James Mill on Method
Source: Editor's Introduction to James Mill, Selected Economic Writings, ed. Donald Winch (Edinburgh: Oliver Boyd for the Scottish Economic Society, 1966).
Copyright information: This book is published online with the kind permission of the copyright holders, The Scottish Economic Society.
MILL ON SCOPE AND METHOD
All of Mill's writings as a political scientist, educationist, historian and an economist were based on deeply-held convictions concerning the importance of ‘theory’ or abstract principles; and there is good reason to believe that he had considerable influence on the methodological thinking of two of the major contributors to the classical tradition, namely Ricardo and John Stuart Mill. Since there has been so much criticism of the classical writers for their adherence to a priori methods, and of Ricardo's ‘vice’ in this respect in particular, it may be of interest to consider James Mill's views on methodological questions here.
Mill's reviews of the works of others are replete with criticisms of those whom he regarded as having been insufficiently ‘philosophical’ or ‘speculative’.1 He regarded the inability to generalise or to move beyond immediate experience as an ‘infirmity of the mind’;2 and it was for his own highly-developed powers of ‘ratiocination’ that his works were praised by admirers and denounced by opponents. These powers are obvious in all his writings, not least in his History of British India where he agreed with Gibbon in regarding mere facts as the least interesting part of the historian's material.3 The ‘abstract’ quality of the Elements of Political Economy has already been noted. Indeed it is evident in the earliest of Mill's economic writings, his essay on the corn bounty, which was basically an attack on those who, like James Anderson, rely solely on arguments based on undigested ‘experience’.4 Mill saw no conflict between ‘theory’ or ‘abstract speculation’ and ‘practice’ or ‘experience’. As far as he was concerned, ‘good abstract principles are neither more nor less than the accumulated results of experience, presented in an exceedingly condensed and concentrated state’.5 The only distinction worth making was between ‘comprehensive and profound’ principles and ‘narrow and empirical’ ones.6 His son was taught at an early stage the fallacy of the popular view of this question.
I recollect also his indignation at my using the common expression that something was true in theory but required correction in practice; and how, after making me vainly strive to define the word theory, he explained its meaning, and showed the fallacy of the vulgar form of speech which I had used: leaving me fully persuaded that in being unable to give a correct definition of Theory, and in speaking of it as something which might be at variance with practice, I had shown unparalleled ignorance.7
Halévy has said that ‘Mill during the long walks which he loved to take with Ricardo was chiefly concerned to give him lessons in method’.8 This statement cannot be documented, but from the evidence of the Mill-Ricardo correspondence there is little doubt that Ricardo's eagerness to believe in the applicability of clear-cut principles made him the perfect subject for Mill's teachings. As Mill said proudly of his pupil: ‘as soon as your understanding is convinced, there is perfect certainty.’9 Ricardo and Mill were completely at one on questions of method. Ricardo's Reply to Bosanquet is conducted as an attack on those who claim to be ‘all for fact and nothing for theory’.10 He recognised that one of the crucial points separating him from Malthus was the different emphasis which they placed on theory and practice. ‘If I am too theoretical which I really believe is the case,—you I think are too practical. There are so many combinations,—so many operating causes in Political Economy, that there is great danger in appealing to experience in favour of a particular doctrine, unless we are sure that all the causes of variation are seen and their effects duly estimated.’11 He considered that one of Malthus's great mistakes lay in thinking that political economy was ‘not a strict science like mathematics’.12 Ricardo has been strongly criticised for the very qualities which Mill believed to be essential in a thinker, and which he did his best to encourage in Ricardo. Perhaps the first contemporary to make this criticism was J. L. Mallet, who distrusted Ricardo's ‘entire disregard of experience and practice’, and said of the Principles that it was ‘almost a sealed Book to all but men capable of pursuing abstract reasoning by a strict and mathematical analysis’.13 The only difference between Mill and Ricardo was that whereas Ricardo's confidence in ‘strong cases’ shows itself mainly on economic questions, Mill's confidence extended to every subject upon which he wrote. There seems little reason, therefore, to quarrel with Halévy's conclusion that Mill ‘did not so much give [Ricardo] a doctrine as develop in him the doctrinal leaning and make him a doctrinaire’.14
