. Parmenides (cf. ELEATICS) gave an apparent logical demonstration that reality must be one. He seems to have thought that anything real must exist at all times and in all places, and there were many attempts in the next century or so to distinguish between the real, which obeyed one or both of these demands, and the merely apparent or derivative. In the atomism of the late fifth century BC, as in the EPICUREAN system which derived from it, the atoms of which things were made were real, while their shifting colours and temperatures were merely attributed to them by us. Against this background, the notion of substance as one of the CATEGORIES is first explicitly discussed by Aristotle. ‘Substance’ is the traditional translation, followed here, for the Greek ousia. Some scholars, however, think ‘substance’ a bad translation, partly because of its later treatment by Locke (see below). They prefer terms like ‘being’ or ‘entity’ or ‘essence’.
Aristotle seems to use ‘substance’ in two main senses (though some see instead two kinds of sustance here). In the first sense a substance is a particular concrete object, like Socrates or this horse, while in the second sense it is the FORM or essence which makes a substance in the first sense the thing it is. Socrates is what he is because the flesh of which he is made has taken on the form of man and not, say, that of horse. In his Categories Aristotle uses primary substance for the former sense of ‘substance’ and secondary substance for the latter sense (or rather for something approximating to that). Socrates is a primary substance and man is a secondary substance. There is, however, a problem about this second sense of ‘substance’, where a substance is a form, for it is not clear how forms are related to UNIVERSALS. Aristotle denies, in his Metaphysics, that anything universal can be a substance. The modern sense of ‘substance’ as ‘stuff’ (water, iron, etc.), though uncommon in philosophy, seems to be an amalgam of these two senses.
There is, however, a third possible sense of ‘substance’, which Aristotle mentions but dismisses, namely matter, or what remains when one removes the form or properties of something. For Locke, as traditionally interpreted, the substance of something is the substrate which remains when we remove all its properties, a rather mysterious ‘something…we know not what’ which underlies the ‘accidents’ or properties of things. We can only know something by describing it, i.e. giving its properties, which is impossible if it itself has none. But this interpretation of Locke has been challenged, and anyway substances (plural) for him were the things themselves, including their properties.
Descartes defined substances as what can exist without depending on anything else, except God, who alone can create or destroy them and who Himself is the only substance in the strictest sense. For Descartes therefore, as for Parmenides, substance was permanent, so that the material world formed but a single substance, since only it, not things in it, was permanent (once God had created it). Properties, or attributes and modes, as they were called, depended for their existence on the substances in which they inhere. A substance therefore is a substrate for properties, but for Descartes it is not something isolated somehow from all its properties and so unknowable. Modes are, roughly, ways in which an attribute can be possessed, rather like the DETERMINATE of a determinable. Leibniz, who thought of substances as living things, connected substance with the notions of actuality and activity. These thinkers disputed about the number and kinds of substances, the main classification being into material and spiritual substances, though Spinoza claimed that there could be only one substance, generally known as ‘God or nature’.
This notion of a substance as what exists in its own right or independently raises certain difficulties. Firstly, what things are distinguishable and definite enough to count? Cloud? Rainbows? Shadows? Secondly, what sort of independence is in question? Is a hand a substance although it must be a hand of someone? Is a father a substance when described as ‘father’, or only when described as ‘man’, since a father must be the father of children? In other words, do fathers as such form one kind of substance as men or horses do? Or is the independence not logical, as here, but of some other kind—perhaps metaphysical, as Descartes thought in calling God the only substance in the full sense, other things being only derivatively so because they owe their existence to God? In what sense is substance prior, since one can no more have a substance without attributes than attributes without a substance? How indeed is sub-stance related to attribute? Is a substance a mere bundle of attributes? If so, what binds the bundle together? Or does a substance underlie all its attributes, which leads back to the unsatisfactory view of an unknowable substrate? If substrate are what attributes apply to, what can be said about attributes of attributes? Surely we can say of an attribute, as well as of a substance, that it is desirable, or rare, or say that scarlet is brighter than crimson.
These considerations suggest a further ambiguity in the term ‘substance’. Is there perhaps an absolute or metaphysical sense where the contents of the world are divided into substances and other things (attributes, relations, etc.), and a relative or logical sense where whatever we are talking about is the substance and what we say about it are the attributes? (Cf. CATEGORIES.) Some philosophers have rejected the metaphysical sense and supposed that substance is at bottom a linguistic notion (cf. METAPHYSICS. This is one form of nominalism). But if we accept the metaphysical notion of substance, should we count among substances any abstract things like UNIVERSALS or propositions (see SENTENCES)?
Aristotle, Categories, (trans. with commentary by J.L.Ackrill in Clarendon Aristotle series, 1963, translation but not commentary reprinted in J.Barnes (ed.), Oxford Translation of Aristotle, 2nd edn, Princeton UP, 1984), esp. chapter 5 but cf. chapter 7, Metaphysics, books 7 (or Z), 8 (or H), 12 (or Λ).
M.Ayers, ‘Substance: prolegomena to a realist theory of identity’, Journal of Philosophy, 1991. (Defends a notion of substance based on six criteria.)
Descartes, Principles of Philosophy, part I, §§ 51–3, 56. Cf. also § 5 of appendix to Replies to Objections, II. (For substance as permanent see J.Bennett, Kant’s Dialectic, Cambridge UP, 1974, § § 19–21.)
W.C.Kneale, ‘The notion of a substance’, Proceedings of the Aristotelian Society, 1939–40. (General historical discussion, followed by tentative defence of substance.)
J.Locke, Essay, book 2, chapter 23. (For challenge see M.R. Ayers, Locke, Routledge, 1991, vol. 2, part 1, and also J.Bennett, ‘Substance, reality, and primary qualities’, American Philosophical Quarterly, 1965, reprinted in C.B.Martin and D.M.Armstrong (eds), Locke and Berkeley, Macmillan, n.d.)
D.M.MacKinnon, ‘Aristotle’s conception of substance’, in R.Bambrough (ed.), New Essays on Plato and Aristotle, RKP, 1965.
G.Martin, Leibniz: Logic and Metaphysics, Manchester UP, 1960, trans. 1964, § 28. (Leibniz on substance.)
A.Quinton, The Nature of Things, RKP, 1973. (Extended modern treatment of four relevant problems.)
R.D.Sykes, ‘Form in Aristotle: universal or particular?’, Philosophy, 1975. (Can serve as elementary introduction to these topics in Aristotle.)
D.Wiggins, Sameness and Substance, Blackwell, 1980. (Fairly difficult modern treatment of substance.)
R.S.Woolhouse, Descartes, Spinoza, Leibniz: the Concept of Substance in Seventeenth-Century Metaphysics, Routledge, 1993.
This is the complete article, containing 1,226 words
(approx. 4 pages at 300 words per page).
