Abstract This article analyses the influence of Robbins’s thought on the subsequent development of economics imperialism. First, the analysis of some passages that have been overlooked up to the present, in which the author explicitly expresses his disbelief in the omnipotence of the economic method, makes it possible to argue that Robbins did not share the confidence exhibited by the advocates of economics imperialism in the capacity of economics to explain non-economic phenomena. Second, this article shows that Robbins describes the influence of real scarcity on human behaviour in terms of a (static) constrained maximization problem, thereby confusing issues of method and scope in his writings. This confusion facilitated the view that economics is a method without a proper subject matter and its later expansion into other fields. In conclusion, although Robbins cannot be counted as an earlier promoter of economics imperialism, the misunderstandings implicit in his writings paved the way for the emergence of this intellectual movement. 1. Introduction In the past 60 years or so, there has been increasing interest among economists in topics that were previously considered to be intrinsically non-economic and thus inaccessible to them (see Ierulli and Tommasi, 1995, p. 1). As a consequence, an ever-increasing body of research has appeared whose topics range over issues such as the family, suicide, religion, politics, law, biology and so on.1 As Sen (1989, p. 326) pointed out, the common denominator in these studies is that they approach these topics from the methodological standpoint of economic science, to which I would add, in whatever way economic methodology is conceived. The large variety of topics that are currently covered by our science has led economists to the conviction that ‘nothing is sacred’, as emphasised by the title of Barro’s book. This conviction is widespread because ‘economic reasoning is not just mathematics and [can] be applied to a wide variety of social problems … I think that no forms of social interaction—including religion, love, crime and fertility choice—are immune from the power of economic reasoning’ (Barro, 2003, p. xiii). Therefore, our science is no longer thought of as being defined by its subject, but by its method (see Becker, 1976; or Ierulli and Tommasi, 1995, p. 1). This relentless expansion of economic science into other areas of knowledge has become known as economics imperialism.2 Looking at this academic phenomenon with intellectual curiosity, at least one obvious question comes to mind: What prompted economists to change their minds about the questions they could address with their science? When this issue is investigated by reviewing the economic literature, different types of evidence are encountered that invariably point to Robbins’s Essay on the Nature and Significance of Economic Science (hereafter Essay), and especially to the definition of the science it contains, as the origins of the expansionist program of economics. In this sense, the substantivist-formalist debate in Economic Anthropology constitutes an example of circumstantial evidence that could have helped to build up the former interpretation of Robbins’s Essay. Being one of the earliest skirmishes between economics and other social sciences, this well-known intellectual clash confronted the followers of the conception of anthropology offered by Polanyi—who argued that ‘orthodox’ economic theory could only be applied to the study of market societies, which are but one of the different ways the economy can be collectively organized3—with another group of anthropologists, who considered that economic theory was applicable to the study of any kind of society, whether or not its economy was organized through a market system. Though this dispute was scarcely relevant to the subsequent appearance of economics imperialism, the fact that this second group of authors generally agreed with Robbins’s definition4 clearly favours the conclusion that his conception of economics fostered the expansion of this science. More relevantly, a similar impression follows from reviewing the work of more recent authors who endorse or directly engage in imperialistic ventures, because they usually cite Robbins’s definition with approval and disregard the definitions of the science proposed by other economists (see Becker, 1976; Stigler, 1984; and Hirshleifer, 1985). In the vast literature on Robbins’s Essay, the relationship between his definition of the science and the emergence of economics imperialism is also described in various ways, which differ in the role granted to Robbins in the history of this intellectual movement. At one end of the spectrum are those authors who consider that Robbins’s definition simply made room for economics imperialism. For example, Backhouse and Medema (2009A, p. 813) found a correlation between the acceptance of Robbins’s definition within economics and the expansion of the economic science into other disciplines and, consequently, they maintain that this definition ‘laid a foundation that could be seen as justifying … the imperialism of economist’s ventures’ (Backhouse and Medema, 2009A, p. 805). Mäki (2009, p. 358) similarly argued that Robbins’s definition ‘opens the door wider’ than other definitions, which explains why it has been ‘particularly suitable for explanatory expansion’ (p. 356). In a similar vein, Medema (2015, p. 71) wrote that ‘the artistic license for economists to cross over into subjects traditionally noneconomic in nature came via Lionel Robbins’s Essay on the Nature and Significance of Economics Science in 1932’. A few lines below this