Hume's philosophy is best understood when we consider it as, in the first place, a criticism of Berkeley, just as Berkeley's had been a criticism of Locke. It will be remembered that the founder of subjective idealism discarded the notion of material substance as an "abstract idea," an unintelligible figment devoid of any sensuous or imaginative content. The only true substances are the subjects of what we call experience communicating through sensation with God, the infinite spirit whose eternal consciousness is reality itself. Hume applied the same tests to spiritual substance, and found that it equally disappeared under his introspective analysis. He begins by dividing the contents of consciousness into two classes, impressions and ideas--the second being copies of the first, and distinguished from them by their relative faintness. Now, from these perceptions (which he called thoughts) Descartes had passed by an immediate inference to the ego or self, which he affirms as the primary fact of consciousness, using it as a basis for sundry other conclusions. But Hume stops him at once, and will not grant the existence of the metaphysical self--that is, a simple and continued substance, as distinguished from particular states of consciousness. We are, he declares, "nothing but a bundle of different perceptions, which {80} succeed each other with an inconceivable rapidity, and are in a perpetual flux and movement." "There is properly no _simplicity_ in it [the self] at one time, nor _identity_ in different [times]; whatever natural propensity we may have to imagine that simplicity and identity." So much being assumed, Berkeley's whole argument for a new theology founded on subjective idealism is bound to collapse, as also is the argument for natural immortality derived from the supposed simplicity and identity of the thinking substance. Modern critics have rightly insisted, as against Hume, that isolated perceptions without a self are abstractions not less unintelligible than a self without perceptions. But the metaphysical argument for human immortality has not benefited by this more concrete interpretation of epistemology; and probably Hume was really more interested in destroying this than in maintaining the sceptical paradox which does not recur in his later writings.