emphasis he places on *consciousness as an intrinsic fea- ture of the mind, put him at odds with behaviouristic, functional, and other materialistic theories of mind. For Searle, although the mind emerges from the body, it pos- sesses an ineliminable subjective character with which materialistic accounts cannot adequately deal. In relation to this claim, he uses his famous *Chinese room argument to show that even though a ‘system’ (a computer and a person) inside a room can manipulate Chinese symbols, it does not necessarily operate on the level of meaning. To do that, mental (intentional) concepts need to be intro- duced into the system. n.f. J. R. Searle, Speech Acts (Cambridge, 1969). —— The Rediscovery of the Mind (Cambridge, Mass., 1992). secondary qualities: see primary and secondary qualities. seeing as. In his later writings, Wittgenstein showed an interest in the phenomenon to which the Gestalt psycholo- gists had drawn attention, of seeing (or hearing, or, . . .) something assomething. The *duck-rabbit is an example: a picture that can be seen either as a duck or as a rabbit. Part of Wittgenstein’s interest in this phenomenon had to do with his rejection of a naïve account of *perception; he took the interpretation of what is seen to be less separable from seeing itself than empiricist philosophers had been wont to think. But perception was not his only concern. We see one continuation of a number-series as ‘more natural’ or ‘sim- pler’ than another; see one grouping of objects in a class as ‘cutting Nature at the joints’, another not; and so on. Our use of concepts depends on ‘seeing as’. r.p.l.t. *illusion, arguments from. L. Wittgenstein, Philosophical Investigations, tr. G. E. M. Anscombe, 3rd edn. (Oxford, 1967). self. The term ‘self’ is often used interchangeably with *‘person’, though usually with more emphasis on the ‘inner’, or psychological, dimension of personality than on outward bodily form. Thus a self is conceived to be a sub- ject of consciousness, a being capable of thought and experi- ence and able to engage in deliberative action. More crucially, a self must have a capacity for *self-consciousness, which partly explains the aptness of the term ‘self’. Thus a self is a being that is able to entertain first-person thoughts. A first-person thought is one whose apt expression in language requires the use of the first-person pronoun ‘I’, or some equivalent *indexical expression. However, it may not be right to insist that a self be capable of express- ing its thoughts in language—even its first-person thoughts. Happily, we possess locutions for ascribing first- person thoughts to others without implying that they are capable of articulating those thoughts. One such locution is the ‘he himself’ construction. Thus if I ascribe to Fred the thought that he himself is fat, I ascribe to him a thought whose apt expression in English by Fred would be ‘I am fat’, though I do not imply that Fred is capable of so expressing that thought. Note that we must distinguish this thought from a similar third-person thought that Fred might have about himself, whose apt expression in Eng- lish by Fred might be ‘Fred is fat’ or ‘That person is fat’ (the latter said by Fred in reference to a person he sees reflected in a mirror, not realizing that it is himself that he sees). It is plausible to require of a self not only a capacity to entertain first-person thoughts but also the possession of certain kinds of first-person knowledge. For example, it seems right to insist that a self must know, of any of its pre- sent, conscious thoughts, experiences, and actions, that they are its own. This is why the response of Mrs Grad- grind (in Charles Dickens’s Hard Times), when asked on her sick-bed whether she was in pain, strikes us as so bizarre: ‘I think there’s a pain somewhere in the room, but I couldn’t positively say that I have got it.’ Our possession of such self-knowledge is connected with the phenom- enon of ‘immunity to error through misidentification’ (Sydney Shoemaker). An example involving memory is provided by the apparent absurdity of supposing that I might accurately remember (as it were, ‘from the inside’) a meal in a restaurant attended by a number of people, and yet be in some doubt about whether I was one of those people. (As against this, however, Derek Parfit has argued that we could in principle inherit ‘quasi-memories’ from other people, including first-person ‘memories’ of what they, but not we, had done.) So far we have largely been concerned with the meaning of the term ‘self’, that is, with the essential characteristics of selfhood. But metaphysicians are also interested in exploring the nature of the self, that is, what sort of *thing the self is, if indeed it is a ‘thing’ at all. In traditional terms, a distinction may be drawn between substantival and non- substantival theories of the self, the former contending that the self is a *substance, either physical or non-physical, the latter that it is a mode of substance. Philosophers like Hume, who regarded the self as ‘nothing but a bundle of different perceptions’, effectively treat the self as belonging to the category of modes. A problem with the Humean approach is that perceptions—that is, thoughts and experi- ences—seem to depend for their identity upon the identity of the selves who possess them, which implies that percep- tions are modes of selves and hence that the latter have the status of substances vis-à-vis their thoughts and experi- ences, rather than being reducible to them. e.j.l. *homunculus. D. Parfit, Reasons and Persons (Oxford, 1984). S. Shoemaker, Self-Knowledge and Self-Identity (Ithaca, NY, 1963). B. Williams, Problems of the Self (Cambridge, 1973). self, bundle theory of the: see bundle theory of the self. self-consciousness. One view of self-consciousness would be that it is the *consciousness of a special kind of object, ‘the *self’. In reply, it has been claimed that just as the eye cannot see itself, so the self, understood as a subject of awareness, cannot be aware of itself as an object. Accord- ing to Schopenhauer, for example, the suggestion that a 860 Searle, John R. subject can be an object to itself would be ‘the most mon- strous contradiction ever thought of’. More cautiously, it might be argued that the core of the intuitive notion of self- consciousness is what might be called introspective self- awareness, and that one cannot be introspectively aware of oneself as an object. Sydney Shoemaker’s defence of this view of introspective self-awareness is to point out that in those cases in which one might be said to be conscious of oneself as an object—seeing oneself in a mirror, for example—one always has to identify the presented object as oneself. Since identification always carries with it the possibility of misidentification, first-person statements based on such awareness are not ‘immune to error through mis- identification’ relative to the first-person pronoun. Yet, it seems to be a requirement on introspective self-awareness that it is capable of grounding first-person statements that are immune to this kind of error. To say that a statement of the form ‘I am F’ is immune to error through misidentification relative to the first- person pronoun is to say that the following is not possible: one knows that someone is F, but one’s statement is mis- taken because, and only because, the person one knows to be F is not oneself. For example, if one were to judge ‘I am in pain’ on the basis of feeling pain, it could not happen that the person one knows to be in pain is not oneself. If self-ascriptions of mental states are immune to error through