statement, Medema concluded that ‘there can be no question, however, that Robbins defined economics in a manner that naturally allowed for its extension beyond the analysis of standard market phenomena’ (p. 71). Other authors have granted a more decisive influence (or active role) to Robbins in that his definition oriented the discipline towards the study of all human behaviour. This position can be first found in Scoon (1943, p. 311), for whom Robbins’s definition extended the scope of economics into other areas of human behaviour such as the political, the military, the legal and so on. More recent authors who hold this view include Hausman (1992, p. 91), who considered that Robbins’s definition makes economics a study of most human behaviour, rather than the study of a particular domain; Giocoli (2003, p. 86), who argued that ‘following Robbins’s definition, the discipline does not apply solely to the analysis of the market system, but is concerned with every instance of means-ends behaviour regardless of the surrounding institutional framework’; or Marciano (2009, p. 128), who held that the outcome of Robbins’s reasoning is the suppression of the limits of the subject matter of the science. Finally, at the other end of the spectrum, there are those authors who consider Robbins to be an imperialist; that is, they implicitly or explicitly suggest that Robbins shared the imperialists’ conception of the science. For example, Harcourt (1979, p. 243) suggested that Robbins was one of the two purveyors (with Wicksteed) of the views defended by economist imperialists. Similarly, and taking into consideration Siegers’s (1992, p. 531) account, the interdisciplinary character of economics ‘look[s] back on a solid tradition’, which—according to this author—begins with Robbins and integrates the contributions of other authors, such as Becker, Hennipman and Lindendberg (p. 544). Finally, the sociologist Lars Údehn (2001, p. 123) argued that ‘Robbins was … a precursor of theoretical economic imperialism, and also made some suggestions in this direction…’. As suggested by this brief review, although it seems unquestionable that the emergence of economics imperialism in economic theory is related to Robbins’s Essay, the actual position of Robbins regarding the limits of economic analysis is open to question, despite the fact that it is a widespread opinion that Robbins would have endorsed the conception of economics as a ‘universal’ science. In other words, conventional wisdom considers that the belief in the omniscience of economic reasoning that characterises many instances of economics imperialism has its roots in Robbins. Kizner (2000, pp. 258–59) spelled this general perception out5 when he stated that ‘In one sense, Becker’s position reflects an old lesson which economists learned from Lionel Robbins in 1932 … the self-same economizing, allocative aspect of human behaviour which is salient in, say, the commercial areas of life conventionally dealt with in economic theory, is present, Robbins taught us, also in the religious or cultural fields of human endeavour … It might seem, then, that Becker’s extension of economic analysis … is simply the consistent application of the “Austrian” (Robbins-Mises) position’. However, a few lines below this statement, Kirzner criticized this view and explicitly said that he intended ‘to dispel such an impression. The Austrian tradition in economics, which both Robbins and Mises were articulating, does not by itself lead to the economic imperialism of which Becker has, not without cause, been accused’ (Kizner, 2000, p. 259). Here Kirzner is clearly referring to certain Austrian views (e.g. Von Mises; see O’Brien, 1990, p. 162; or Loasby, 1991, pp. 54–55) the young Lionel Robbins was sympathetic to, which argued for a universalistic conception of scarcity and choice. Kirzner separates this Austrian standpoint from the ones that, sharing a similar stance on the ubiquity of scarcity and choice, additionally advocated the adoption of a mathematical-optimization framework to study them—which are the ones that Fine and Milonakis (2009, pp. 30–31) associate with the neoclassical marginalist approach. The possibility that Robbins can be properly ascribed to the Austrian camp, as proposed by Kirzner, suggests that an even more radical reading of his position on the applicability of economics outside its traditional field stems from his writings—one that does not trace the roots of this intellectual enterprise back to his writings. This is a clear indication that it remains an open issue whether this author should be properly classified as an imperialist. This article attempts to contribute to this debate by suggesting that the influence Robbins’s thinking exerted on the later expansion of the science follows a different (and more intricate) channel than the simpler one commonly acknowledged in the literature. Clarifying this influence is relevant not only to obtain a clearer picture of Robbins’s thinking, but also to obtain a better understanding of the recent history of economics and the actual reasons why economics has invaded other fields of research. This article is organized as follows: the next section presents textual evidence from Robbins—some of which has passed unnoticed up to the present—that demonstrates that the views he held about the economic method clash with those defended by the proponents of economics imperialism, as will be shown; Section 3 takes advantage of the results of my previous research—which shows that there is an implicit confusion between the method and the subject matter of economics in Robbins’s writings—and analyses the reasons why Robbins could have confused these different aspects of the science; Section 4 explains in greater detail the implications of this confusion and sketches a plausible (and indirect) way in which it may have facilitated the appearance of economics imperialism; and Section 5 provides the conclusions. 