In False Prophets of Economics Imperialism, Watson (2024) tells the story of how the pursuit of mere derivational unification has enabled economic imperialism and the spread of market models. By derivational unification, he means the ability of ‘deriving large classes of explanandum sentences from a parsimonious set of premises, theoretical structures, or inferential patterns (Mäki, 2009, p. 363)’. He contrasts this with ontological unification, which is about discovering that a small number of entities, causes and mechanisms indeed drive a diverse set of phenomena. To add some meat to this, we can use Lagrange's equations to describe both systems of gravitating masses and electric circuits. Yet, this does not mean that we have achieved unification of the two phenomena in any physically interesting sense. Similarly, the fact that market models can describe different phenomena does not mean that we have achieved unification in an economically interesting sense. If derivational unification is not enough, how then can we justify the widespread use of market models? According to Watson (p. 10), ‘only the generation of more plausible explanations of actual lived experiences would allow economics to pass such a test’. However, economics does not deliver such explanations. Building on this diagnosis of our current state, the book tries to explain how things could go so wrong. Watson's narrative starts with primers on the history of mathematics and philosophy of science. It then runs from Jevons, via Robbins and Samuelson, to Arrow and Debreu to show that economic imperialism is the outcome of a highly contentious process in which wrong turns were taken at every stage. The chapters draw on a rich set of sources from different disciplines that Watson effortlessly weaves into an entertaining broadside against economics. In this regard, this book is for scholars whose fields have felt the reach of economic methods and, in particular, market models. It offers a detailed and at times provocative account of the pre-history of economic imperialism, presenting an alternative narrative to the triumphant stories told by its advocates. Consequently, Watson does not set out to convince those who welcome the transfer of economic models into other domains. Instead, he provides a lens and solace for those who question these incursions. Despite these explicit intentions, the book might also offer open-minded and thick-skinned economists a perspective on the assumptions, limits, and consequences of applying market models and other mathematical models beyond their original context. Despite these virtues, we would like to point out some features of the book that will give market modellers the feeling that they are being treated unfairly. For instance, consider Robbins definition of economics as the ‘science which studies human behavior as a relationship between ends and scarce means which have alternative uses (p. 139)’. While Watson spends a great deal of time explaining where Robbins' definition may have originated, considerably less effort is devoted to making Robbins' definition precise and to assessing its merit. We are told that Robbins' ‘unflinching commitment to demarcating economics as an autonomous subject field dedicated to studying a separate category of human behaviour seems to be a good example of the bogus scientific unification that the later Pareto warned against (p. 147)’. The problem with this depiction of Robbins is the level of generality and detail at which it is presented. At times, it is hard to distinguish Watson's depiction of Robbins's view from Mill's opposing vision of economics. For Mill, two counter-motivations are inseparably bound to the pursuit of wealth: aversion to labour, and enjoying (costly) consumption (Hausman, 2008, p. 43). He, therefore, proposes that economics should only focus on the ‘laws of such phenomena of society as arise from the combined operations of mankind for the production of wealth, insofar as those phenomena are not modified by the pursuit of any other object’. While Mill understood economics as the study of the production of wealth, he also suggested focussing on a ‘separate category of human behaviour’, that is maximising wealth while minimising labour and self-denial. Yet, Mill's definition predates the marginalist and is therefore not part of the pre-history of economic imperialism that Watson tries to tell. That left us wondering whether an economics that had followed Mill more closely would also be a (‘bogus’) imperialist science according to Watson, and how he would assess attempts to study the motives behind economic behaviour more closely. Would such a move be welcomed? Afterall, behavioural economists have also voiced their imperialist ambitions (e.g. Buyalskaya et al., 2021). Watson also depicts the forbears of economic imperialism as pursuing clear agendas, whereas a more charitable depiction would portray them as interested in genuine and open inquiry. For instance, Watson (p. 138) tells us that Robbins ‘wished … for economists' practice to revolve around a self-consciously analytical definition, thus moving … from a “field-determined” to a “discipline-determined” conception’. Yet, this is not merely a wish of Robbins. He instead argued that a definition focused on ‘material wealth’ or ‘the exchange economy’ cannot capture the valuable work already being done under the label ‘economics’ in the 1930s. For example, work on the economic aspects of war and education. What was needed was an explanation of why economists studied these phenomena. Robbins' answer to this puzzle is that it is the relationship between the tools of war and the services of the schoolmaster to given wants that makes them the subject of economics. On top of that, Watson's discussion of the mathematical foundations of economic imperialism leaves us somewhat puzzled. Watson distinguishes ‘mathematics by definition’ and ‘mathematics by description’ (p. 71). Watson takes this distinction to be of central importance for metatheoretical decisions concerning mathematical modelling in economics (Watson, p. 71). Yet, while there is plausibly a useful distinction in the practice of mathematical modelling that could be captured by these terms, Watson advances a stronger claim: If there is no ‘single mathematics’ (p. 68), then there can be no single mathematical economics and, consequently, no single approach to the mathematisation of economists' market models. On this basis, he concludes that it makes little sense to think in terms of ‘better’ models displacing those that have ‘self-evidently had their day’ (p. 69), and more generally that it makes no sense to evaluate mathematical models as superior to alternative tools and practices within a discipline. While we agree that mathematical models are not inherently superior to other methods, the connection Watson draws between these methodological claims in economic practice and proof-theoretic and foundational discussions in mathematics remains somewhat unclear and, at certain points, appears to rest on imprecise and potentially misleading simplifications. In particular, Watson appears to associate what he calls ‘mathematics by definition’ with figures such as David Hilbert, whom he characterises as wanting to rid intuition out of mathematics (e.g. p. 47). Yet, Hilbert did not seek to eliminate intuition altogether. Rather, he sought to restrict its role to the domain of finite mathematical objects and processes (Tait, 1981). This already obscures the conclusion Watson wants to draw from this treatment of the history of mathematics. More fundamentally, Watson conflates distinct levels of analysis. Hilbert's Program was explicitly metamathematical, concerned with proving consistency of formal systems. The failure to secure finitistic consistency proofs for second-order arithmetic does not preclude using arithmetic to model economic behaviour. Economic models do not aim at proof-theoretic completeness, nor are they evaluated according to consistency proofs in the Hilbertian sense. Invoking Hilbert to cast doubt on the representational content of concepts such as ‘consumers’ or ‘workers’ (p. 60) in economic models risks committing a category error. In light of these points, we think that Watson's critical account will generate more antagonism than need be and significant questions about scope remain unaddressed. How does Watson's analysis apply to contemporary cases such as behavioural economics, which operates with fundamentally different methodological commitments, yet also harbours imperialistic ambitions? Moreover, Watson's historical reconstruction elides important nuances in his treatment of figures such as Robbins and Hilbert. While the book makes a valuable contribution by opening up important debates, his alternative narrative would benefit from taking the problems those scholars tried to solve more seriously. Nevertheless, we think it is a valuable book for readers who want to see the optimistic narrative of economic imperialism challenged through a reconstruction of the choices made throughout the history of economics. Those seeking a fair assessment of those choices just have to look elsewhere. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.
Beck et al. (Sun,) studied this question.