misidentification, then the awareness on which they are based may be introspective, but could not be awareness of oneself as an object. Kant expressed this point by saying that the self as it is in itself cannot be ‘intuited’ or perceived by means of *inner sense. Since, for Kant, knowledge of an object requires both a concept and an intuition of it, he concluded that know- ledge of the self as it is in itself is impossible. Kant did not, however, accept the Humean idea that there is no more to self-consciousness than consciousness of subjectless mental occurrences. Instead, he argued that consciousness of self consists in an ability to ascribe one’s thoughts and experi- ences to oneself. The self-ascription of experiences was in turn claimed to require experience and knowledge of objects other than oneself. A variation on this suggestion is the idea, associated with P. F. Strawson, that for one to be able to ascribe experiences to oneself, one must also be able to ascribe them to subjects other than oneself. A somewhat different approach would be to claim that self-consciousness necessarily involves awareness of one’s own body. Since bodily self-ascriptions such as ‘My legs are crossed’ appear to be immune to error through misidentification when based on awareness of one’s own body ‘from the inside’, this makes it plausible that such awareness is a genuine form of self-consciousness. If bod- ily awareness is also awareness of oneself as an object, then Shoemaker’s argument may not, after all, be decisive. On the other hand, some have argued that the peculiari- ties of bodily awareness are such as to cast doubt on the idea that it is awareness of oneself as an object. The sug- gestion that self-consciousness requires bodily awareness is also controversial. q.c. *introspection. J. L. Bermúdez, The Paradox of Self-Consciousness (Cambridge, Mass., 1998). Q. Cassam (ed.), Self-Knowledge (Oxford, 1994). G. Ryle, The Concept of Mind (London, 1949). S. Shoemaker, Identity, Cause, and Mind, expanded edn. (Oxford, 2003). —— The First-Person Perspective and Other Essays (Cambridge, 1996). P. F. Strawson, Individuals (London, 1959). L. Wittgenstein, The Blue and Brown Books (Oxford, 1958). self-control. Traditionally, a capacity to conduct oneself as one judges best when tempted to do otherwise. Self-control is the contrary of weakness of will or *akrasia. Aristotle distinguishes self-control (enkrateia) from temperance (so¯phrosune¯). The latter, a moral virtue, is possessed only by individuals who have no improper or excessive desires regarding bodily pleasures and pains; self-controlled individuals have such desires, but they characteristically resist them, acting as they judge best. On more recent views, self-control may be exhibited in any sphere in which motivational states compete with a per- son’s values, principles, or practical judgements, including practical and theoretical reasoning and the gathering and assessment of evidence for motivationally attractive or unattractive hypotheses (e.g. the hypothesis that one is popular or that one’s spouse has been unfaithful). a.r.m. A. R. Mele, Autonomous Agents: From Self-Control to Autonomy (Oxford, 1995). self-deception. Everything about the concept of ‘self- deception’ is controversial among philosophers, begin- ning with its definition. That human beings play a large and often wilful role in perpetuating their own ignorance and befuddlement is beyond dispute; but how legitimate is the traditional characterization of the activities sub- sumed under this role as ‘self-deception’? Dictionaries define the term unilluminatingly as the act of deceiving oneself or the state of being deceived by one- self. Since deception involves intentional misleading, such a definition invites the question precisely how one can both intend to be misled by oneself and succeed in such an endeavour. Can the *self perhaps be divided into a deceiv- ing and a deceived part, as in Freud’s view of the uncon- scious keeping information from the conscious self? Or must one adopt Sartre’s paradoxical view, in Being and Nothingness, that ‘I must know, as deceiver, the truth that is masked from me as deceived’? Many reject such views as logically or psychologically impossible. Some claim that ‘self-deception’ refers to one or more of four restrictions on perception, none of which need involve the paradox of simultaneously deceiving and being deceived: first, the ignorance resulting from our necessarily limited capacity to respond to incoming inform- ation; second, the ‘psychic numbing’ that constitutes a reflex response to prolonged exposure to facts which would, if truly confronted each time, be difficult to bear— as when children shield themselves from fully responding self-deception 861 to the violence they witness within the family or on televi- sion; third, mechanisms of denial whereby we may end up deceived about information that would otherwise be too painful to confront, even though we are not consciously deceiving ourselves; and, fourth, processes of more con- scious avoidance such as procrastination, rationalization, and compartmentalization. Advocates of political and religious doctrines have fur- ther disputed the nature of what we hide from ourselves. The greater their zeal in promoting particular truths, the more tempting it becomes for them to assume that non- believers are not merely in error but actually engaged in blocking truths they would otherwise have to acknow- ledge as utterly self-evident. In practice, this assumption easily leads to indoctrination and worse, as witch-hunts ancient and modern make clear. A final controversy about self-deception, however defined, has to do with its desirability. The injunction of the Delphic Oracle—‘Know thyself’—that underlies much philosophy has long been pitted against dismal sus- picions of what we would find if we took the Oracle ser- iously. The drive for attaining greater understanding about ourselves and our role in the world has clashed with the fear of inviting revulsion or misfortune by probing too deeply. Some have further claimed that judicious self- deception is conducive to better mental and physical well- being, as if to underline Jonathan Swift’s (ironical) remark, in A Tale of a Tub, defining happiness as ‘the perpetual Possession of being well Deceived’. The continuing debate over the desirability of self- deception reveals two incompatible views of optimal human functioning. These views, in turn, generate incom- patible conceptions of the role of all involved in therapy: to what extent and by what means should they encourage fuller self-understanding, or on the contrary promote in patients what they take to be life-enhancing false beliefs? If therapists choose the latter path, they run up, once more, against one of the paradoxes of self-deception: for how can they be honest with patients about their intent and about any illusory belief they wish to encourage? But if they can- not, why should their patients trust them? s.b. *lying. Sissela Bok, ‘Secrecy and Self-Deception’, in Secrets: On the Ethics of Concealment and Revelation (New York, 1992). Herbert Fingarette, Self-Deception (London, 1969). Mike Martin (ed.), Self-Deception and Self-Understanding: New Essays in Philosophy and Psychology (Lawrence, Kan., 1985). Alfred R. Mele, Self-Deception Unmasked (Princeton, NJ, 2001). self-defeating theories. In the simplest sense, a theory is self-defeating if the truth of the theory would imply the falsity of the theory. However, the expression is usually applied to theories that purport to guide action in some sense—particularly normative ethical theories. A theory is self-defeating