2. Robbins’s views on the power and applicability of the economic method Mäki’s (2009) analysis of economics imperialism provides us with a suitable framework for our investigation. According to this author, two types of economics imperialism should be distinguished. On the one hand, we have economics imperialism, which he defines as ‘a form of economics expansionism where the new types of explanandum phenomena are located in territories that are occupied by disciplines other than economics’ (Mäki, 2009, p. 360). On the other hand, there is economics imperialism*, ‘a form of economics expansionism where the new types of explanandum phenomena are located in territories that are occupied by disciplines other than economics, and where economics presents itself hegemonically as being in possession of superior theories and methods, thereby excluding rival theories and approaches from consideration’ (p. 374). This author considers that although the first type of imperialism is not necessarily wrong or intrinsically misguided, the second form should be avoided. Hence, we can make use of Mäki’s descriptions to investigate whether some features of these two types of economics imperialism are present in Robbins’s texts. 2.1 Robbins and economics imperialism* Let us first consider whether Robbins’s conception of the science fits within the economics imperialism* category as depicted by Mäki. The distinctive trait of authors whose perspective can be put under this heading is their conviction in the hegemonic character of economics, such that they regard this science as the ‘queen of the social sciences’.6 Thus, if Robbins were an economics imperialist*, one should expect to find passages where he explicitly claimed the superiority of the economic method or at least defended its sufficiency to deal with economic issues. However, the evidence in his texts points in exactly the opposite direction, because in some passages that have been overlooked up to the present, Robbins (1949) explained that, even in the spheres in which the economic interpretation can be applied, ‘there are still significant limitations to the extent to which our technique, unaided by other knowledge, can yield results which are practically helpful’ (p. 102). More specifically, he explicitly stated that a knowledge of economics by itself is not a sufficient guide to the practicability of policies because it is possible to devise a plan which is perfect from the analytical point of view, but that is administratively impracticable.7 To avoid misinterpretation, Robbins explained that he did not mean that economists should remain aloof and abstain from participating in the grand debate of political economy; on the contrary, he explicitly recommended that economists engage in these discussions, but at the same time he advised them to supplement their technical training with other kinds of knowledge—e.g. history, political philosophy, public administration and so on (p. 104). Similar ideas are displayed by Robbins (1953) in a subsequent paper in which he criticised Sir Dennis Robertson’s position that the series of changes economic science was experiencing at that time would diminish its utility and make it impossible for economists to give helpful advice. After refuting this view and arguing that economists can provide limited but relevant knowledge to the solution of social problems, in a passage that has been equally overlooked, Robbins concluded that their differences stem from the fact that he had ‘a narrower view of what can be got out of economics alone and exhort the economist, if he hopes to be at all useful, deliberately to look outside’ (p. 111).8 In the concluding paragraphs of this paper, Robbins warned against the risks of putting too much expectation on what can be got out of economics alone. He also praised ‘the great masters’ of economics, because their educational background included much more than mere economic theory, such that they were able to invoke history and politics with authority. Moreover, there is evidence that in later life Robbins held the same views regarding the necessity of complementing education in economics with the study of other disciplines (see Robbins, 1998, pp. 244–45).9 Furthermore, in another work, Robbins (1939) expressed in passing his misgivings regarding the validity of the conclusions derived from the application of the economic method, and admitted the possibility that a logically consistent theory (or model) could produce a false explanation from an empirical point of view because ‘to establish a logical possibility is far from establishing an historical influence or a present tendency’ (pp. 120–21).10 Note that these ideas sharply contrast with the views expressed by the proponents of economics imperialism*, to whom it is just the logical consistency of its approach which provides economics with the power to be applied to other territories (e.g. see Lazear 2000, p. 99). Contrary to this position, the previous paragraph shows that Robbins considered that it is possible to produce logically correct models that do not actually explain reality. No doubt Robbins was referring to the field of applied policy in the textual evidence cited—which he named ‘Political Economy’ and considered to be different from the pure science of ‘economics’ (see Colander, 2009, and note 8 above). The fact that Robbins’s