Evidently the process of argEtts presupposes that things have a fixed nature, and that this fixed nature is apprehensible. Knowl edge rests, therefore, upon the principle of contradiction ("the same essential characteristic cannot both belong and not belong to the same class of things") ; and this principle cannot itself be proved, because it is the condition of all proof. It is, however, sufficiently established by showing that its denial would be tanta mount to denying that the real has any permanent character and to asserting that knowledge of the real is consequently impos sible.
Clearly Aristotle was assuming that the human mind is endowed from the start with capacities which enable it naturally to advance from crude sense intuition to the intuitive grasp of first principles. He was assuming that the innate discriminating power he calls per ception naturally effects a certain abstraction from particularising circumstances (Post. Anal. ii. IQ). Then, on the basis of these first generalities, and through the aid of imagination and memory, we collect by induction instances which agree and reach the stage of empirical knowledge, a stage in which grounds or reasons are, however, not as yet definitely contrasted with their consequences. And when this stage is reached, it becomes possible to take a further step, to pass from the mere contingency of fact to the necessity of reason, to apprehension of the grounds of what is offered in experience.
But more. Let it be assumed that reason, aided by the prepara tory work of induction, has seized the essential mark of some species or type of existence. In the first place, however, it has to be admitted that these essential marks are not all that enters into the concrete nature of this type of existence, that there are in addition accidental features concerning which we are told knowledge is impossible. In the second place, amongst the neces sary characteristics themselves a distinction has to be made between the essential, those without which the type in question would not be what it is, and the derivative, those which can be proved to be necessary because dependent on the essential. Yet the nat_ire of this dependence is left in obscurity. The essential attributes are called grounds of the derivative, but what the rela tion is that objectively corresponds to the relation of ground and consequent Aristotle has nowhere attempted to determine.
A further point calls for notice. Aristotle's theory both of knowledge and of real existence was dominated by the thought of the subject-predicate relation. A thing, an existent, was regarded as that about which predication may be made, the predication finding expression in one or other of the so-called "categories." The fundamental category, that of substance (obo-La), is the con crete individual, not an atom, but the individual as a member of a natural kind, an infima species; while the other nine categories indicate the general differences among all the predicates whereby our knowledge of the individual is expressed. It is only the con crete individual that has the characteristic of never appearing as a predicate. In so defining the ultimate subject of predication Aristotle brings forward in all its difficulty his fundamental doc trine that the individuals are always members of a natural kind. For, defined from this point of view, the ultimate subject of predication would seem to be rather the numerical individual than the natural kind. But the numerical plurality of the members of a natural kind Aristotle persistently regarded as a final acci dent, so to speak, attaching to the real subjects of predication in consequence of their existence in the world of generation. And since knowledge is always knowledge of the universal, we seem to be landed in the predicament of saying that the numerical individual is beyond the range of knowledge.