The Cambridge Resolution of the Methodenstreit
Marshall was not inclined to engage in methodological debate, and yet his methodological approach arguably had a great impact. This impact is divided between those (non-mainstream) economists who focus on his professed methodology and his analysis of evolutionary contexts (e.g. as in industrial districts), on the one hand, and those (mainstream) economists who have pursued the deductivist methodology he used in his formal theoretical marginal analysis, on the other. Groenewegen (1995: 415; italics in original) quotes Marshall, in a letter to Neville Keynes, as follows: ‘I take an extreme position as to the method & scope of economics. In my new book I say of methods simply that economics has to use every method known to science’. This pluralist methodology included a particular view of mathematics which followed in the Newtonian-Scottish-Malthusian methodological tradition. First, the emphasis was on geometry rather than algebra or calculus. Second, mathematics was not to be used as an engine of enquiry, but rather as a substitute for verbal argument to be set aside when communicating theories with the aid of real-life examples. This reflected both that, like Hume, Marshall sought to make connections across the learned-vulgar divide and his view that political economy had a human and concrete nature.
Marshall’s evolutionary view of the subject matter of economics limited the scope for general laws arrived at either by pure deduction or pure induction. Marshall explicitly argued against ‘economic man’ as an appropriate basis on which to build theory, on the grounds of its contradiction by observation. Deductivists in contrast proceeded by accepting the concept of economic man as ‘self-evident’, and as fruitful given the scope for application of mar- ginalist analysis on the basis of this conception. Marshall’s evolutionary view required him to engage in detailed study of real social contexts in order to grasp the nature of particular evolutions. Yet, for all his use of historical evidence, Marshall was critical of the English Historical School’s empiricism
(pure inductivism) on the grounds that the evolutionary nature of the subject matter precluded identification of permanent uniformities (Groenewegen 1995: 310). It was rather Hegelian analytical history[1] which he saw as playing an important methodological role, whereby the operation of the mind was brought to the identification of conjectural patterns in history. From Hegel too Marshall absorbed the organicist idea of the whole being more than the sum of the parts.
The presentation of the Principles thus involved a methodology which differed in important respects from the other builders of the marginalist revolution. Yet it was those elements of the content of the Principles which were most consistent with the development of marginalism elsewhere, rather than Marshall’s professed methodology, which had the greatest impact on economics as a whole. Indeed, Whitaker (2008: 364) argues that ‘his [Marshall’s] method was in the general deductive tradition of John Stuart Mill’. His key theoretical contribution as far as most economists are concerned was the conceptual framework he developed for marginalist comparative static analysis of exchange, which spawned theoretical developments, generating law-like conclusions about the consequences of atomistic individual behaviour. For Marshall, perfect competition and laissez-faire were intended simply as starting points for the analysis, but they became the benchmark for the bulk of mainstream analysis.
The fact that Marshall’s methodology has been open to conflicting interpretations is due in large part to the conflict between the methodology implicit in much of his theorising and the methodology he professed (Pratten 1998). But, whatever his impact on methodology (intended or unintended), Marshall is also important for his indirect methodological contributions. First is the part he played in establishing the Economics and Politics Tripos at Cambridge and of moving economics from the Moral Sciences Board, with profound effects on the future development of economics at Cambridge. In promoting the idea of economics as a mature discipline, Marshall put great emphasis on continuity of ideas (as in the first Preface to the Principles), rather than discontinuities. As a result, he distracted attention from the ways in which his own methodology differed from the other approaches in Britain and further afield.
It was in this respect that Marshall is also important for promoting his protege, John Neville Keynes, the father of John Maynard Keynes. Keynes’s methodological treatise, The Scope and Method of Political Economy, served to support Marshall’s efforts to establish economics as a discipline (Deane 1983)[2] and has had an impact on the wider discipline which continues to this day. For Neville Keynes, the subject matter of economics was wealth: ‘By economic activities are meant those human activities that direct themselves towards the creation, appropriation, and accumulation of wealth; and by economic customs and institutions, the customs and institutions of human society in regard to wealth’ (Neville Keynes 1904: 2). This definition is not inconsistent with the classical view of economics as being chiefly concerned with production, distribution and exchange, or Marshall’s (1890: 14) more vague ‘ordinary business of life’. But Keynes’s definition was to be superseded by the Robbins definition in terms of scarcity, which diverted attention from customs and institutions to rational choice and deductivist methodology.
