to enable him to fulfil this role. In such a society the rulers will possess the wisdom to guide the rest in the light of the good and the true. In the good city there will be all the usual trades and crafts, and the majority of citizens should be trained to perform them, presumably learning reading, writing, counting, and the particular skills appropriate to their trade. But educators will notice that some youths are suited by temperament to guard and guide the city; those singled out need to be both brave and gentle. To produce soldiers of the right disposition, those selected will receive an education in music and gymnastics, based on the trad- itional classical Greek models. Music, which includes liter- ature, will have to be uplifting and moral, rather than effeminate or disorderly. Poetry which shows the gods behaving disreputably and music which is barbaric or effete are ruled out. Gymnastics will train both bodies and characters. The most selfless and steadfast of the guardians are to be educated further to become rulers of the city. For this, they will need to become philosophers, lovers of wisdom, skilled in science and reasoning (or what Plato calls dia- lectic). Both rulers and soldiers are to be brought into what is in effect an armed camp within the city, and taken away from their parents. For the rest of their lives they will possess everything in common, including wives and children. The presumption is that the offspring of the initial guardians will share their qualities and form the next generation of guardians. Within the camp no one will know who his or her par- ents are. There will also be no distinction between the sexes, women being selected as guardians as much as men, and educated in the same way. Future generations of guardians will be told that they have been bred for the city. Their philosophizing is not for their own satisfaction only, but is so that they can rule and instruct the rest. Simply to recount Plato’s proposals may seem to have little philosophical point. There are, though, certain themes which have recurred in educational thinking since Plato’s time: the idea that education and individual lives are ideally for the sake of the state, not for the sake of the individual alone; the idea that education is as much about the building of character as of intelligence in our sense (which is the case even with the rulers’ philosophizing, which is all directed towards a type of static wisdom which is coterminous with moral goodness); and the idea that education is capable of transforming individual minds and characters so as to produce acceptance of a revolutionary communistic project. Plato’s doubtless exaggerated assessment of the power of education nevertheless leads him to write scathingly and brilliantly of forms of education of which he disap- proves. He writes of the schoolmaster in a democratic society who ‘fears and flatters his pupils’ and the pupils who consequently despise him, of old men who, ridicu- lously, condescend to the young, ‘imitating their juniors in order to avoid the appearance of being sour or despotic’. For Plato, although education is communistic, it cannot be democratic or, in the modern sense, child- centred. Even though within us there are the seeds or traces of wisdom, wisdom eludes the grasp of the young, who are wayward and blind and who have to be trained over many years to have the right dispositions and thoughts. Education then cannot proceed on the basis of the current interests of the young. The philosophers, indeed, have a duty, which is painful for them, of descend- ing into the *cave metaphorically occupied by the unen- lightened, so as to instruct and rule them. And wisdom is 228 economics and morality fixed and one. So a pluralistic approach to value or educa- tion is rejected. Like much else in philosophy, philosophy of education continually rehearses Platonic themes: authoritarian or child-centred? dictatorial or pluralistic? collectivistic or individualistic? And there is also continual worrying at the relationship between what is already inside the child and what is to be received from without. Of course, strange transmogrifications have happened on the way. The train of thought now known as liberal education, as reflected in the writings of Michael Oakeshott, say, would agree with Plato in rejecting child-centredness, but disagree with him on the fixity of knowledge and on the nature of the state. It would agree with Plato on the importance of learning what has been discovered, but disagree with him on the closure of traditions of thought. And while it would give some thought to the importance of moral education, it would tend to view education in far more intellectualist terms than Plato, claiming that intellectual disciplines and their content are worth learning for themselves irrespect- ive of any moral improvement they bring. We have no systematic treatise by Aristotle on educa- tion, but he would have shared Plato’s suspicion of teach- ers who are afraid to appear despotic. He also made some suggestive remarks about the need to inculcate the right dispositions in the young before encouraging them to reflect on morality and politics. (Otherwise their reason- ing will be clever rather than wise.) And he endorsed the classical stress on music and gymnastics in terms similar to Plato. But, although he saw the happy individual as play- ing a role in public