if attempting to achieve what the theory says ought to be achieved is bound to fail because of that attempt. The best-known example is sometimes referred to as the paradox of hedonism. Hedonism tells us that hap- piness is the ultimate goal, but clearly if we spend our lives single-mindedly seeking happiness, we are unlikely to achieve it. *Parfit introduces distinctions between individ- ually self-defeating theories (self-defeating when one per- son acts according to the theory) and collectively self-defeating theories (self-defeating when a group of people act according to the theory), and between indir- ectly self-defeating theories (self-defeating when the aims of the theory are consciously adopted by the agent) and directly self-defeating theories (self-defeating when the aims of the theory are successfully achieved by the agent). Some forms of consequentialism seem to be indirectly self-defeating in the same way that hedonism is indirectly self-defeating. e.j.m. D. Parfit, Reasons and Persons (Oxford, 1984). self-determination, political. The rule of a particular group of people—nation or religious community or, more simply, the residents of a place— over their own affairs. Self-determination is not the same as self-government, which usually implies some version of *democracy. A group of people, freed, say, from imperial rule, might choose the government of a king, an oligarchy, or a clerical élite and, assuming that the choice is not coerced from out- side, this would still be called self-determination. A right to self-determination is a right to make choices of that sort. In recent times, this right is most often claimed on behalf of a nation. (*Nationalism.) But the character and standing of the ‘self ’ in ‘self-determination’ is often a matter of dispute. In principle, the right was invented for the sake of existing collective selves, but it may also happen that collectivities are invented in order to exercise the right. m.walz. *homeland, right to a. Alfred Cobban, The Nation State and National Self-Determination (New York, 1970). Dov Ronen, The Quest for Self-Determination (New Haven, Conn., 1979). self-interest: see egoism, psychological. self-love: see Butler. self-regarding and other-regarding actions. A distinc- tion among actions which becomes important if one attempts to formulate *liberalism by defining an area of conduct in which society has no business to interfere; as does J. S. Mill, when he says that ‘the only part of the con- duct of any one, for which he is amenable to society’ is that which concerns the interests of others. Critics claim the distinction cannot be made. j.m.s. *liberty; state intervention; public and private. J. S. Mill, On Liberty. Sellars, Roy Wood (1880–1973). American critical realist, evolutionary naturalist, materialist, and socialist who taught at the University of Michigan. Knowing, for Sellars, is an activity which, in disclosing objects by means of 862 self-deception ideas, is about external things and consequently tran- scends the cognitive organism. ‘The sensory complex arises in the brain under patterned stimulation of the sense organs and has the role of guidance of response. Such guidance is a transcending role . . . we do not need to get mystical about transcendence.’ Evolutionary *naturalism is not reductive, since nature undergoes cumulative change in which new patterns emerge. ‘Matter is . . . exist- ent in its own right. And I shall think it in terms of the cat- egory of substance’, not process. For Sellars, *‘socialism is a democratic movement whose purpose is in securing of an economic organization of society which will give the maximum possible at any one time of justice and liberty.’ His son, Wilfrid Sellars, acknowledged close philosophical affinity with his father, though he wrote in the idiom of a different generation. p.h.h. *Critical Realism; materialism. R. W. Sellars, Critical Realism (Chicago, 1916). Symposium in Honor of Roy Wood Sellars (Philosophy and Phenom- enological Research, 15; 1954). Sellars, Wilfrid (1912–89). American philosopher notable for his thoroughgoing investigations in metaphysics and the philosophy of mind. He distinguishes between the manifest image of man as a being with beliefs, desires, and intentions, and the scientific image of him as an embodied being subject to study by physicists, biochemists, and physiologists. The task of reconciling those two images is a major problem in the philosophy of mind. Typical of Sellars’s own approach to the problem is his verbal behav- iourist account of thought and meaning in terms of the functional role of linguistic items. (*Functionalism.) Thought is inner speech which is modelled on overt speech, and overt speech is the exercise of a capacity to use words and sentences appropriately in relation to the world and to each other. Thus nothing repugnant to the scien- tific image is invoked. o.r.j. *myth of the given. W. Sellars, Essays in Philosophy and its History (Dordrecht, 1974). —— Empiricism and the Philosophy of Mind, with intro. by R. Rorty and study guide by R. Brandom (Cambridge, Mass., 1997). semantic ascent. The move from talk about the world to talk about the semantic properties of a language (e.g. the move from ‘Snow is white’ to ‘“Snow is white” is true’). This is said to involve ascent because of the doctrine that the semantic properties of a language L cannot, in general, be expressed in L itself, but only in a higher *metalan- guage. The move is useful because ascent to a semantic level enables one to express certain kinds of generaliza- tions that are otherwise inexpressible. Thus, the sentence ‘Every axiom of Peano arithmetic is true’ makes, it is argued, a claim about numbers. But, since Peano arith- metic contains infinitely many axioms, the claim cannot be expressed, without resorting to semantic ascent, by any finite sentence. a.gup. W. V. Quine, Philosophy of Logic (Englewood Cliffs, NJ, 1970). semantics. InFoundations of the Theory of Signs (1938) C. W. Morris divided the general study of signs (*semiotics) into three branches. These are *syntactics, or the study of the relation of signs to other signs; semantics, or the study of the relation of signs to the things they represent; and *pragmatics, or the study of the relation of signs to their users. Semantics is thus the general study of the interpret- ation of *signs, and in particular the interpretation of the sentences and words of languages. Following Carnap, it is commonly divided into pure semantics, or the study of artificial and formally specified languages in the abstract, and applied semantics, or the study of natural, empirically given languages such as English or French. The language studied is called the *object language, and the language in which interpretations are given, the *metalanguage. A semantic statement typically mentions a sentence or other term of the object language, and says what it means, or refers to, or what otherwise provides its interpretation using the metalanguage. An object language can function as its own metalanguage, at least to an extent circum- scribed by the need to avoid semantic *paradoxes. A *for- mal semantics is a fully systematic description of the way in which an object language is to be interpreted, standardly given by a recursive account of the way in which larger meanings or truth-conditions for entire and progressively more complex sentences depend upon the interpretations assigned to their elements. The fundamental problems for semantics are first to discover what linguistic categories we need to distinguish, and then the kind of description of the function of terms that is appropriate. The great