The tedious genius

These three volumes contain the entire surviving correspondence of Alfred Marshall, the dominant figure in both Cambridge and British economics in the decades around the turn of the century. The editor, John Whitaker, carried out the painstaking task of assembling the 1,148 letters over the course, as he notes, of "a long decade".

Unlike his principal protege, John Maynard Keynes, Marshall led a very modest and cloistered existence. He entered St John's College, Cambridge as an undergraduate in 1862 and, apart from a few years around 1880, he spent the rest of his long life in that institution. He wrote occasional letters to the national press, but his role in public affairs was minimal. In the university, his one substantial achievement was the creation of a separate economics tripos. But, apart from the years immediately prior to that, his involvement even in university affairs was small. As Whitaker notes in his introduction "Marshall was not cut out to be a successful university politician".

A consequence of this blameless and retiring life is that a good part of his correspondence deals with everyday trivia. The volumes contain correspondence both to and from leading economists of the day, such as Fisher, Keynes, Pigou, Robertson and Wicksell. But many of these are really no more than polite notes acknowledging the receipt of a book or paper, or some such thing.

The editor seems to have realised the handicap under which he worked when he writes of Marshall: "Being both sensitive to criticism and extremely reluctant to enter into controversy, or even serious debate - after 1890 he seems to have taken little interest in the work of others - his general reaction was to rewrite to enhance the clarity of his statements. Only in a few cases were criticisms addressed explicitly."

The volumes are not likely to appeal to many noneconomists and one cannot help but feel that an edited selection of Marshall's correspondence would have been more attractive. For among the tedium, there are many gems.

One piece of treasure is the protracted correspondence which Marshall conducted via the letters page of The Times with Karl Pearson, one of the founders of modern statistical analysis. Just one of Marshall's replies occupies four full pages of these volumes, or well over 2,000 words. Hardly less delightful is the topic of the controversy, namely whether or not the wages of "drinking men" are less than those of "nondrinkers".

The feeling of a bygone age is no less forcefully contained in a letter to Lord Reay, when Marshall writes "my only confident dogma in economics is that every short statement on a broad issue is inherently false. It was in 1903 that the chancellor of the exchequer set me two questions ... it is now 1909 and the answers are not yet ready". Marshall did, however, send an immediate short reply to the chancellor, saying "I am telling my servant to send you a copy of a paper I wrote in 1890".

But there is clearly a serious point here which is very relevant today. There is no shortage of economists in the City ready to proffer instant advice to the chancellor, and to give confident predictions of the consequences of their tips. Such shallow behaviour simply gives economics a bad name. Marshall would surely have been appalled both at this conduct and at the fact that it is essentially regarded as part of the public relations budget of City firms. Still, he would have been pleased to know that modern developments in nonlinear signal processing theory can be used to confirm the soundness of his judgement.

An especially valuable aspect of the volumes is the occasional apercu on how economics should be conducted. Marshall may have been a dull dog as a person, but he was an outstanding economist. In correspondence with Edgeworth, another famous name in economics, he remarks on the "mischievousness of an academic education in abstract economics not continued into real economics". For all his seeming unworldliness, Marshall was constantly concerned with the need to illustrate principles with examples from real life.

Perhaps the most interesting single letter of the whole volumes is a regrettably incomplete one of 1906 in which Marshall sets out his views on the use of mathematics in economics. His views are worth quoting at length: "I have had a growing feeling that a mathematical theorem dealing with economic hypotheses is very unlikely to be good economics, and I go more and more on the rules - 1. Use mathematics as a shorthand language rather than as an engine of inquiry. 2. Keep to them until you have done. 3. Translate into English. 4. Then illustrate by examples that are important in real life. 5. Burn the mathematics. 6. If you can't succeed in 4, burn 3. This last I do often".

This was by no means a diatribe of the innumerate against the use of maths per se. Marshall himself had extensive training in the subject, and graduated as second wrangler in the maths tripos of his day. His views permeated the Cambridge faculty for many years, being held with probably even greater fervour by Keynes, who himself was also initially a mathematician. These two great economists must be turning in their graves at some of the scribblings in simple calculus which pass for economics in certain parts of Cambridge today.

Marshall's views on this issue have great current relevance for the discipline of economics. The most powerful criticisms of the free market model of economic orthodoxy have come from within the discipline itself, from the imaginative use of rigorous mathematics to explore more deeply the assumptions which are required for the theorems of the free market approach to hold. But all the best papers in this area, such as von Neumann's proof of the existence of competitive equilibrium or Radner's proof of the information requirements of such an equilibrium in a world in which time exists, do satisfy Marshall's criteria. In particular, with a bit of thought they can be translated back into English. The same cannot be said of much of the content of economics journals in recent years. Sifting through the 1,000-plus pages of these three volumes, there are other fascinating remarks, made almost in the nature of asides.

There is little doubt that all serious libraries should have these volumes. But stricter editorial control could have led to a work in which Marshall's comments on economics - which after all are the only real point of interest in the letters - are made much more accessible.

Paul Ormerod is chairman of Post-Orthodox Economics and visiting professor of economics, University of Manchester.