. There are two main applications of the term ‘naturalism’ in modern philosophy, one in metaphysics, epistemology and philosophy of mind and the other in ethics. What they have in common is the belief that the universe is all one, in the sense that all objects in it and all aspects of it are equally accessible to study by scientific method.
The empiricist tradition of Locke, Hume and J.S.Mill and others made no sharp distinction between what would now be called philosophy of mind and psychology. Questions about the justification of mental states such as beliefs tended to be assimilated to questions about their psychological origins. Towards the end of the nineteenth century and beginning of the twentieth, this tendency was fiercely attacked, in particular by Bradley, Frege and (eventually but not at first) by Husserl, usually under the name of PSYCHOLOGISM. The empiricist revival that dominated the first half of the twentieth century accepted and indeed insisted on this distinction between epistemology and psychology, and the difficulty of finding epistemological justifications for various beliefs once psychological substitutes were discounted led to scepticism.
More recently, mainly under the influence of Quine, there has been a revival of naturalism but in a more self-conscious form, called naturalizedEPISTEMOLOGY, in that epistemological questions are not just tacitly treated in a psychological manner but the change is avowed and deliberate. Among other writers this is also true of Hume, but Hume abandoned the search for justification, and tried to explain rather than justify our beliefs in such things as real causal connections and the existence of a world outside our own minds. (How far he either accepted those beliefs in fact or thought they could be disproved is disputed, but he claimed that scepticism, however irrefutable in theory, was impossible to accept in practice outside the philosophical study.) The recent revival is more inclined to claim that the psychological account itself constitutes a kind of justification. Cf. here reliabilism (see EPISTEMOLOGY).
The ethical version of naturalism claims that there is no unbridgeable gulf between ethics and other studies. It takes two main forms, that ethical terms can be analysed into non-ethical terms, and that ethical conclusions can be logically derived from non-ethical premises.
Modern discussions start from two famous attacks on these two forms respectively, by Moore and Hume. Moore insisted that ‘good’ is indefinable, and that the questions what ‘good’ means and what things are good must be sharply distinguished, e.g. pleasure may be good, and even possibly the only good thing, but ‘good’ does not mean ‘pleasant’ or ‘producing pleasure’, etc. He called the neglect of this distinction the naturalistic fallacy, and said goodness is a nonnatural quality. (He never successfully explained ‘nonnatural’, which is now not used of qualities. For ‘nonnatural’ see MEANING.) One of Moore’s arguments was the open question argument: whatever definition of ‘good’ was offered, it would always be an open question whether what satisfies the definition is good. It is not clear, however, whether the naturalistic fallacy lies in defining ‘good’, defining ‘good’ in non-ethical terms, defining any ethical notion in non-ethical terms, defining any notion in terms of notions in a different sphere, or indeed simply confusing one notion with another (cf. Prior). In this last form the fallacy is sometimes called the definist fallacy. It is difficult, however, to mark off different spheres or notions, and the errors, if any, are not strictly fallacies—no fallacious inference is involved.
Hume, as usually interpreted, attacked naturalism by denying that conclusions whose main verb is an ‘ought’, or an equivalent, can be logically derived from premises not containing such a notion; he maintains what is now called the is/ought distinction. Hume was primarily concerned with the moral OUGHT, but the question also arises of whether his view applies to other ‘oughts’ (e.g. ‘doctor’s orders’).
The alleged distinction underlying both these attacks is often if loosely, called the fact/value distinction, though ‘ought’ is not strictly a value term (and the fact/value and is/ought distinctions are sometimes separated: Wiggins, who accepts the former but rejects the latter).
Moore made his distinction between different qualities, all of which objectively belonged to their possessors: whether a thing is good no more depends on the observer than whether it is, say, spherical. Various later writers, however, have tried to make the same distinction in other ways, asking how words have meaning, and what the speaker’s own attitudes contribute. Thus a distinction grew up between straightforward descriptions which claim to state facts and impart knowledge, and utterances whose purpose is to express or evince emotions or attitudes, to issue prescriptions or recommendations, or to evaluate. Utterances of the former kind, and the words used in them, had descriptive, factual or cognitive meaning; those of the latter kind had emotive, evaluative, prescriptive or in general non-cognitive meaning, and were subject to a SPEECH ACT analysis. The terms in the former list in general coincide (but for ambiguities in ‘factual’ see FACTS). Those in the latter list do not all coincide—to express emotion is not the same as to prescribe action or attitudes—but they share the property of being contrasted with the former list. Many utterances, however, especially ethical ones, had both kinds of meaning. ‘He is courageous’ might mean ‘He knowingly takes great risks’ (factual), ‘and I hereby express approval of his doing so’ (emotive), or ‘I hereby recommend you to follow suit’ (prescriptive). ‘He is rash’ might have the same factual meaning but the opposite non-cognitive meaning, i.e., as above, but with ‘disapproval’ for ‘approval‘ and ‘forbid’ for ‘recommend’. Emotivists and prescriptivists analyse ethical utterances in this way, emphasizing emotive and prescriptive meaning respectively, but descriptivists think that ethical utterances have meaning in the same way as factual ones do, i.e. they state facts, even if of an ethical kind. Descriptivism and naturalism are in practice closely similar, but Moore was a descriptivist but not a naturalist.