advances in the subject came with realizing, for instance, that ‘Some men are mortal’ is semantically quite different from ‘Aristotle is mortal’: the phrase ‘some men’ does not function as a name or ‘term’ interpreted as referring to some men. The difference in function is clearly seen when we look at the different kinds of inference such expressions create. The theory of this dif- ference (quantification theory) is well understood, but other semantic problems have proved less tractable. Are we content to say of a *name, for example, that it refers to its bearer? In that case we see no difference between two names that have the same bearer. Or is some more fine- grained description needed, separating what is said about each of two such names? The former option makes for a more simple and logically more tractable system (exten- sional semantics) while the latter initiates a search for prin- ciples governing the more fine-grained (intensional) features that separate terms with the same extension, but which mean different things. Controversies in semantics frequently centre on the use of various devices, such as possible worlds, to provide the necessary interpretations. But it is generally accepted that the more fine-grained the discriminations or contexts that a language permits, the richer are the categories and descriptions that a semantics must adopt in representing its structure. Even when these problems are solved, others remain for a full philosophical semantics. For any semantics is apt to deal in terms such as *reference, predication, and semantics 863 *truth, and perhaps in addition the richer intensional con- cepts of *meaning, *sense, and *synonymy. And even if we are quite happy using such terms, the question remains in virtue of what they apply (for instance, do predi- cates mean what they do in virtue of shared universals, and what are *universals and how do we apprehend them?). If we consider a pure or formal specification of a language as an abstract structure, then the equivalent problem will be the question what is necessary for it to be correctly attributed to a population. Divisions rapidly arise over whether the appropriate empirical grounding is given by one kind of fact or another. These problems sep- arate semantics in a narrow sense from the wider concerns of the philosophy of language. s.w.b. *semiotics; language, problems of the philosophy of; Montague. R. Carnap, Introduction to Semantics (Cambridge, Mass., 1947). P. Ludlow (ed.), Readings in the Philosophy of Language (Cam- bridge, Mass., 1997). semantics, formal: see formal semantics. semantic theory of truth. This theory was developed by Tarski, who was particularly concerned to overcome the semantic *paradoxes to which talk of *truth gives rise in natural languages, such as the *liar paradox. He held that truth could only be adequately defined for a language which did not contain its own truth-predicate. Calling such a language, L, the object language, Tarski undertook to provide a definition by *recursion of truth-in-L, the def- inition being formulated in an appropriate *metalan- guage. For such a definition to be satisfactory, Tarski held, it would have to enable one to prove all true equivalences of the form ‘S is true-in-L if and only if p’, where ‘S’ is a structural specification of a sentence of L and p constitutes the correct translation of that sentence into the metalan- guage. He showed how this task could indeed be carried out for certain artificial, formalized languages, but believed that the method could not be extended to pro- vide a definition of truth for any natural language, such as English. e.j.l. *snow is white. S. Haack, Philosophy of Logics (Cambridge, 1978). semiotics. General theory of *signs. Peirce distinguished three kinds of sign: icons, which are like the objects signi- fied (e.g. naturalistic paintings); natural signs (e.g. clouds signify rain); and conventional signs (e.g. red for danger, and at least the majority of words). Semiotics is usually divided into three fields: *semantics, the study of mean- ing; *syntactics, the study of (surface ‘grammatical’ and also ‘deep’) structure; and *pragmatics, which deals with the extra-linguistic purposes and effects of com- munications. a.j.l. C. W. Morris, Signification and Significance (Cambridge, Mass., 1968). Sen, Amartya K. (1933– ). Indian economist and philoso- pher at Harvard, winner of the Nobel Prize for Economics in 1998. Working on the foundations of welfare and development economics, Sen is a leading theorist of social choice. In the debate following *Arrow’s paradox, Sen has been a critic of *welfarism, which appraises the value of outcomes wholly in terms of individuals’ preferences between them. Sen has argued for a consequentialist ethics that incorporates respect for rights in its doctrine of the good. He raised the ‘paradox of the Paretian liberal’— an inconsistency, given plausible background assump- tions, between the welfarist claim that if everyone prefers an A to a B, then A must rank above B in a social ordering, and a condition of minimal liberty that each agent possess a personal sphere where her preferences dictate the social ordering. Sen has worked on the nature of personal *well- being and the measurement of poverty. t.p. A. K. Sen, Choice, Welfare and Measurement (Oxford, 1982). —— On Economic Inequality, new edn. (Oxford, 1997). Seneca, Lucius Annaeus (c.2 bc–ad 65). Stoic, tutor to Nero, chief administrator of the Roman Empire with Bur- rus ad 54–62, and author of ten Moral Discourses, 124 Moral Epistles, a satire on Claudius, nine tragedies, and a work on natural philosophy. At worst Seneca is an unoriginal philosopher and a contrived stylist. At best, in the Epistles and Discourses (note particularly ‘De providentia’, a Stoic dissertation on suffering, and ‘Ad Marciam de consola- tione’, addressed to a mother on the death of her sons), he writes with a vividness of illustration and a persuasive bril- liance unrivalled in philosophy: philosophy is practical goodness; excessive passion is evil; external goods are ulti- mately valueless; life is infinitely worth while; tragedy can be overcome or endured. Driven to suicide by Nero, his reputation and his life alike were blighted by his infamous pupil. j.c.a.g. *Stoicism. V. Sørensen, Seneca: The Humanist at the Court of Nero (Edinburgh, 1984). sensation. The subjective aspect of *perception—usually taken to denote the sensory (as opposed to conceptual) phase of a perceptual process. In hearing a concert, for instance, the sensation is the conscious auditory event pre- ceding whatever thoughts and beliefs (if any) the sensa- tion arouses in the perceiver. One might hear—thus have a sensation caused by—a French horn without coming to know or believe that it is a French horn. One might misidentify it as a trombone or not have any thoughts at all about it. This, presumably, is what happens with ani- mals and young children. They can hear French horns. They can, therefore, have sensations—perhaps even sen- sations similar to ours—without these sensations neces- sarily producing beliefs similar to ours. Perhaps (though this is controversial and depends on just what is meant by having a *belief) they can have sensations (be sentient) without having any beliefs at all. 864 semantics Aside from the sensations (visual, auditory, olfactory, and so on) associated with the various sense modalities, there is also a wide variety of other sensory-like phenom- ena that are ordinarily classified as sensations: twinges, tickles, pains, itches, thirst, hunger, feelings of sexual arousal, and so on. If there is any feature that distinguishes