misgivings about the economic method were only related to its capacity for solving real-world problems does not affect the validity of the conclusions reached, for one of the defining characteristics of economics imperialism* is the view that the economic approach alone suffices to shed light on all kind of applied problems—not to mention on questions of applied policy.11 In this sense, Robbins himself put into practice this theoretical distinction during the many times he became involved in issues of applied policy during his life. For example, at the beginning of the 1930s he served on a committee assembled by the Economic Advisory Council to discuss the causes and remedies of the Great Depression; during the Second World War he joined the Department of War, first as an economic assistant, later as the director of the Economic Section, and in 1944 he participated in the Bretton Woods Conference; finally, after the war, he also worked at the National Art Gallery and in the Committee of Higher Education (see Howson 2011).12 2.2 Robbins and economics imperialism In view of the foregoing passages, one can safely hold that Robbins’s conception of the science was at odds with that entailed by economics imperialism*. Nevertheless, there remains the question of whether his views on economics implied the universal applicability of the economic method; that is, it remains to be elucidated whether he can be counted as an earlier promoter of economics imperialism. In fact, as we saw in the introduction, some authors really regard Robbins as a purveyor of this intellectual movement. Even Údehn (2001, p. 123) presented some textual evidence from Robbins in which he suggests that at least some part of political science falls within the scope of his definition of economics, which therefore seems to support the conclusion that Robbins was suggesting a direction in which economics could be expanded. However, some uncertainty emerges about the validity of this conclusion if we consider the whole section of the chapter from which that text is excerpted, because a completely different picture emerges from Robbins’s texts. In fact, in the seventh section of Chapter 5 of his Essay, Robbins stated that economics cannot be used to explain changes in the data that economists take as given in their analyses, which is exactly the opposite view to that of economics imperialism. In these pages, Robbins successively denied the capacity of economics to predict changes in population, techniques and political science, although in passing he made some minor concessions regarding the applicability of economic analysis.13 Moreover, Robbins closed this chapter by stating that there is a region of human behaviour that Economics must take as ultimate data (Essay, p. 135). In other words, Robbins was pronouncing against the possibility of endogenizing the elements that are considered to be exogenous in our explanations, which is exactly what economics imperialism proposes, for he was denouncing that we, as economists, are not well equipped to study just those aspects of reality we take as given. This is the ultimate reason why, for example, Robbins argued that economic analysis ends when the process of explanation comes across tastes since ‘both individual valuations and technical facts are outside the sphere of economic uniformity … from the point of view of economic analysis, these things constitute the irrational element in our universe of discourse’ (Essay, p. 106, emphasis in the original).14 This position drastically contrasts with the opinion that Becker and Stigler (1977) exhibited with regard to the same matter. According to these authors, economists should not argue about tastes, not because we lack the proper tools to investigate them, but ‘for the same reason that one does not argue over the Rocky Mountains—both [tastes and the Rocky Mountains] are there, will be there the next year, too, and are the same to all men’ (p. 76). Instead, they argued that our toolkit suggests other directions by which to generate plausible explanations: to look at changes in the variables that determine the restrictions faced by the economic agents, mainly prices and income, in whatever ‘subtle forms’ these variables may take.15 When additional evidence is considered, the widespread interpretation of Robbins’s position on the ability of economics to explain ‘non-economic’ facts becomes untenable. In the notebooks, correspondence and drafts consulted by Howson to examine the origins and nature of Robbins’s Essay, there is an excerpt in which Robbins argued that his Essay ‘always intended to be a sort of preliminary manifesto designed to forestall the criticism that I did not know where the borderline between the different disciplines really lay’ (Howson, 2004, p. 417). Once more, this statement runs against the views expressed by the promoters of economics imperialism, who usually advocate for a unified social science under the leadership of economics (e.g. Brunner, 1987; Hirshleifer, 2002, p. xi; or Radnitzky, 1992, p. 3), whereas Robbins was clearly suggesting that there are limits between the different social disciplines that should be respected. Even more revealing is the statement made by Robbins in his 1949 paper that economic reasoning has limited power, in the sense that there are problems to which ‘its technique is inapplicable’ (Robbins, 1949, p. 100). Again, this assertion clashes with the opinions usually expressed by economist imperialists, who consider that economics is applicable to the study of all types of human behaviour (see Frey, 1999). In the same paper, Robbins also alerted economists to the danger of ‘overdoing