The Cambridge History of English and American Literature in 18 Volumes (1907–21).
Volume XIV. The Victorian Age, Part Two.
§ 8. System of Logic
Mill’s System of Logic was published in 1843, and ran through many editions, some of which—especially the third (1850) and the eighth (1872)—were thoroughly revised and supplemented by the incorporation of new, mainly controversial, matter. It is probably the greatest of his books. In spite of Hobbes’s treatise, and of the suggestive discussions in the third book of Locke’s Essay, the greater English philosophers almost seem to have conspired to neglect the theory of logic. It had kept its place as an academic study, but on traditional lines; Aristotle was supposed to have said the last word on it, and that last word to be enshrined in scholastic manuals. English thought, however, was beginning to emerge from this stage. Richard Whately had written a text-book, Elements of Logic (1826), which, by its practical method and modern illustrations, gave a considerable impetus to the study, and Hamilton’s more comprehensive researches had begun. From them Mill did not learn much or anything. What he set himself to work out was a theory of evidence in harmony with the first principles of the empirical philosophy; and this was an almost untouched problem. He may have obtained help from Locke; he acknowledges the value for his thinking of Dugald Stewart’s analysis of the process of reasoning; he was still more indebted to his discussions with a society of friends. Thus he worked out his theory of terms, propositions and the syllogism; and then the book was laid aside for five years. When he returned to it, and proceeded to analyse the inductive process, he found rich material to hand not only in Sir John Herschel’s Discourse on the Study of Natural Philosophy (1830), but, also, in William Whewell’s History of the Inductive Sciences (1837). After his theory of induction was substantially complete, he became acquainted with, and derived stimulus and assistance from, the first two volumes of Comte’s Cours de philosophie positive (1830). These were the chief influences upon his work, and their enumeration serves to bring out the originality of his performance. His work marks an epoch in logical enquiry, not for English philosophy only but in modern thought.
The reputation of Mill’s Logic was largely due to his analysis of inductive proof. He provided the empirical sciences with a set of formulae and criteria which might serve the same purpose for them as the time-worn formulae of the syllogism had served for arguments that proceeded from general principles. In this part of his work he derived important material from Whewell, much as he differed from him in general point of view, and he found his own methods implicitly recognised in Herschel’s Discourse. The importance and originality of Mill’s contribution, however, cannot be denied. His analysis is much more precise and complete than any that had been carried out by his immediate predecessors. He seeks to trace the steps by which we pass from statements about particular facts to general truths, and also to justify the transition: though he is more convincing in his psychological account of the process than in his logical justification of its validity. When he is brought face to face with the fundamental problem of knowledge, as Hume had been before him, he does not show Hume’s clearness of thought.
Mill’s work is not merely a logic in the limited sense of that term which had become customary in England. It is a theory of knowledge such as Locke and Hume attempted. The whole is rendered more precise by its definite reference to the question of proof or evidence; but the problem is Hume’s problem over again. The ultimate elements of knowledge are subjective entities—“feelings or states of consciousness”—but knowledge has objective validity. The elements are distinct, though the laws of association bind them into groups and may even fuse them into inseparable wholes—but knowledge unites and distinguishes in an order which is not that of laws of association. The theory of knowledge, accordingly, has to explain how our thinking, especially in the transition from assertion to assertion which we call “proof,” has validity for objective reality, and, in doing so, it has to give a tenable account of the universal principles postulated in these transitions. In Mill’s case, as in Hume’s, this has to be done on the assumption that the immediate object in experience is something itself mental, and that there are no à priori principles determining the connections of objects. In his doctrine of terms and propositions, Mill emphasises the objective reference in knowledge, although he cannot be said to meet, or even fully to recognise, the difficulty of reconciling this view with his psychological analysis. He faces much more directly the problem of the universal element in knowledge. He contends that, ultimately, proof is always from particulars to particulars. The general proposition which stands as major premiss in a syllogism is only a shorthand record of a number of particular observations, which facilitates and tests the transition to the conclusion. All the general principles involved in thinking, even the mathematical axioms, are interpreted as arrived at in this way from experience: so that the assertion of their universal validity stands in need of justification.
In induction the essential inference is to new particulars, not to the general statement or law. And here he faces the crucial point for his theory. Induction, as he expounds it, is based upon the causal principle. Mill followed Hume in his analysis of cause. Now the sting of Hume’s doctrine lay in its subjectivity—the reduction of the causal relation to a mental habit. Mill did not succeed in extracting the sting; he could only ignore it. Throughout, the relation of cause and effect is treated by him as something objective: not, indeed, as implying anything in the nature of power, but as signifying a certain constancy (which he, unwarrantably, describes as invariable) in the succession of phenomena. He never hesitates to speak of it as an objective characteristic of events, but without ever enquiring into its objective grounds. According to Mill, it is only when we are able to discover a causal connection among phenomena that strict inductive inference is possible either to a general law or to new empirical particulars. But the law of universal causation, on his view, is itself an inference from a number of particular cases. Thus it is established by inductive inference and yet, at the same time, all inductive inference depends upon it. Mill seeks to resolve the contradiction by maintaining that this general truth, that is to say, the law of causation, is indeed itself arrived at by induction, but by a weaker form of induction, called per enumerationem simplicem, in which the causal law is not itself assumed. Such a bare catalogue of facts, not penetrating to the principle of their connection, would not, in ordinary cases, justify an inference that can be relied on. But Mill thinks that the variety of experience that supports it in this case, its constant verification by new experience and the probability that, had there been any exception to it, that exception would have come to light, justify our confidence in it as the ground of all the laws of nature. He does not recognise that these grounds for belief—whatever their value may be—all assume the postulate of uniformity which he is endeavouring to justify.
A later and more comprehensive discussion of his philosophical views, especially in a psychological regard, is given in his Examination of Sir William Hamilton’s Philosophy and of the principal philosophical questions discussed in his writings. This work was published in 1865; and, as his habit was, the author amplified it greatly in subsequent editions by replies to his critics. In this case the criticisms were exceptionally numerous. The book focused the whole controversial energy of the period belonging to the two opposed schools, the intuitional and the empirical; and, in spite of its controversial character, it became the leading text-book of that psychological philosophy which had been adumbrated by Hume. It is a work which shows Mill’s powers at their most mature stage. He criticises with severity the theory which he sets out to examine; but he is alive to the awkward places in his own position. Among the numerous doctrines on which he left the impress of his workmanship, none excited more attention at the time of the book’s publication, or are of greater permanent importance, than his doctrines of the external world and of the self. There is nothing fundamentally original about his views on these topics; but his discussion of both illustrates his ability to see further into the facts than his predecessors, and his candour in recording what he sees, along, however, with a certain disinclination to pursue an enquiry which might land him definitely on the other side of the traditional lines. Mill’s doctrine is essentially Humean, though, as regards the external world, he prefers to call it Berkeleyan; and here he is the inventor of a phrase: matter is “permanent possibility of sensation.” The phrase is striking and useful; but a possibility of sensation is not sensation, and the permanence which he attributes to the possibility of sensation implies an objective order: so that the reduction of matter to sensation is implicitly relinquished when it appears to be affirmed in words. Mind, in somewhat similar fashion, is reduced to a succession of feelings or states of consciousness. But the fact of memory proves a stumbling-block in his way; he cannot explain how a succession of feelings should be conscious of itself as a succession; and he implicitly admits the need of a principle of unity. Thus, he almost relinquishes his own theory and only avoids doing so explicitly by falling back on the assertion that here we are in presence of the final inexplicability in which ultimate questions always merge.