this odd assortment of mental entities, it is, perhaps, their introspectively salient quality. The sound of a French horn is utterly unlike the look or feel of a French horn. If the sensation is identified, as it typically is in the case of perceptual awareness, with the way things sound, look, and feel, then these sensations, though they are all of the same thing (a French horn), have an intrinsic, an intro- spectively obvious, quality that distinguishes them from one another. This is quite unlike such *propositional atti- tudes as thought, belief, judgement, and knowledge. Beliefs differ only with respect to their content—the proposition believed—not in their intrinsic quality or ‘feel’ to the person having them. Sensations, on the other hand, can be of, about, or directed upon the same thing (a French horn)—thus having, in this sense, the same con- tent—and yet remain entirely different. As a result, thoughts are classified in terms of their *intentionality, what they are of or about, while sensations are specified in terms of their intrinsic character, what they feel or seem like to the person having them, quite apart from what, if anything, they are of or about. A second feature of sensations that sets them apart from such discursive events as reasoning, thinking, knowing, and remembering is that sensations are, in the first instance at least, independent of the conceptual or intel- lectual assets (if any) of the subject. One cannot want chocolate, believe that there are chocolates in the box, or remember that one ate chocolate without understanding what chocolate is. One can, however, taste chocolate, smell it, and see it—and in this sense have chocolate sensations—while remaining completely ignorant of what chocolate is. In this way sensations constitute a primi- tive level of mental existence. They occur at a level—pre- sumably in certain animals—at which discursive thought and reason are, if possible at all, not well developed. One does not need the concept of an itch or a pain, the capacity to have itch-thoughts and pain-beliefs, in order to feel itches and pains. Though sensations, unlike thoughts, differ from one another in some intrinsic way, their epistemological status remains moot. Is one directly aware of (say) a visual sensa- tion when one perceives, in a perfectly normal way, an external object? If so, is one aware of two things in normal perception—the external object (we say we perceive) and the internal sensation which it (the object) arouses in us? Or is one directly aware of only one thing, the sensation, while the external object is reached (known? perceived?) by some inferential or constructive mental process (thus being known or perceived indirectly) as the *representa- tive theory maintains? Or is one only aware of the external object, the internal sensation being known only by inference, as *naïve realism asserts? If so, how is one’s knowledge—which seems direct—of the character of sensations to be understood? f.d. A. Clark, A Theory of Sentience (Oxford, 2000). F. Dretske, Knowledge and the Flow of Information (Cambridge, Mass., 1981). C. Peacocke, Sense and Content (Oxford, 1983). B. Russell, The Analysis of Mind (London, 1921). sense, manifold of: see manifold of sense. sense and reference. Standard translations of Frege’s terms Sinn and Bedeutung, originating in his 1892 paper ‘Über Sinn und Bedeutung’. The reference of an expres- sion is the entity it stands for: referring expressions stand for objects, predicates stand for *functions (in the math- ematical sense, which Frege called ‘concepts’), and sen- tences stand for truth-values. Referring expressions and predicates combine to form whole sentences, whose references are a function of the references of their parts. Senses are ‘modes of presentation’ of *references: the terms ‘Cicero’ and ‘Tully’ have the same reference but dif- ferent senses. Sense was initially introduced by Frege to solve the puzzle of identity: if ‘Cicero’ has the same refer- ence as ‘Tully’, then how can ‘Cicero is Tully’ be inform- ative when ‘Cicero is Cicero’ is not? The senses of the parts of sentences combine to form the senses of sentences, which Frege called ‘thoughts’. t.c. *connotation and denotation; meaning. Gottlob Frege, ‘On Sense and Meaning’, in Translations from the Philosophical Writings of Gottlob Frege, ed. P. T. Geach and Max Black (Oxford, 1980). sense awareness: see awareness, sense. sense-data. Subjective entities (allegedly) having the properties the perceived external object (if there is one) appears to have. In seeing a white circle under red light and at an oblique angle, the sense-datum would be red and elliptical (the way the white circle looks). According to sense-data theorists, one perceives an external object, a white circle, but what one senses (is acquainted with, directly apprehends) is a red ellipse: the subjective sense- datum. Then, if one is clever (and knows about the funny lighting), one infers, on the basis of the sense-data one directly apprehends, that there is (probably) a white circle causing the red, elliptical sense-datum. In this way our knowledge of sense-data is supposed to provide a founda- tion for all empirical knowledge. f.d. *perception; phenomenalism; representative theory of perception. B. Russell, The Problems of Philosophy (New York, 1959), ch. 1. C. D. Broad, Scientific Thought (London, 1923), chs. 7 and 8. sensibility. In one sense this can mean a set of individual or collective dispositions to emotions, attitudes, and feelings. As such, sensibility is relevant especially to value theory, including ethics, aesthetics, and politics. Arguably, there are at least three important interrelated types of judgement one sensibility 865 can make about a sensibility: that some constitutive emo- tions can be criticized or justified against criticism in various ways (e.g. are ‘irrational’, ‘exaggerated’, ‘well-founded’, etc.); that some constitutive emotions ought to be regu- lated in certain ways, in light of criticism; and that individual or collective responsibility is appropriate for some of the emotions, in light of the possibility of regulation. e.t.s. *aesthetic attitude; taste. Ronald B. De Sousa, The Rationality of Emotion (Cambridge, Mass., 1987). sentences: see statements and sentences. sentential calculus. Where a *proposition is understood to be a completely interpreted indicative sentence of a lan- guage, ‘sentential calculus’ and *‘propositional calculus’ may be used interchangeably. Where, as in Frege, a proposition is an abstract entity which is the sense or content of a sentence, those objects are represented by sentences. Different sentences in a given or a different language may express the same proposition. Given the elusiveness of such entities, the logic of inter- preted sentences remains the vehicle for presenting the logic of propositions as in the propositional calculus. r.b.m. B. Mates, Elementary Logic (Oxford, 1972). sentential function. An expression which can be joined to another expression or expressions to form a sentence. Sen- tential functions include *connectives, such as ‘and’, which form a complex sentence from a sentence or sen- tences. Predicates are also counted as sentential functions since, for example, the predicate ‘ . . . is wise’ when joined with the singular term ‘Socrates’ forms the sentence ‘Socrates is wise’. a.d.o. sentimentalism: see moral sense. sentiments. A sentiment is an attitude, in favour of or against people and their actions, which may involve both *judgement and *emotion. The term ‘sentiment’ has also been used, as by Hume and Smith, to refer to a possible basis for our moral attitudes. In this use sentiment is a feel- ing which the objects of moral appraisal evoke in us; as a possible basis for our moral attitudes, sentiment is opposed to reason. t.p. A. Smith, The Theory of Moral Sentiments (London, 1759). Serbian philosophy. 