the claim for economic interpretations’. With this expression, Robbins referred to a tendency he detected at that time and which he described in the following passage: In the century of the coffeehouse intellectuals everything must be a hundred per cent. In all sorts of connections, we are witnessing an overvaluation of the economic which is liable to lead to interpretations no less myopic and ill-proportioned, no less likely to confuse our understanding of the complex springs of action that any which tended to ignore it. … The world is a very complicated affair. To understand it at all we must simplify. But woe betide us if we simplify to such an extent that essential ingredients are left out. (Robbins, 1949, p. 102) Although it is obvious that this paragraph cannot be read as a direct attack on economics imperialism, for Robbins was clearly referring to the forms of economic interpretation existing at that moment, it provides indirect evidence on this point. Note that, had Robbins championed the universal applicability of economic reasoning, one would have expected different statements emerging from his pen; that is, assertions that would run no risk of producing myopic and ill-proportioned explanations due to considering all problems from the economic point of view. Moreover, this reading is also in accordance with Robbins’s (1979, p. 997) recognition that although much of human conduct may have an economic aspect, it is extra-marginal as regards systematic study. The textual evidence presented in this section leads to the conclusion that there is a case for the claim that Robbins did not share the conception of the science of economics imperialism* and did not advocate economics imperialism—or, at least, this textual evidence makes this interpretation plausible. However, this conclusion immediately raises the following question: How is it possible that Robbins’s definition has been read as an invitation to expand the economic method into other fields of research, given his stand on the universal applicability of economics? 3. Discovering the foundations of the imperialistic reading of Robbins’s writings The fact that some of the most relevant proponents of economics imperialism (and economics imperialism*) have seen in Robbins’s definition a justification to expand the limits of our science is an indication that there must be something in Robbins’s writings that suggests this possibility. On the other hand, given the foregoing analysis, it seems plausible that Robbins did not mean that economics can be fruitfully applied to study all kinds of social (or non-social) problems, so the rationale for this reading must rest on some confusion or obscure point that lurks in his thinking. A good candidate for this hidden misunderstanding is the illicit association between the concepts of ‘alternative uses’ and ‘multiplicity of ends’ which, although already denounced by Souter (1933, pp. 380–82), has passed unnoticed up to the present. In fact, as I have shown in a previous paper (Falgueras-Sorauren, 2017), this association lies at the root of the confusion that operates in Robbins’s thought between the conditions which, whenever they are present, make human behaviour exhibit the (economic) aspect that is relevant to the science (real scarcity), and the theoretical elements used by the method of the science to capture in a stylized way the problem of economising that appears when the former conditions apply (formal scarcity). The fact that Robbins mixed these two different concepts in his definition of economics makes the method of the science indistinguishable from its subject matter in his writings.16 Clearly, this is the kind of confusion which favours the view that economics can be expanded without limits. Thus, given that the possibility that the imperialistic reading of Robbins is grounded in this muddling of notions has been already well established in my previous paper, in this section I will focus my attention on finding out the reasons why Robbins could have mixed these two concepts. 3.1 Two different approaches to the study of Robinson Crusoe’s basic economic problem Robbins devoted the first chapter of his Essay to fulfilling one of the three aims he had set for his book; that is, to substitute the old and erroneous ‘materialistic’ conception of economics that prevailed at that time for a different one which better described the subject-matter of the science (see O’Brien, 1988, p. 23). To this end, Robbins developed several arguments that show the absurdity of the conclusions regarding different aspects of economics that logically follow from the adoption of those materialistic definitions (Essay, pp. 4–9). Nevertheless, the ultimate reason for this rejection is advanced by Robbins in his examination of Cannan’s exposition of the conditions of wealth for isolated man and society, which is the example this author used to demarcate the sphere of economics (Essay, p. 9). Using the same theoretical device as Cannan (the Robinson Crusoe economy17), Robbins showed that, if one accepts that economics deals with material welfare, one should then admit that Robinson Crusoe’s problem of the division of his time between activities aimed at increasing material welfare and activities aimed at increasing non-material welfare constitutes an economic problem. As even Cannan (1928, p. 17) recognized, the more time Robinson Crusoe devotes to the first type of activities, the greater his wealth (and material welfare) will be. However, Robbins then concluded that this problem of ‘deciding between the “economic” and the “non-economic”’ (Essay, p. 15, emphasis in the original) cannot be fully accommodated