Historical criticism, in its appreciation of literary evidence, may be of four kinds. It may (1) examine the record from a dogmatic point of view, pronouncing on it by reference to prepossessions directed against the facts; or (2) make use of the same method, but direct the attack against the evidence on which the record rests; or (3) it may examine whether the record is contemporary with the events narrated; or (4) consider its internal agreement with itself or with fact. We have instances of each of these methods in the examination of the literary evidence on which miracles are believed. The first, the prepossession concerning the philosophical impossibility of miracles, is seen in Spinoza; the second, the impossibility of using testimony as a proof of them, in Hume; the third, the question whether they were attested by eyewitnesses, is the ground which Bolingbroke touches; the fourth, the cross-examination of the witnesses, is seen in Woolston. Of these, the first most nearly resembles the great mass of the deist objections to revelation, being philosophical rather than critical. The second forms a transition to the two latter, being philosophy applied to criticism, and is the form which deism now took. The two latter are those which it subsequently assumed.(478) These remarks will explain Hume’s position,(479) and show how he forms the transition between two modes of inquiry; his point of view being critical, the cause of it philosophical. His speculations in reference to religion are chiefly contained in his Essays on the Human Understanding. A brief explanation is necessary to show the dependence of his theology on his philosophy. The speculations of Locke, as we have before had occasion to notice, gave an impulse to psychological investigations. He clearly saw that knowledge is limited by the faculties which are its source, which he considered to be reducible to sensation and reflection; but while denying the existence of innate ideas, he admitted the existence of innate faculties. Hartley carried the analysis still farther, by introducing the potent instrument offered by the doctrine of the association of ideas. Hume, adopting this principle, applied it, in a manner very like the independent contemporaneous speculations of Condillac in France, to analyse the faculties themselves into sensations, and to furnish a more complete account of the nature of some of our most general ideas, such, for example, as the notion of cause. The intellectual element implied in Locke’s account of the process of reflection here drops out. Faculties are regarded as transformed sensations; the nature of knowledge as coextensive with sensation. According to such a theory therefore, the idea of physical cause can mean nothing more than the invariable connexion of antecedent and consequent. The notion of force or power which we attach to causation becomes an unreality; being an idea not given in sensation, which can merely detect sequence.