1863–1945. Emerging after the foundation of Belgrade University in 1863, Serbian philosophy gained a reputa- tion through the work of Branislav Petronijevic´, whose articles were later cited as authoritative in such works as Lee’s Zeno of Elea (Cambridge, 1936) and Boyer’s The Con- cepts of the Calculus (New York, 1939). 1940s–1960s. The philosophical tradition was dismantled in post-war Yugoslavia by the communist regime. The official establishment of ‘humanist Marxism’ followed the 1953 ideological cleansing of ‘dogmatists’. The critical atti- tude of members of the ‘Praxis Group’ such as Svetozar Stojanovic´ to the governing regime led to their dismissal from Belgrade University in 1975. The international reputation of ‘Praxis Marxism’ partly rests on non- philosophical grounds. Non-Marxist approaches were not tolerated and Alexandar Kron’s work in formal logic represents the only such achievement of that time. 1970s and since. Although both politicians and Marxist aca- demics became more tolerant in the early 1970s, it is hard to understand how a critical mass of analytically orien- tated, practically self-taught non-Marxist students was reached so quickly. The ‘September Meetings’ in Dubrovnik, established in the early 1980s by David Charles, Timothy Williamson, and their Belgrade col- leagues, resulted in what was later called the ‘England–Belgrade axis’. The distinctively analytic char- acter of Serbian philosophy is underlined in that fourteen of the sixteen Yugoslav contributors to the collection cited below are, or have been, active at Belgrade University. Hopefully, though now either spread throughout the world or still working at home under unfriendly condi- tions, some will make important contributions to meta- physics, logic, epistemology, philosophical psychology, ethics, philosophy of action, and philosophy of science. Very well trained in philosophical analysis and symbolic logic, they are particularly successful in using thought experiments and the *reductio ad absurdum method. m.a. *Croatian philosophy; Slovene philosophy. A. Pavkovic´ (ed.), Contemporary Yugoslav Philosophy: The Analytic Approach (Dordrecht, 1988). set theory. The property of being human is said to ‘pick out’ or ‘determine’ the set of all human beings. This has subsets—the sets of Scots, English, etc.—and members— e.g. David Hume and Jane Austen. At least normally, if not always, a set is not a member of itself: thus the set of City University philosophers is not itself, alas, another philoso- pher, who could help increasing numbers of students. It is an abstract object. Our basic logical thoughts often embody relations between sets, subsets, and members, for example in syllo- gistic argument. Thus ‘All robots are musical’ says ‘The set of robots is a subset of the set of musical things’; or, every member of the first set is a member of the second. Between 1874 and 1897 Cantor developed an astonishingly rich the- ory of infinite sets, including ones whose members are ordered, and sets having even more members than the so- called ‘denumerably’ infinite set of all integers—thus prov- ing the existence of ‘higher’ infinities. Later, Russell and Whitehead tried to show that pure mathematics is a branch of the logic of sets, and is thus *analytic. Set theory has applications within many areas of mathematics. It is therefore extremely embarrassing that our simplest intuitive thoughts about sets very quickly lead to contra- diction. For if every property determines a set, then the set 866 sensibility (R) of all ‘normal’ sets, namely, ‘those which are not mem- bers of themselves’ is, if a member of itself, then not, and vice versa (*Russell’s paradox). Alternative set theories are formal, symbolic expressions of relationships between sets which attempt to avoid contradictions with minimal loss of intuitive acceptability. The Russellian approach rejects as malformed symbolic expressions of both ‘S is a member of itself ’ and its denial. Sets are put in hierarchies, and one can only meaningfully express membership rela- tions between sets of immediately neighbouring levels. Such an axiomatization may be consistent, but only through inordinate loss of expressive power. Zermelo– Fraenkel–Skolem set theory only allows the construction of sets from properties when certain other conditions obtain: these entail the non-existence of R. Von Neumann– Bernays– Gödel set theory is more comprehensive but even more complex. It allows the existence of R, but it is not a member of any other set (it is then called a ‘class’). This seems counter-intuitive: for if R exists, then why should there not be a merely two-membered set contain- ing, say, R plus the set of all philosophers?. a.j.l. Abraham Fraenkel, ‘Set Theory’, in Paul Edwards (ed.), The Ency- clopedia of Philosophy (New York, 1967). Paul Halmos, Naïve Set Theory (Princeton, NJ, 1960). Michael Potter, Set Theory and its Philosophy (Oxford, 2004). sex. Biological feature distinguishing males and females in respect of their reproductive roles (contrast *gender). Thus, by extension, sex is thought of as a biological drive which gives rise to activity that typically results in repro- duction, or as that activity itself. This suggests that the kind of explanation required for such activity is a bio- logical one, occasioning such protestations as ‘My sex life is not my fault: I’m programmed by my genes’. As well as presupposing a crude determinism, this underplays the role of *culture in giving rise to multifarious forms of sex- ual activity (e.g. *homosexuality). Yet sexual desire has usually been viewed as a blind desire, i.e. one the desir- ability of whose object is not apparent to reason. It is perhaps for this reason that the character of Freudian explanations of behaviour in terms of sexual desire (and their scope) remains mysterious. p.g. *sexual morality. Roger Scruton, Sexual Desire (London, 1986). sex, philosophy of. The theoretical examination of human sexuality, desire, and pleasure. Analysis has focused on the attempt to establish non-moral standards to rank sexual behaviours as ‘natural’ or good, as opposed to perverse or bad. Central to this examination is deter- mining whether sexual desire is localized simultaneously within the phantasmatic as well as the sensory boundaries of the skin. Thomas Nagel reworks Sartre’s notion of ‘a double reciprocal incarnation’ into an account of sexual desire as a serial unfolding of nested desirings of two reflective persons. On this view, the category of perverted sex would extend to include any solitary sexual practice. This analysis of sexual desire in terms of reciprocating psy- chic structurings has ramifications for justifications of sado-masochism. Patrick Hopkins argues that rape simu- lations are logically distinct from rapes, since certain sex- ual behaviours cannot be properly individuated except by their links with a shared sexual imaginary. Robert Solomon not only continues this mentalist reading of sex- ual desire, but refigures bodily gestures as having seman- tic content: an intimate behaviour is a ‘natural expression’ that can be perverse if untruthful or feigned. Alan Gold- man, on the other hand, rejects any account of sexual desire that is not straightforwardly ‘bodily’. For Goldman, sexual desire is directed toward the physicality of another person and involves only a minimal, short-lived psycho- logical component. On this account, the