Categories

4. Are categories universal?

The impression that Aristotelian and Kantian theories have a vital disagreement in common is strengthened if we move beyond the Western tradition in which they are such towering presences. Angus Graham (1989) endorses the contention, most frequently linked with the name of the famous linguist Benveniste, that the Aristotelian categories represent nothing more than yet another bogus reification of Indo-European linguistic forms. For example, ‘quantity’ in Aristotle’s list is the translation of a Greek word which functions as both an interrogative (‘how much?’) and an indefinite adjective (‘so much’). Thus it is argued that peculiarities of Indo-European grammar have encouraged the construction and success of a categorical scheme which, viewed from a broader linguistic perspective, is nothing more than the deluded inflation of quite parochial traits of familiar languages into general features of reality itself.

Graham goes on to try to substantiate this charge of Indo-European parochialism and to show that moderate versions of ‘linguistic relativism’ are acceptable – that is, that the narrowly linguistic properties of a language (its syntax, its degree of inflection and so on) encourage or discourage certain thought-patterns in its users. He sketches a scheme of Chinese categories derived from Chinese interrogative patterns, on the assumption that Aristotle had unreflectingly extrapolated his from Greek ones. For example, if the Aristotelian category of substance is the natural linguistic correlate of the question ‘what is it?’, the analogous Chinese interrogative ho solicits an answer in terms of lei, kind or type of thing (for example, human) rather than essence or basic nature (for example, rational biped). His global hypothesis is that the Chinese preference for ‘which?’ over ‘what?’ questions betrays a holistic cast of mind, felt throughout Eastern philosophy, originating in the inevitable congruence of Chinese thought with Chinese language.

Graham’s version of linguistic relativism is arresting but very hard to vindicate. Quite apart from the obstacle that in the Topics the category ‘what it is’ groups non-substantial subjects as well as substances, Graham’s relativistic interpretation presupposes that Aristotle proceeded in blinkered ignorance of the (alleged) fact that his native language variously stimulated and inhibited his tendency to propound ontological doctrines. But inspection of his works amply demonstrates that Aristotle is acutely aware of the linguistic traps into which philosophers are prone to fall. After all, in large measure the goal of the Topics is to alert us to such dangerous possibilities, and Aristotle’s exercises in linguistic analysis often reach a pitch of extreme sophistication. This is not to say that gauging the impact of language on category theories is fruitless: only that they must not be reduced to a paltry imposition of fortuitous linguistic structure onto a world actually untouched by such strictly local projections. Relativistic objections do not foreclose the debate between Aristotelian confidence and Kantian caution.