predicable. But logicians give a different meaning to the word. They divide propofitions into certain claffes, according to the relation which the predicate of the propofition bears to the fubject. The first clafs is that wherein the predicate is the genus of the fubject; as when we fay, This is a triangle, Jupiter is a planet. In the fecond clafs, the predicate is a Species of the fubject; as when we fay, This triangle is right-angled. A third class is when the predicate is the specific difference of the fubject; as when we fay, Every triangle has three fides and three angles. A fourth when the predicate is a property of the fubject; as when we fay, The angles of every triangle are equal to two right angles. And a fifth class is when the predicate is fomething accidental to the subject; as when we fay, This triangle is neatly drawn.

Each of these claffes comprehends a great variety of propofitions, having different subjects, and different predicates; but in each clafs the relation between the predicate and the fubject is the fame. Now it is to this relation that logicians have given the name of a predicable. Hence it is, that Sf 2 although

although the number of predicates be infinite, yet the number of predicables can be no greater than that of the different relations which may be in propofitions between the predicate and the fubject. And if all propofitions belong to one or other of the five claffes above mentioned, there can be but five predicables, to wit, genus, Species, differentia, proprium, and accidens. These might, with more propriety perhaps, have been called the five claffes of predicates; but ufe has determined them to be called the five predicables.

It may also be obferved, that as fome objects of thought are individuals, fuch as, Julius Cæfar, the city Rome; fo others are common to many individuals, as good, great, virtuous, vicious. Of this last kind are all the things that are expreffed by adjectives. Things common to many individuals, were by the ancients called univerfals. All predicates are univerfals, for they have the nature of adjectives; and, on the other hand, all univerfals may be predicates. On this account, univerfals may be divided into the fame claffes as predicates; and as the five claffes of predicates

above

above mentioned have been called the five predicables, fo by the fame kind of phrafeology they have been called the five univerfals; altho' they may more properly be called the five claffes of univerfals.

The doctrine of the five univerfals or predicables makes an effential part of every fyftem of logic, and has been handed down without any change to this day. The very name of predicables fhews, that the author of this divifion, whoever he was, intended it as a complete enumeration of all the kinds of things that can be affirmed of any subject; and fo it has always been understood. It is accordingly implied in this divifion, that all that can be affirmed of any thing whatever, is either the genus of the thing, or its fpecies, or its fpecific difference, or fome property or accident belonging to it.

Burgerfdick, a very acute writer in logic, feems to have been aware, that strong objections might be made to the five predicables, confidered as a complete enumeration: but, unwilling to allow any imperfection in this ancient divifion, he endeavours to restrain the meaning of the word predicable, fo as to obviate objec

ན

tions. Those things only, fays he, are to be accounted predicables, which may be affirmed of many individuals, truly, properly, and immediately. The confequence of putting fuch limitations upon the word predicable is, that in many propofitions, perhaps in most, the predicate is not a predicable. But admitting all his limitations, the enumeration will ftill be very incomplete: for of many things we may affirm truly, properly, and immediately, their existence, their end, their cause, their effect, and various relations which they bear to other things. These, and perhaps many more, are predicables in the ftrict sense of the word, no less than the five which have been fo long famous.

Altho' Porphyry and all fubfequent writers, make the predicables to be, in number, five; yet Ariftotle himself, in the beginning of the Topics, reduces them to four; and demonstrates, that there can be no more. We fhall give his demonftration when we come to the Topics; and fhall only here obferve, that as Burgerfdick juftifies the fivefold divifion, by reftraining the meaning of the word predicable; fo Ariftotle juftifies the fourfold divifion,

vifion, by enlarging the meaning of the words property and accident.

After all, I apprehend, that this ancient divifion of predicables with all its imperfections, will bear a comparison with those which have been fubftituted in its ftead by the most celebrated modern philofophers.

Locke, in his Effay on the Human Understanding, having laid it down as a principle, That all our knowledge confists in perceiving certain agreements and difagreements between our ideas, reduces these agreements and disagreements to four heads: to wit, 1. Identity and diverfity; 2. Relation; 3. Coexistence; 4. Real Existence (a). Here are four predicables given as a complete enumeration, and yet not one of the ancient predicables is included in the number.

The author of the Treatife of Human Nature, proceeding upon the fame principle that all our knowledge is only a perception of the relations of our ideas, obferves, "That it may perhaps be esteemed