within the materialistic description of economics because it involves the use of one immaterial means: time. This drawback justifies the abandonment of the materialistic definitions of the science. In subsequent pages, Robbins tried to determine the actual conditions under which Robinson Crusoe’s basic economic problem has an economic aspect. Note that this requires discovering the features which force Robinson to economise and are not under his direct control; otherwise he could discard the need for economising by simply modifying these features (see Falgueras-Sorauren, 2017). To put it differently, using Crespo’s (2013, pp. 764–65) distinction of concepts, the problem is one of determining the factors that, once present, make Robinson’s behaviour have an ‘economic’ aspect, rather than one of discovering the elements that Robinson takes into account to act ‘economically’—for both words do not refer to the same thing. Since there are some types of Robinson’s resources that have to be economised, the problem should be approached from the analysis of the features that this class of resources exhibits. From this perspective, it turns out that the characteristics that impel Robinson to economise on available means are the limitation in their supply and the existence of alternative uses, as Robbins himself points out (Essay, p. 12ff.). These two attributes suffice to capture the conditions under which scarcity appears in reality and shapes human behaviour, which explains why I included them in the notion of real scarcity (see Falgueras-Sorauren, 2017). However, Robbins also considered Robinson Crusoe’s basic economic problem from a slightly different angle: the perspective of the actor who has to economise. This change of viewpoint should not raise any eyebrows for, as Robbins himself recognized, his Essay was not an attempt to settle a completely new definition of the science, but simply to state, as simply as possible, ‘propositions which are the common property of most modern economists’ (Essay, p. 15; see also Robbins, 1976, p. 146). From 1870 onwards, the professionalization of the science and the international exchange of ideas extended the conception that economics was based on maximizing behaviour (Backhouse, 1985, pp. 123–27). However, this conception of the science described the economic problem from the perspective of the person who is managing with scarce resources and is struggling to get the most from them.18 This is plainly evident in Wicksteed’s Common Sense, which was one of the main sources used by Robbins to elaborate his Essay. Wicksteed understood that economics is about problems of ‘administration of resources’ (see Wicksteed, 1946, pp. 18–94). Similarly, Menger (1994, pp. 94ff.), who was another major influence on the development of Robbins’s Essay, basically understood human economic activity to be a process of economizing on goods or resources. The point is that, from this perspective, the focus of attention shifted from locating the objective characteristics of resources that impel us to economise on them, to one which described the elements that the chooser takes into account to administer them. As a result, Robbins introduced one additional element (multiple ends) that is not an actual determinant of real scarcity, but is relevant to analyse the problem of administration of resources which their scarcity gives rise to. 3.2 Some theoretical assumptions introduced by Robbins in the description of the subject matter of the science In embracing the methodological perspective adopted by the economists of the time, Robbins not only introduced an irrelevant element in his description of the actual determinants of economic behaviour, but he also slid in some theoretical assumptions that accompany it and serve to scientifically capture the problem of administration of resources. Robbins’s well-known defence of introspection as a method to obtain valid economic knowledge (O’Brien, 1990, p. 159; Crespo, 1998; Hands, 2009) probably led to the introduction of these theoretical assumptions being overlooked even to Robbins himself, because this methodological procedure does not critically examine the exact nature of the findings gained by introspection; rather, they are simply considered as being obvious. An initial theoretical assumption that Robbins slipped in is his categorical distinction between ends and means. Robbins conceived actual human behaviour as if there were some elements that, considered by themselves, would always play the role of ends, whereas there were other elements that would always play the role of means. Although this clear-cut distinction is not explicitly defended by Robbins, it is implicit in some of the positions he advances in his writings. For example, it is at the heart of his rejection of ends from the sphere of economics, which he introduced to keep the science distinct from ethics (Essay, pp. 24ff.). This rejection required an unambiguous categorization of human action into ends and means, for only if there are elements that always play the role of ends, and they can be neatly distinguished from those which always play the role of means, the necessary separation of ethics and economics implies the expulsion of ends from the scope of the science. At first sight, the foregoing conception of human action seems plausible, a priori, for two simple reasons. First, since the ultimate ends pursued with our actions19 provide part of the criteria to evaluate human action, the association of ends with ethics seems natural. Second, in our practical activities, we all experience such a clear and drastic division of means and ends. As Crespo (2007, p. 