NATURE AND CHARACTERISTICS OF INFERENCE 411 



objected that in every syllogistic reasoning the real ground of our 

 conclusion, the real evidence by which it is supported, is not to 

 be found in the universal premiss of the syllogism, but lies rather 

 in those particular facts of sense experience which are the source 

 from which the universal premiss was derived : so that, after all, it 

 might fairly be contended that the conclusion of the syllogism is 

 established not so much by means 0/&quot;the universal, as in and with 

 the latter, and by means of the self-same facts from which we 

 derive the latter : this being, therefore, rather a record of an in 

 ference completed, than a starting-point for an inference that is 

 still to be accomplished ? 



In answer to this, let \isfirst point out that it does not involve 

 the syllogism in any petitio principii to hold that the particular 

 facts of sense experience are the ultimate source both of its pre 

 misses and of its conclusions, provided we insist that the universal 

 premiss is not a mere enumeration of those facts, and the conclu 

 sion itself necessarily one of them. Then, secondly, the process by 

 which the universal is got from the particulars is not a process of 

 logical inference at all, but a process of intellectual abstraction and 

 generalization which is sometimes a simple intuition, as in the 

 case of self-evident mathematical axioms, while it is sometimes 

 much more complex and laborious, as in the discovery of laws 

 of physical causation by means of induction (197). J Thirdly, it 

 is proper to distinguish between the logical grounds or evidence 

 for the conclusion of a logical inference, and the ultimate psycho 

 logical and metaphysical motives for our accepting and assenting 

 to those logical grounds themselves. 2 And, while insisting that 

 the logical evidence for the conclusion of a syllogism lies partly in 

 its major and partly in its minor premiss, we may freely admit that 

 the mind could never have come into possession of either premiss 

 without facts of sense-experience from which to abstract its con 

 cepts. In this sense, particular facts form the ultimate sources, 

 not, however, the logical grounds or evidence, of the conclusions 

 of our syllogisms : 3 but in this passage from particular fact to 



1 Induction, of course, involves inference in various ways. Although the appre 

 hension of an abstract law of causation, by way of hypothesis, is not an inference, 

 yet the process of verifying it by eliminating all alternative hypotheses, may be 

 set forth in the form of a mixed alternative syllogism. Cf. infra, 209, 212 ; JOSEPH, 

 op. cit., pp. 405 sqq. 



2 C/. VENN, Empirical Logic, pp. 124-9, where he discusses the grounds 

 of our belief in the Uniformity of Nature. 



3 ibid., pp. 377-9, where the author discusses this whole question from the 

 point of view of induction. 