affairs, he saw the gaining of knowledge by individuals as an end in itself and not to be justified in terms of the contribution this might enable an educated individual to make to the state. For Aristotle men have intrinsic desire to know and understand, which it is part of their nature to pursue. Here Aristotle was close to Socrates’ dictum that the unexamined life is not worth living. He was also closer to the individualism of Socrates than to the Plato of the Republic. During the Christian era, Platonic themes resurface, notably in the writings of St Augustine. Human nature needs to be turned to the light because of original sin and its enmeshment in body. The young are not yet rational. They need instruction in the basic subjects. The content of what is taught is censored, and the aim of elementary edu- cation will be the prevention of idleness. Only in the higher stages, with the study of philosophy and theology, will anything like full rationality be possible. Augustine sees God as the source of all truth and analyses learning quasi-Platonically as a form of opening ourselves to an inner divine illumination, an opening which, as with Plato, does not preclude didactic teaching methods. While not accepting Augustine’s Platonic view of knowledge, most Christian writers on education up to the time of Luther and St Ignatius of Loyola saw education in terms of the salvifically necessary transmission of truths established and revealed. Even with the recovery of pagan learning in the Renaissance the stress is on imparting to the young what has been learned. There is, though, some- thing of a sea-change, in the seventeenth century, with its dismissal of past authority, and stress on individual experi- ence. Thus Francis Bacon, who believed that the truth about nature would be manifest to the individual who engaged in presuppositionless observation, opposed the foundation of Charterhouse school on the grounds that its curriculum was to be based on the ancient classics. He wrote: ‘what happiness it would be to throw myself into the river Lethe, to erase completely from my soul the memory of all knowledge, all art, all poetry; what happi- ness it would be to reach the opposite shore, naked like the first man.’ Education should, therefore proceed by the learner making his own observations and discoveries, without external direction. Locke was not so sanguine as Bacon about the possibility of learning much about the world through the senses, although he believed with Bacon that we have no other access to the world. Locke, accordingly, emphasized the moral aspects of education, at the expense of the intellectual and scientific. But he agreed with Bacon’s thoroughly utilitarian approach to knowledge and to education, both, in Bacon’s terms, to be directed to ‘the relief of man’s estate’. Throughout the seventeenth and eighteenth centuries, schools and universities went on traditionally. The new philosophies of Bacon and Descartes which stressed indi- vidual discovery and reasoning and which downplayed didactic instruction had little effect on curriculum or peda- gogy. It is indeed doubtful that any philosophy had any significant effect on the practice of education prior to the beginning of the nineteenth century, when the ideas of Rousseau and his followers Pestalozzi and Froebel began to make an impact. Rousseau’s Émile (1762), in common with many of his other writings, is a sustained criticism of civilization as it existed in Rousseau’s time. Although, in Rousseau’s words at the start of Émile: ‘God made all things good, man meddles with them and they become evil.’ We are born free, but live in chains. Our first natural impulses are always right, yet society, by encouraging envy and vanity, makes us into civilized monsters, suffering and causing suffering. The child, moreover, is not a miniature adult, but a creative being with its own particular needs and desires, which should be allowed to ‘indulge its sports, its pleasures, its delightful instincts’. Émile was a heady brew, combining nature-worship, child-centredness, an emphasis on doing and discovery at the expense of reading and being taught, and a pervasive hatred of the existing order of things, from which the child must be protected. Its actual proposal, to allow each Émile to develop ‘naturally’ and in isolation from society, under the exclusive tutelage of a Rousseauian guide or, in today’s terms, ‘facilitator’, is as impracticable as Plato’s. Nevertheless despite an ineradicable lack of clarity about what Rousseau means by nature and uncertainty about the benefit to be gained by following nature’s impulses, Émile’s influence can be seen in every primary school in the Anglo-Saxon world. (France, perhaps surprisingly, has so far remained education, history of the philosophy of 229 . arranged as to enable him to fulfil this role. In such a society the rulers will possess the wisdom to guide the rest in the light of the good and the true. In the good city there will be all the usual. of Socrates than to the Plato of the Republic. During the Christian era, Platonic themes resurface, notably in the writings of St Augustine. Human nature needs to be turned to the light because. regularities require: they are bound to have the attitudes that lead by ordinary psychology to suitable actions; or they are bound, at whatever cost to their attitudinal coherence— they may ‘go on the blink’—to