373) suggested, ‘at some stage … of human actions we need to stop the deliberation on ends … and start looking for means, taking the ends as something that has been previously decided’. However, as soon as one pauses to consider human action more closely, the theoretical character of this clear-cut distinction between ends and means comes to light.20 In fact, as Parsons (1934, pp. 523–24) pointed out, human behaviour is actually articulated into a complicated chain of means-ends relationships in which only two elements admit such a radical classification: those which fall under the categories of the ‘ultimate end’ and the ‘ultimate means’. However, in the middle section of the chain, one end or purpose is the means to further ends (Kaufmann, 1933, p. 383); hence, a given link of the chain plays different roles depending on the situation at hand (see Rivett, 1955, p. 219). This consideration also shows that the association between ends and ethics is theoretical and does not necessarily hold in practice, for in this middle section it is perfectly possible to have purely technical ends, which further implies that it is also possible to speak of economic ends, both from an individual and a social viewpoint (e.g. see Prachowny, 1994). In fact, Robbins himself is inconsistent when he recognizes that the price system is a social means for producing and distributing goods (which is an economic goal).21 The second theoretical assumption that Robbins slid into the description of real economic phenomena was the existence of hierarchically ordered goals (Essay, pp. 13–14).22 Robbins thought it a matter of everyday experience that individuals can and do actually order their preferences. To him this fact was so obvious that it did not require controlled experiments to establish its validity (Essay, pp. 75–76, 79). Once more, this a priori assumption may have prevented him from realising that he was claiming too much. As Sudgen (2009, p. 869) correctly explained, it seems reasonable to suppose that, when a person makes a choice between two options A and B, she has been able to rank them according to her ‘feelings’ of preference. However, it does not follow from this ‘fact of experience’ that ‘all the feelings of preference and indifference that a person would have, conditional on every different conceivable choice problem ... can be integrated into a single preference ordering’ (p. 869), which is what Robbins defended in his Essay. However, the existence of a preference ranking is so obvious to Robbins that he introduced this element as an additional determinant of the emergence of an economic problem in reality. Sudgen (2009, p. 868) also pointed out that this leads him to nearly claim that the existence of a preference ordering can be deduced from his proposed definition of economics.23 Note that it is misleading to include the existence of a hierarchy of ends among the required elements for human behaviour to exhibit an economic aspect, because, using the same example as Robbins, Robinson’s behaviour is affected by real scarcity even if he were unable to rank his preferences and, thereby, to make a decision. Furthermore, even if he could not make up his mind and remained paralysed like the ass of the fable, he would still incur an opportunity cost; that is, the outcome he could have obtained if he had made a decision. This opportunity cost reflects the fact that, by not choosing at all, Robinson still spends time doing nothing, and time is scarce because it has other productive uses and not because Robinson is capable of making a choice.24 To stress this idea, a state of indecision does not eliminate the influence of real scarcity on human behaviour, because it is the existence of (productive) uses for time—the resource the chooser is using, and in this case wasting—and not her ability to make decisions which makes it scarce. As before, the failure on the part of Robbins to grasp this point led to some inconsistencies in his thought. For example, there are passages in which Robbins held that our behaviour is affected by scarcity even if we are not perfectly rational or even irrational (see Essay, pp. 90–94; Robbins, 1934, p. 98). But then, the existence of a complete preference ranking, which amounts to being perfectly rational,25 cannot be a determinant of the emergence of real economic problems, as Robbins claimed (see above). In conclusion, Robbins’s ambiguous approach to the problem of discovering the determinants of scarcity in practice led him to surreptitiously (or unconsciously) take the presence of multiple ends on the part of the chooser as one of the causes of this phenomenon. Additionally, his reliance on introspection masked the simultaneous introduction of some theoretical assumptions attached to this last element. This combination of facts may help us understand why Robbins mixed the concepts of real and formal scarcity both in his definition of the science and his writings, which in turn explains how he slipped theoretical elements into the description of real economic facts, consequently blurring the distinction between the method and the subject matter of the science. 