The 19th-century search for harmony in economics: Stuart Mill as a key player
The 19th century search for harmony in economics: Stuart Mill as a key player 
After the abstract leaning of Ricardo’s contribution to economics, the principal contributor to mainstream of the discipline is Stuart Mill, more than just the craftsman of the classical synthesis. He did indeed contribute to new theories and to new concepts that serve both Marshall and the marginalist (Ekelund and Hébert 1997). The importance of his contributions, albeit different, can be considered as tantamount to the one of Cournot for marginalists.
Mill is the transmission belt of the positivist view of science (reinforced by empiricism) and the view that the social phenomena are best described by mechanisms inspired by physics. In his Principles, no interest was devoted to technology and history. Indeed, technology was just referred to in passing. To see some
John Stuart Mill is, in a particular way, struggling –or rather undecided–between two conceptions of economic method: the one from Ricardo and part of the work of Classical Economists and another one more historical and case sensitive in leaning, as can be found in A. Smith and later on in a very partial way in the historical schools.
The work of Stuart Mill on method and logic is relatively neglected in the economic literature, even in the history of economic thought, but this is also the case in the literature on logic (Kneale and Kneale 1962). His impact, however, is much more far-reaching in the economic profession than on contemporaneous logicians. His work, A System of Logic, is more than an intellectual curiosity; it contains a methodological discussion of science in general and on the ounce called moral sciences and that roughly correspond in scope to the current social sciences. Mill here is developing a similar task to the one that Comte did in the Positive Philosophy: the methodological setting of science and in each respective case sociology and, we could say, political economy. Mill, while firstly fascinated by Comte, was more cautious on his later development of positivism as we have discussed it. His admiration for Comte’s first masterwork, the Cours de Philosophie positive, did not fade away (Mill 1865). Mill inherited Comte’s task when he tried to reconcile empiricism with a certain view of “abstraction”(Mill 1889). Nevertheless, in his later opus –and certainly central work– The Principles, he has almost completely ignored the logical and methodological considerations.
There is, however, an interesting evolution from Comte to Mill, the latter considered progress not as improvement, but as simply change that does not reverse to its starting point (Mill. A System of Logic, Book VI, Chapter X, p 913).
“The circumstances in which mankind are placed […] form the characters of the human beings; but the human beings, in their turn, mould and shape the circumstances for themselves and for those who come after them.”
However, the point of view of Stuart Mill on history is much less generous than what we can infer from the previous citations. His treatment of history is rather an attempt of normalization of the study of history, very similar to the one we find in the standardization in the history of techniques. His pretense to conciliate classical political economy with the point of view of the historian is a failure. We can see the normalization process and the failure if we compare his book Unsettled Questions with his Principles that does not make much methodological queries, a little bit like he had never written A System of Logic. Stuart Mill negates much history, in a very similar way as Comte did, in his quest for abstraction that he will pass to other economists, like Marshall, even if they have sympathy for the historical schools. History, according to Mill does not introduce essential variations in the economy and society, which is precisely what he must avoid (Denis 2000). Stuart Mill is convinced of the universal validity of the classical political economy (Denis 2000) (Mill 1889, Hodgson 2001). Mill is however more relativist than Comte when, in his book on Comte he says that: “we have no knowledge of anything but phenomena; and our knowledge of phenomena is relative, not absolute. We know not the essence, nor the real mode of production, of any fact, but only its relations to other facts in the way of succession or of similitude.” Mill is clearly positivist here and disdains metaphysics as Comte did. According to him, “all phenomena without exception are governed by invariable law”; it does sound like metaphysics or ontology of the object of study.
To be fair with Stuart Mill, on page 84 of his book about Comte, he says that: “[…] society proceeds in its development […] by the accumulated influence of past generations over the present. The human beings themselves, on the laws of whose nature the facts of history depend, are not abstract or universal, but historical human beings; already shaped, and more what they are, by human society” (Mill 1865).
We have seen the failure of Comte about history. If Mill criticizes the social statics of his master, and he still praises Comte for the treatment of the social dynamics or the laws of the evolution of human society; this does not save Mill’s point from the criticisms addressed to Comte. Mill, on page 113 of the same work, “find no fundamental errors in M. Comte’s general conception of history”. That is the lesson that, unfortunately, has passed to other generations of economists. Time is then enclosed in ceteris paribus clauses. Marshall, author of the next influential textbook on economics, talk about pound, when referring to time, a kind of enclosure for the disturbing effects, as he called them, that we could label emergent effects. In the second essay I will treat in more depth the hesitations of Marshall on the conceptions of time (Mill 1865).
