Categories
Publié le 25/02/2010
Extrait du document
Categories are hard to describe, and even harder to define. This is in part a consequence of their complicated history, and in part because category theory must grapple with vexed questions concerning the relation between linguistic or conceptual categories on the one hand, and objective reality on the other. In the mid-fourth century BC,Aristotle initiates discussion of categories as a central enterprise of philosophy. In the Categories he presents an 'ontological' scheme which classifies all being into ten ultimate types, but in the Topics introduces the categories as different kinds of predication, that is, of items such as 'goodness' or 'length of a tennis court' or 'red', which can be 'predicated of' subjects. He nowhere attempts either to justify what he includes in his list of categories or to establish its completeness, and relies throughout on the unargued conviction that language faithfully represents the most basic features of reality. In the twentieth century, a test for category membership was recommended by Ryle, that of absurdity: concepts or expressions differ in logical type when their combination produces sentences which are palpable nonsense. Kant, working in the eighteenth century, derives his categories from a consideration of aspects of judgments, hoping in this manner to ensure that his scheme will consist exclusively of a priori concepts which might constitute an objective world. The Sinologist Graham argues that the categories familiar in the West mirror Indo-European linguistic structure, and that an experimental Chinese scheme exhibits suggestively different properties, but his relativism is highly contentious.
«
his list, or why one should feel confident that it is comprehensive ( Aristotle §7 ).
To complicate matters further, there are compelling grounds for the belief that, despite the tradition, the genera of
being with which the Categories deals are not (Aristotelian) categories.
The word which we transliterate
'category' is employed by Aristotle to mean 'predicate' or '(type of) predication' , so that a 'category theory' ought
to be a theory of predication, not of being.
And in a relatively unfamiliar text, Topics (I 9), this is just what we do
find.
After saying that the types of predication must be distinguished, Aristotle provides a list identical to that in
the Categories - except that the first item is not 'substance' , but 'what it is' (103b 21-3 ).
In the Categories ,
individuals such as Socrates are substances pre-eminently, and species such as human being are substances
secondarily: in the Topics , in contrast, colour is the 'what it is' of red no less than human being is the 'what it is'
of Socrates.
What the Topics delivers, then, is a theory of predication according to which the predicates which can characterize
any subject whatsoever will fall into ten ultimate kinds.
Again, defence of this list in particular (or of the
possibility of constructing any non-arbitrary, correct list) is lacking; but reflection on why the Topics was written
might make the absence of this defence considerably less shocking, though at the cost, ironically enough, of
undermining the status which the categories of the Categories have historically enjoyed.
The Topics as a whole
comprises prescriptions for the classification and analysis of a vast array of the sorts of argument then current in
Greek philosophy.
The intention is combative as well as constructive: Aristotle also teaches how to detect and
dismember what he considers sorts of fallacious reasoning, and prime among these is ignorant or malicious
exploitation of confusion in manner of predication.
For example, what is good about food? That it does something
- it produces pleasure.
What is good about human beings? That they are 'of a certain quality, such as temperate or
courageous or just' (Topics I 15.107A ).
So, if viewed as a digest of types of predication already familiar from the
practices of Greek dialectic, the categories of the Topics require no defence beyond reasonable fidelity to the range
of predicates actually found in the typical dialectical repertoire.
In particular, there need be no presumption that
predicational categories rooted in philosophical practice will correspond to a significant, let alone universal,
ontological classification.
This is not to say that predicational and ontological categories are unrelated.
Because Aristotle moves with
disconcerting freedom from language to the world, he takes it for granted that predicate expressions do usually,
without the possibility of radical misrepresentation, refer to real entities.
So a scheme of predicational categories
might indeed suggest at least the outline of a corresponding ontological scheme, one justified in detail and scope
precisely to the extent that the original classification of predicates, and its extension, are well-founded.
If this is the
correct account of how these theories developed in Aristotle, what prompted the substitution of ontological
'substance' for predicational 'what it is'? Perhaps the governing idea of the Categories is that substance alone can.
»
↓↓↓ APERÇU DU DOCUMENT ↓↓↓
Liens utiles
- DOMAINES D'ERREURS CATEGORIES D'ERREURS LEXICAL GRAMMATICAL (morphologie) (lié à la morphosyntaxe) 1) oubli d'un ou plusieurs phonogramme(s) Ex : mot
- THE TRANSCENDENTAL ANALYTIC: THE DEDUCTION OF THE CATEGORIES -KANT