Understandably, the naturalist controversy is associated with another, concerning whether there are objective and agreed procedures for arriving at conclusions in matters of value or duty. Descriptivists and most naturalists claim that there are, even if they are hard to elaborate. They take the utterances in question to say something true or false about objective reality, and so to state facts. Their opponents may deny that there are such procedures. Hence the attempt common among POSITIVISTS and linguistic PHILOSOPHERS to confine ethics to certain questions only: see ETHICS. Alternatively they allow that there are such procedures, but have special difficulty in elaborating them, since special kinds of reasoning seem needed to support conclusions claiming to do something other than state facts.
Recently the fact/value distinction has come under fire. There seems to be little unity on the value side, which has to cover expressions of emotion, commendations, prescriptions, etc., in each case sometimes moral and sometimes non-moral. Where a similar distinction seems viable it may still be relative (Anscombe). Emotivists and prescriptivists face objections connected with UNIVERSALIZABILITY. The theory that words like ‘good’ or ‘ought’ get their meaning by commending or prescribing, i.e. by their use in SPEECH ACTS, also faces difficulties (see GOOD).
The view that moral conclusions cannot be logically derived from non-moral premises may be called a weak non-naturalism or weak anti-naturalism if it admits that there are limits to the possible contents of ethical statements, i.e. to what we can intelligibly be said to commend or prescribe as a duty. Would we, for example, understand anyone who seriously thought it his duty to blink every five seconds, regardless of circumstances? It is a strong non-naturalism if it denies that there are such limits. (Cf. the question whether just anything can be GOOD). The phrase autonomy of morals or of ethics can apply to both these versions of non-naturalism, and also to the view that practical moral questions can be separated from theoretical ethical analysis because any position on the former can be combined with any position on the latter. This last view too has been attacked because of universalizability, among other things.
Perhaps the most central question is whether there is any way of establishing ethical or other valuational conclusions. If there is, there may seem to be little objection to using terms like ‘true’ and ‘fact’ in connexion with such conclusions, i.e. to adopting descriptivism; but this view has recently been disputed.
‘Naturalism’ is also used in other ways, notably in art, where it might be most obviously described as the attempt to represent nature (including human nature) as it is rather than in various stylized, symbolic, or romantic guises; cf. REALISM.
G.E.M.Anscombe, ‘On brute facts’, Analysis, vol. 18, 1958. (Argues that distinction between ‘brute’ and ‘institutional’ facts is relative.)
R.F.Atkinson, ‘The autonomy of morals’, Analysis, vol. 18, 1958. (Defends fact/value distinction against some attacks of a logical nature.)
C.Beck, ‘Utterances which incorporate a value statement’, American Philosophical Quarterly, 1967. R.W.Newell, ‘Ethics and description’, Philosophy, 1968. (Two attacks on fact/value distinction.)
*P.Foot (ed.), Theories of Ethics, Oxford UP, 1967. (Includes several notable discussions of naturalism.)
D.P.Gauthier, ‘Moore’s naturalistic fallacy’, American Philosophical Quarterly, 1967. (Discusses Moore and Prior and offers own view on naturalistic fallacy.)
A.Gewirth, ‘Positive “ethics” and normative “science”’, Philosophical Review, 1960. (Compares ethics and science as fields of enquiry.)
J.Griffin, ‘Values: reduction, supervenience, and explanation by ascent’ in D.Charles and K.Lennon (eds), Reduction, Explanations, and Realism, Clarendon, 1992. (Throws doubt on the natural/nonnatural distinction, and thereby on supervenience accounts of value: what exactly supervenes on what?)
*W.D.Hudson (ed.), The Is/Ought Question, Macmillan, 1969. (Discussions of Hume’s attack on naturalism, plus some essays on evaluation.)
W.D.Hudson, Modern Moral Philosophy, Macmillan, 1970. (General introduction. Includes Hare’s later views in discussing prescriptivism, etc. Cf. also p. 171 on terminology.)
F.Jackson, G.Oddy and M.Smith, ‘Minimalism and truth aptness’, Mind, 1994. (Argues against a too ready acceptance of terms like ‘true’ and ‘fact’ in connection with ethical statements.)
P.Kitcher, ‘The naturalists return’, Philosophical Review, 1992. (Extended survey of the revival of naturalism in metaphysics and epistemology.)
Midwest Studies in Philosophy, vol. 19, 1995. (Journal issue devoted to philosophical naturalism.)
G.E.Moore, Principia Ethica, Cambridge UP, 1903, chapter 1, esp. § B. (Locus classicus for naturalistic fallacy.)
L.Nochlin, Realism, Penguin, 1971. (Naturalism in art.)
D.Papineau, Philosophical Naturalism, Blackwell, 1993. (General discussion, relating naturalism to physicalism and applying it to problems in philosophy of mind and epistemology.)
A.N.Prior, Logic and the Basis of Ethics, Oxford UP, 1949. (Naturalistic fallacy in its historical setting.)
W.V.O.Quine, ‘Epistemology naturalised’ in his Ontological Relativity and Other Essays, Columbia, 1969. (Major influence in revival of naturalism.)
A.K.Sen, ‘Hume’s law and Hare’s rule’, Philosophy, 1966. (Uses universalizability to defend naturalism.)
S.J.Wagner and R.Warner (eds), Naturalism: A Critical Appraisal, Notre Dame UP, 1993. (Specially written essays mainly critical of naturalism from wide variety of viewpoints, with introduction and summaries.)
D.Wiggins, ‘Truth, invention, and the meaning of life’ (pamphlet), Proceedings of the British Academy, 1976, reprinted in his Needs, Values, Truth, Blackwell, 1987. (See esp. p. 96 (of reprint) for separation of fact/value and is/ought distinctions. Difficult.)
This is the complete article, containing 1,841 words
(approx. 6 pages at 300 words per page).