The extent of the direct influence of Mill on Ricardo must remain a matter for conjecture, but there is little doubt as to his influence on his son John's thinking on methodological questions. The emphasis of John's education was on equipping him with the tools of intellectual analysis, and so successful was this that John considered himself, in the early stages of his life, to be little more than a ‘reasoning machine’.15 As he wrote to John Sterling in 1831: ‘the only thing I believe I am fit for is the investigation of abstract truth, and the more abstract the better. If there is any science which I am capable of promoting, I think it is the science of science itself, the science of investigation-of method.’16
When John was going through his first ‘mental crisis’, with its accompanying estrangement from his father, Macaulay's famous attack on the Essay on Government appeared. Macaulay had set out ‘to expose the vices of a kind of reasoning utterly unfit for moral or political discussions’.17 It was an extremely effective critique of the a priori, deductive approach, which, as John admitted, gave him much to think about. He was not satisfied with his father's off hand response to Macaulay's strictures: he felt that it would have been better to admit that the Essay was a reform tract and not a ‘scientific treatise on government’. This line of argument would not have appealed to James Mill, for it opens up the gap between theory and practice; he would also have been the last to admit the truth of Macaulay's contention that ‘it is utterly impossible to deduce the science of government from the principles of human nature’.18 But John was more flexible, and did make efforts to come to terms with Macaulay's point of view in his essay ‘On the Definition and Method of Political Economy’, which later formed the basis for Book VI of his Logic. Perhaps the most interesting fact about these efforts is that John ended up far closer to his father's position than might have been expected. He admitted that his father's premises concerning the self-seeking propensity of men were too narrow, but held to the view ‘that politics must be a deductive science’; and in so doing rejected Macaulay's view that political science must be inductive or based on ‘experience’.19 He introduced the distinction between the ‘science’ of economics and the ‘art’ of legislation, and mentioned the possibility of using a posteriori methods in testing hypotheses; but he also upheld the a priori, abstract method as the only appropriate one for the moral sciences.20 As Professor Anschutz has indicated, John's refusal to give up his father's basic position, despite his sympathy for certain elements in the opposition's case, is due partly to his view that the deductive approach was sanctioned by the method of the natural sciences, and partly to his political assessment that such methods were more likely to serve the all-important cause of progress and reform.21
James Mill's fondness for the abstract deductive approach is amply illustrated in the other works reprinted in this volume. The following extract from a dialogue written by Mill at the very end of his life helps to make explicit his views on the importance of political economy as a guide to action. It also provides a full statement of Mill's ideas on the rôle of ‘theory’ in economics; he uses the term to denote what would be called a ‘model’ in modern parlance. Several points of interest emerge from the extract. The most striking, perhaps, is Mill's view that the science of political economy was virtually complete, and his consequent impatience with further controversy; the implication being that those who oppose ‘true’ doctrines do so through ignorance or because the truth is incompatible with their selfish interests. He puts forward no ‘external’ criteria for the establishment of truth, in terms, say, of empirical evidence; the truth seems to be simply what right-minded, qualified economists believe it to be.
See e.g. his remarks on Sir James Steuart quoted above, pp. 24–25.
See History of British India, London, 1826 in 6 volumes, vol. I, p. iv, and vol. II, p. 72.
For further comment on this see below, p. 384.
See above, p. 46–49.
Monthly Review, Apr. 1813, vol. LXX, p. 412.
Autobiography, p. 22. A version of the conversation which made such a mark on John's memory can be found in James Mill's article, ‘Theory and Practice’, Westminster Review, Apr. 1836, vol. XXV, pp. 223–34.
Growth of Philosophical Radicalism, p. 272.
Works, vol. VII, p. 363.
Works, vol. III, p. 160, 181–2, 239, 252.
Works, vol. VI, p. 295.
Works, vol. VIII, p. 331.
Political Economy Club; Centenary Volume, 1921, p. ix and p. 224.
Op. cit., p. 282. See also T. W. Hutchinson, ‘James Mill and the Political Education of Ricardo’, Cambridge Journal, Nov. 1953, vol. VII, pp. 81–100.
Autobiography, p. 76.
See The Earlier Letters of J. S. Mill (ed. F. E. Mineka), vol. I, pp. 78–79.
Miscellaneous Writings, 1860, vol. I, p. 317.
Op. cit., p. 317.
Autobiography, pp. 110–13.
See Essays on Some Unsettled Questions in Political Economy, London, 1844, p. 146.
The Philosophy of J. S. Mill, Oxford University Press, 1953, espec. pp. 17–19; 62; 81–96.