However, a particular understanding of Neville Keynes’s methodology itself has had a lasting impact. In fact, to the extent that mainstream economics textbooks start with a methodological statement, it includes a version of his position on method. Keynes set out a hypothetico-deductivist methodology, which seemed to reconcile the two opposing positions of deductivism and inductivism. This was achieved by treating each as the counterpart of the other, such that rational argument was used for discovery while empirical evidence was used for confirmation of conclusions. But for Keynes induction was not the logical process in the Newtonian tradition of applying the mind to experience, but rather the collection of evidence on repeated instances of constant conjunctions of events. Rather than the transcendent middle way of the Cambridge tradition, therefore, this was simply a synthesis of deductivism and inductivism within a positivist logical framework.
Keynes’s methodological analysis was well suited to seeking a synthetic solution to the Methodenstreit, being carefully worded and heavily qualified and thus open to a range of interpretations. But Keynes’s emphasis was much more on the side of deductivism (Deane 1983; Moore 2003), not least because of his stated aim to identify uniformities (albeit with the ceteris paribus qualifier). Keynes did see evidence as having importance for confirming not only theoretical propositions but also the assumptions on which they are based. Like Marshall, Keynes had doubts about the concept of economic man. But the tendency was for this (realist, Cambridge) aspect of his methodology to be ignored. Once the rational choice assumptions were widely taken as uniformly supported by evidence (or self-evidently true), they seemed no longer to require confirmation; they could therefore provide a universal basis for hypotheses. By trying to set out a synthesis of competing methodological positions, Keynes ended up downplaying Marshall’s emphasis on real evidence, encouraging the deductivist reading of Marshall and supporting the rising tide of deductivist marginal analysis. Inductive argument was effectively limited to testing theories. Although presented as a middle way, in fact Keynes’s hypothetico-deductive methodology encouraged a deductivist approach to establishing hypotheses.
This influence was reinforced by Keynes’s discussion of the distinction between positive, normative, and applied economics (which he had earlier identified as characteristic of the deductivist position). While Hume had already distinguished the first two concepts, Keynes went much further in positing that a value-free economics was possible, to which values could later be added.[3] [4] As Deane (2008: 725) points out, this was a device to insulate the ‘hard scientific core of economic theory.. .from the charges of ideological bias, or immorality, or relativity, as well as from failures in practical economic policies’. But as a result Keynes provided a rationale for deductivist methodology which could be presented as if value-free.
By supporting Marshall’s drive to establish economics as a mature discipline, with intellectual continuity and methodological consensus, Keynes had helped to suppress important methodological issues and provided a foundation for Mill’s increasingly dominant approach. Indeed, many at Cambridge proceeded within that framework. But, while it would seem that an important Cambridge contribution to economic methodology was to develop and promote a methodology which had emerged elsewhere, the next generation of Cambridge economists departed from the apparent consensus and made a series of influential contributions to methodology which can be traced back to the older Cambridge tradition.11
- [1] There is a parallel with the Scottish analytical history tradition which can arguably be traced throughHegel, but it was Hegel who was the direct influence on Marshall (Groenewegen 1995 and Martins 2013).
- [2] Keynes has been seen as setting out a rationalisation of Marshall’s methodology (Whitaker 2008).
- [3] Colander (1992) has revived this framework in order to argue for a pluralist methodology for appliedeconomics, but where the theory to be applied is positive, and derived by means of deductive logic: ‘Theart of economics is applied economics. It relates the lessons learned in positive economics to the normative goals determined in normative economics’ (ibid.: 192).
- [4] While Weintraub (2005) challenges the idea of a continuing Cambridge tradition stemming fromMarshall, given the diversity of approaches within Cambridge, our focus here is on distinctive Cambridgecontributions. What is emerging is that one such contribution is the continuing thread of a distinctivemethodological approach stemming from Newton.
