Substances and aggregates

This chapter will attend first to the concept of substance, which is important in the Dialectic and in the thought of Kant's rationalist predecessors, and will lead on from that to certain other matters which are prominent in the philosophical or historical background of the Dialectic. I can offer only fragments of rationalist thought; but in the hope of making them a little more shapely and less like mere bleeding chunks, I shall sometimes follow a topic through to a natural stopping place, even if that goes a page or two beyond what a consideration of the Dialectic strictly requires.

The basic rationalist notion of substance is clearly stated by Leibniz: ‘When several predicates can be attributed to the same subject, and this subject can no longer be attributed to any other, we call it an individual substance.’ Kant aligns himself with this when he says that the rock-bottom, minimal notion of substance is that of ‘a something which can be thought only as subject, never as a predicate of something else’ (186). I shall often give this a linguistic turn, replacing ‘thought as subject’ by ‘referred to by a substantive or noun-like expression’. Thus transmuted, the rationalist view becomes this: If a complete account of what there is would need some substantival expression referring to the Fs, then the Fs are substances; but not otherwise.

Items which we do in fact handle substantivally, i.e. refer to through nouns or noun-phrases, may fail in either of two ways to qualify as substances.

Firstly, an item may fail to qualify as a substance because it is an aggregate, an assemblage of smaller items which are its parts. A cloud is an aggregate of water-particles, an army is an aggregate of men and weapons etc. That aggregates are not substances follows from the explanation I have given of ‘substance’, at least in the linguistic form of it. We could state all the facts about an aggregate by referring substantivally not to it but only to its parts – capturing the whole truth about a cloud, say, in a set of truths about the histories of individual water-particles, with no mention of the cloud as such. The water-particles may not be substances, but they are closer to being so than the cloud is.