deliberate delay of coitus and bestiality each count as perverse. Other issues involve Kant’s claim that objectification is necessar- ily involved in any sexual relation between persons. Martha Nussbaum separates ‘benign’ objectification from the malignant sorts marked by one or more of the follow- ing: viewing the desired other as essentially replaceable, lacking ‘boundary-integrity’, and failing to be the posses- sor and owner of a unique personal narrative. Michel Foucault challenges this entire analytic schema by running a genealogy of sex. On his view, sex for modern subjects in the West is discursive: subjectivities, bodies, and pleasures are produced by the operations of a particular regime of knowledge/power/pleasure. Fou- cault delineates ‘a scientia sexualis’ that naturalizes the phe- nomena it extracts and organizes through medicalizing the confession, replacing the priest with the psycho- analyst. He claims that any natural/unnatural schematic is a mythic construction, since there is no non-linguistic access to any pre-social human nature. b.t. *sex; sexual morality. Michel Foucault, The History of Sexuality, vol. i: An Introduction, tr. Robert Hurley (New York, 1978). Alan Soble (ed.), The Philosophy of Sex: Contemporary Readings, 3rd edn. (Lanham, Md., 1997). sexism. Thought or practice which may permeate lan- guage and which assumes women’s inferiority to men. The existence of sexism is acknowledged from a variety of ideological perspectives, and sexism may be conceived either as something one encounters instances of, or as a pervasive phenomenon endemic to society. Thus ‘sexist’ is applied pejoratively to individuals and to institutions both by liberal feminists and by feminists who advocate a radical transformation of existing *gender relations. j.horn. *feminism. Mary Vetterling-Braggin, Frederick A. Elliston, and Jane English (eds.), Feminism and Philosophy (Totowa, NJ, 1977). Sextus Empiricus (ad c.200). Sceptic and physician. Sextus, about whose life we know practically nothing, wrote a number of works on the complex history of the Sceptical movement. The surviving works are: Outlines of Pyrrhonism; Sextus Empiricus 867 Against the Dogmatists; and Against the Professors. The second two are usually coupled together and titled Against the Mathematicians, that is, all those who profess any sort of technical knowledge. Outlines of Pyrrhonism provides an account of the philosophy of Pyrrho, including a compari- son of *Pyrrhonism with versions of Academic *scepti- cism. The other works examine at considerable length various dogmatic claims in the arts and sciences and scepti- cal strategies that may be employed to undermine confi- dence in them. These works are therefore a mine of information on many ancient philosophical schools. Sex- tus argues for the superiority of Pyrrho’s Scepticism to that of the *Academy, although the difference between these are disputed in the scholarship. Although Sextus is unre- lentingly critical of all other philosophical positions, he believes that Scepticism has a positive practical purpose, namely, the tranquillity of soul arising from abandoning the quest for knowledge of any sort. l.p.g. Julia Annas and Jonathan Barnes (eds.), The Modes of Scepticism (Cambridge, 1985). Jonathan Barnes, The Toils of Scepticism (Cambridge, 1990). Sextus Empiricus, Outlines of Scepticism, ed. and tr. J. Annas and J. Barnes, 2nd edn. (Cambridge, 2000). Philip P. Hallie (ed.), Sextus Empiricus (Indianapolis, 1985). sexual morality. Principles of right conduct in matters of *sex, or their observance. Two questions arise: What sex- ual acts are morally permissible? With whom are they per- missible? The view that some kinds of sexual act are morally wrong can spring from several sources. The most obvious employs the *consequentialist test of whether they cause harm. Thus some sexually sadistic acts may, by this criter- ion, be condemned, though de Sade himself would reply that they cause less harm than the acts resulting from their repression. In the absence of reliable empirical evidence on the effects of sexual behaviour (e.g. of reading pornog- raphy), non-consequentialist criteria may be turned to. One employs the notion that some kinds of sexual acts are unnatural and therefore wrong. Two problems arise. First, the mere fact that many sexual acts are not con- ducive to reproduction (or to other biological purposes) does nothing to show that they are unnatural, in the sense of tending, or intended, to frustrate what, if anything, is *natural to human beings. Second, even if they were, they would not thereby be shown to be immoral without further premisses such as the Roman Catholic belief that what is contrary to created nature’s purposes is wrong because contrary to its Creator’s will. A more widely acceptable criterion condemns some kinds of sexual acts as failing to treat those with whom they are performed, or oneself, as *persons rather than as objects. Kant seems to have treated all sex like this, hold- ing that ‘sexual love makes of the loved person an object of appetite; as soon as that appetite has been stilled, the per- son is cast aside as one casts away a lemon which has been sucked dry’. Sartre thinks of ordinary sexual desire as aiming to avoid this, but failing, so that either one makes the other an object, as in sadism, or one becomes an object for the other, as in masochism. With Sartre’s pessimism discarded, this has provided an influential criterion, par- ticularly for *feminism. It does, however, require add- itional argument to conclude that depersonalized sex is morally impermissible. The view that a sexual partner should be treated as a person offers one in a series of answers to the question with whom one may, morally, have sexual relations. The most stringent answer restricts sex to *marriage partners, ruling out, inter alia, *homosexuality; the next to those in a relationship of *love, ruling out casual sex; then to those desired and respected as persons, excluding, perhaps, prostitution; and uncontroversially, to consenting adults, ruling out sex with children and animals, who are incapable of informed consent. The first three answers correspond roughly to three general approaches in moral philosophy. The ban on extramarital sex goes with an ethics of *duty. The restriction of sex to love implies an ethics of *care. And the person-centred approach emphasizes an ethics of *virtue, of self-creation rather than spontaneity. It may not be fanciful to suggest that the application of each approach here is a reaction to viewing sex as a potentially disruptive force—disruptive, respectively, to society at large, so that exceptionless formal restrictions need to be imposed; to personal relationships, so that sex must be tied to concern for another’s welfare; and to the individuals themselves, whose integrity as persons is put at risk by it. Few philosophers have, by contrast, developed an ethics of sexuality as something other than an appetite requiring regulation. They have, however, attacked the first two moral restrictions—on consequentialist grounds, like Plato, as socially dysfunctional; on the grounds that they inhibit individual *automony in relationships; and on feminist grounds that they impose a pattern of relation- ships which actually benefits men at the expense of women. As to the third restriction, Nagel bravely main- tains that ‘bad sex is generally better than none at all’. Yet such *essentialist assumptions about sexuality run counter to currently popular views, deriving from femi- nism and *post-structuralism, which see different sexual- ities as constitutive of people’s identities as e.g. a