Augustin Cournot is generally considered as a central figure in the building of modern microeconomics and mathematical modelling. What is not generally recognized in economics is his works on statistics and on methodology and philosophy of science. The economic papers of Cournot appear quite early in his academic career under the auspices of the mathematician Poisson. He recognized different influences from Leibniz, Laplace and Condorcet. His rationalist concept of science comes from them all. His optimism and drive to apply mathematics to human societies are congruent with Leibniz philosophy but come above all from the social mathematics (mathématique sociale) of Condorcet and the probability theory of Laplace. From Laplace, to be more precise, he will use probability theory and mechanical analogies and apply them to the social phenomena in the same way as it was applied in natural phenomena (Louçã 1997).
Those are the influences we can find in Cournot by the time he has written most of his works in economics that did not receive the attention he hoped and this turn him somehow a bitter man. Nevertheless, his methodological work did continue and his position has evolved – quite unexpectedly for a positivist assuming the legacy of Condorcet – towards biology and vitalism (Milhaud 1927). In fact it is not unexpected because Cournot, contrary to Comte did not practice his mental health  and continued to get interested in the development of contemporaneous sciences, not just physics, (he translated a treaty on astronomy), mathematics (he wrote several books on mathematics and statistics, collaborated with Poisson), political economy but also the philosophy of science and biology. This last aspect of his scholarship was and still is completely neglected by economists –Walras valued mostly and almost exclusively his mathematics of wealth- but not by mathematicians (Poincaré 190?), philosophers and epistemologically inclined scientists (e. g. Pierre Duhem).
If Cournot began his career by translating books of astronomy and mechanics and adopted the mechanical metaphor from the world of order conceived by Laplace; later in his life, he recognized explicitly the value of life as a metaphor for the social world. Two other aspects have to be stressed: on the one hand Cournot devoted part of his teaching and research to the problem of randomness or chance (hazard) and, on the other hand, on the issue of historical processes; the latter being associated to biology and life rather than physics. Milhaud characterized Cournot Essay of 1851 as still social physics but referred to social biology for his subsequent works (Milhaud 1927). There appears to be a rupture in Comte’s conception and method of scientific enquiry. He distinguished the inorganic world from the organic one (life). For him the difference is such that it justifies a profound change of method. According to him, there is a discontinuous character between the physical world and the chemical phenomena. The complexity of the phenomena increases when we enter the problem of life, and still more complex is the social world, with history being the extreme, out of reach from science as Cournot conceive it. He thought of extending the biological principle (principe vital) to the evolution of societies. The vital force inspired by biology was the principle that he referred to for studying societies, physics for mechanical forces and biology for life (biological forces).
Cournot sustained that we should take reality as it appears to us and not as we would like it to be (Milhaud 1927). He studied history and considered that it contained an irreducible element; similar to the conception he had about randomness.
 I neglect much of the socialist and Marxian literature here.
 It is interesting to note that the moral sciences, moral had a rather broad meaning – if compared to nowadays – coming from the Latin mos, custom; but, although it concerns a great deal of historical work, few serious and theoretical history was conceivable.
 Citations are from pages 6 and 8 of Mill, Auguste Comte and Positivism. Similar quotations can be found in pages 58 and 80.
 His Researches into the Mathematical Principles of Wealth (1838) appeared at a time he was working on probability theory – of which he will publish an interesting treaty Exposition de la théorie des chances et des probabilités published in 1843. Most of the works that will follow develop his methodology and philosophy for social sciences, and of interest for the present work. Besides his Essay on the foundations of knowledge, we can refer to: his Traité de l’enchaînement des idées fondamentales dans les sciences et dans l’histoire of 1861; his restatement of his 1838 book; Principes de la théorie des richesses, 1863; Les institutions d’instruction publiques en France, 1864; his Considérations sur la marche des idées et des événements dans les temps modernes, 2 vols, 1872; and, finally, his Matérialisme, vitalisme, rationalisme: Études des données de las science en philosophie, in 1875 and the last book on economics he published the year of his death, Revue sommaire des doctrines économiques, 1877.
 What Comte called higiène mentale was his refusal to read any scientific literature in order to be able to write his positivist projects for a new science and the renewal of society. Mill although fascinated by Comte’s Cours de philosophie positive did not consider this stand as fruitful for science as well as the vagaries of Comte’s later philosophical writings and activist pamphlets.
 Cournot did consider that randomness does not disappear even if our knowledge of phenomena increases. On the other hand, he recognized different types of probabilities that might be of interest for the current research in imprecise probabilities. However, Laplace’s influence has been strong and his conception of a posteriori probabilities became mainstream.