lesbian woman or a straight man. These identities, like those of any cultural minority, are regarded as prior to and formative of the particular moralities which apply to them. Thus the attempt to formulate general principles of sexual conduct is viewed (as in a wider *ethical relativism) as the imposition of the morality of one identity group—typically that of straight men—at the expense of members of others. There are, however, many difficulties with this approach: how fine-grained are the relevant sexual identities? Are all iden- tities (e.g. paedophile) to be tolerated, and, if not, why not? Is the implied *libertarianism compatible with social organization? Can the objections to a general ethical rela- tivism be evaded? The absence of a middle way between universalizing and relativizing approaches to sexual moral- ity is sympathetic to contemporary practical, as well as philosophical, uncertainties in this area of life. p.g. 868 Sextus Empiricus R. Baker and F. Elliston (eds.), Philosophy of Sex (Buffalo, 1984). T. Nagel, ‘Sexual Perversion’, in Mortal Questions (Cambridge, 1979). I. Primoratz, Ethics and Sex (London, 1999). A. Soble (ed.), The Philosophy of Sex: Contemporary Readings, 3rd edn. (Lanham, Md., 1997). J. Weeks, Invented Moralities (Cambridge, 1995). Shaftesbury, third Earl of (1671–1713). Named Anthony Ashley Cooper, like his descendant, the nineteenth- century philanthropist, he is normally known simply as Shaftesbury. Partly educated under the politically radical Locke (though he later criticized Locke on both ethics and epistemology), he was an early, if not always consistent, representative, in his Characteristics of Men, Manners, Opin- ions, Times (1711), of the *‘moral sense’ doctrine in ethics, inventing that phrase. For much of the time, though not all of it, he emphasized feeling rather than reason as the source of morality: we approve of, or take pleasure in the contemplation of, virtue, and this is because we are by nature altruistic and not just selfish. Morality with him becomes human-orientated rather than God-orientated, though religion can motivate us further towards it. He also foreshadowed to some extent *utilitarianism, which came to prominence later in the eighteenth century. a.r.l. S. Grean, Shaftesbury’s Philosophy of Religion and Ethics (Athens, Oh., 1967). shame. An emotion that serves as the focal point of ethics in many ancient and non-Western philosophies, but its comparative neglect in many ethical theories is illustra- tive. The Judaeo-Christian tradition and many modern theories place considerable emphasis on *guilt, but the dif- ference between shame and guilt is profound and symp- tomatic of a larger omission in ethics. Guilt (not causal or legal guilt, but the feeling of guilt) is a highly individualistic emotion, a matter of self-scrutiny and self-condemnation. Shame, by contrast, is a highly social *emotion, and it has to do with violating a common trust, ‘letting the others down’. Like guilt, it is self-accusatory, but it is so through the eyes of others, as an inextricable member of a group or a community. The capacity to feel shame has thus been cited as a pre-condition of all the virtues, as in the Ethiopian proverb ‘Where there is no shame, there is no honour’. Thus Aristotle, in his Ethics, takes shame to be a ‘quasi-virtue’. It is not good to feel shame, because it is not good to have done something about which to be ashamed, but to do something wrong and not feel shame is the ultimate proof of a wicked character. r.c.sol. Aristotle, Nicomachean Ethics (Indianapolis, 1985). Shoemaker, Sydney (1931– ). American philosopher at Cornell, known principally for his work in metaphysics and the philosophy of mind. In the former he has argued for the possibility of time without change. He has defended a causal theory of properties, which has as a consequence that the laws of nature are a posteriori necessary rather than contingent; and a causal theory of identity over time. In the philosophy of mind he is a vocal proponent of ana- lytic *functionalism, in defence of which he offers a subtle discussion of *qualia: he denies the possibility of absent qualia, that someone might be functionally identical to us, yet lack qualitative mental states; but accepts the possibility of inverted qualia, that two people may be be functionally alike but differ in their qualitative mental states. In add- ition, his work has covered personal identity, memory, self- consciousness, and dualism. m.g.f.m. S. Shoemaker, The First-Person Perspective and other Essays (Cam- bridge, 1996). —— Identity, Cause, and Mind: Philosophical Essays, expanded edn. (Oxford, 2003). side constraints: see ends and means. Sidgwick, Henry (1838–1900). British moral philosopher, who developed the most sensitive, sophisticated (and com- plicated) account of *utilitarianism in the nineteenth cen- tury. Sidgwick was educated as a classical scholar at Cambridge, resigned his college position because of reli- gious doubts in 1869, but later became the first secular Pro- fessor of Philosophy at Cambridge (1883). He was the professor when McTaggart, Russell, and Moore were phil- osophy students. Sidgwick wrote on many areas, but his only great work is The Methods of Ethics (1874; and then five other editions in his lifetime). This is not intended as a defence of utilitarianism so much as an account of the ways in which it is possible to reach a rational basis for action. Starting with common sense, Sidgwick identifies three such methods: *intuitionism, universal hedonism (i.e. utili- tarianism), and individual *hedonism (i.e. *egoism). He finds that the particular maxims of common-sense moral- ity do not meet the criteria he lays down for something being an intuitively self-evident principle; but that these are met by certain ‘absolute practical principles’ of a more abstract nature, such as that future good is as important as present good, or ‘that the good of any one individual is of no more importance, from the point of view (if I may say so) of the Universe, than the good of any other’. With such principles he manages to reconcile intuitionism and utili- tarianism. However, he thinks that egoism is also an intui- tive principle of action, which would only be made compatible with utilitarianism by the work of God. Being reluctant to introduce God for this purpose, Sidgwick had no solution for what he called the ‘dualism of *practical reason’, and hence ended the first edition with the sombre words that ‘the prolonged effort of the human intellect to frame a perfect ideal of rational conduct is seen to have been foredoomed to inevitable failure’. r.h. J. B. Schneewind, Sidgwick’s Ethics and Victorian Moral Philosophy (Oxford, 1977). Bart Schultz (ed.), Essays on Henry Sidgwick (Cambridge, 1992). sign and symbol. A distinction first explored in these terms by C. S. Peirce. Signs are a highly general category, including natural indications of things. Spots are a sign of measles, clouds a sign of rain to come. A sign of a state of sign and symbol 869 . Unmasked (Princeton, NJ, 2001). self-defeating theories. In the simplest sense, a theory is self-defeating if the truth of the theory would imply the falsity of the theory. However, the expression. semantics, or the study of the relation of signs to the things they represent; and *pragmatics, or the study of the relation of signs to their users. Semantics is thus the general study of the interpret- ation. usually applied to theories that purport to guide action in some sense—particularly normative ethical theories. A theory is self-defeating if attempting to achieve what the theory says ought to be achieved