4. The implicit association of real scarcity with static constrained maximization problems and its consequences for the later development of economics Having clarified the reasons why Robbins mixed up the method and the subject matter of the science, let us proceed to the final step of our investigation: to trace out the way this confusion may have induced a subset of the subsequent generation of economists to consider that the method of economics could be applied without limits. First, this aim requires obtaining a more detailed picture of the way real scarcity is formally captured by the notion of formal scarcity. Regarding this point, it is not difficult to recognize in the theoretical assumptions conveyed by this notion26 the constitutive elements of a constrained maximization problem. In this type of problem, the clear-cut ends-means separation is captured by the maximization of a given utility function subject to a set of constraints on the other, because, as Killingsworth (1988, p. 1) pointed out, these elements represent ‘what agents want’ and ‘what they can get’, respectively. In addition, the objective function provides the criteria for evaluating the different alternatives (consumption bundles) faced by the agent, whereas the constraints capture both the limitation in the available quantity of the resources and the alternative uses that exist for them, as is clear from any standard microeconomics textbook. The tension between these two elements formally captures the problem of administration of resources that is generated by their scarcity. Therefore, the confusion between real scarcity and formal scarcity in Robbins’s writings implicitly conveys the idea that, by simply arranging the analysis of human behaviour in terms of a constrained optimization problem, the influence of real scarcity on human action is completely captured, so a real scarcity administration problem takes place in reality and the economic science can fruitfully illuminate it. This assimilation of concepts manifests not only in the different passages in which Robbins described the economic aspect of human activity,27 but also in the alternative characterization of the economic problem that appears in Chapter 6 of the Essay, in which Robbins depicted it in terms of the dual formulation of a utility maximization problem; that is, the problem of minimizing the expenditure required to secure given ends (see Essay, p. 145). Note that this conclusion is also in line with the existing view that Robbins was defining economics in terms of a constrained maximization problem (e.g. see O’Brien, 1990, p. 158; Colander, 2009; and Hodgson, 2008, p. 136).28 4.1 Do static constrained optimization problems completely capture real scarcity?29 Before presenting a brief sketch of the impact this confusion had on the later development of the economic science, and to avoid misunderstanding, let me clarify that I am not rejecting the use of constrained maximization problems as a method to formally study the influence of real scarcity on human behaviour; instead, I am arguing that one should refrain from inferring that real scarcity influences human behaviour in some relevant way based on the mere application of a constrained maximization technique to the analysis of human action. It does not follow from the use of this mathematical technique to the formal study of human behaviour that a corresponding problem of administering real scarcity, which can be fruitfully illuminated by the science, takes place in reality,30 or even that this theoretical construction captures all the features of real scarcity. This last point comes to light as soon as we reconsider Robbins’s exemplification of Robinson Crusoe’s basic economic problem. In the discussion that follows his description of this problem, Robbins explained that when the individual has ample means to reach a given end, his behaviour is not affected by real scarcity, and his conduct ‘assumes none of the forms which are the subject of economic science’ (Essay, p. 13). This is incorrect for two different reasons. First, in this example, Robbins ignored the fact that, even if the economic problem is not relevant to the individual chooser—because she has ample means to do whatever she wants—it still can be relevant to the rest of the society—because the means the individual chooser is using are affected by real scarcity.31 As a result of this omission, Robbins failed to notice that the level at which the problem of scarcity becomes relevant can be different in each particular case: it would sometimes be at the individual level, whereas in other circumstances it would only be at the social level, but in most cases it would be relevant at both levels.32 The absence of the social level at which the problem of scarcity manifests is reflected in the fact that Robbins, in his characterization of the economic problem, remained silent on the identity of the chooser who was forced to economise—as Buchanan (1964, p. 214) lamented. This silence on the choosing agent left the imprint on economics that any problem raised by real scarcity could be framed and analysed in terms of a constrained maximization problem, irrespective of the social level which was being considered (see Buchanan, 1964, pp. 215–16).33 Second, the example is incorrect because it implicitly associates the emergence of real scarcity with the size of the cost34 it imposes on human behaviour, whereas the relevant point to this question is whether means are limited and have alternative uses. When these conditions are fulfilled, our behaviour is affected by real scarcity, and there exists a cost of using means. On the other hand, the quantity of limited means that we have at our command determines the size of the cost that real scarcity imposes on our behaviour: the greater this quantity, the lower this cost, because the number of alternative uses that must be sacrificed is also reduced.35 Obviously, as this cost diminishes, our choice becomes less constrained, such that we may even not feel constrained by real scarcity, although in fact we are. That is, whenever limited means have alternative uses our behaviour is constrained by real scarcity, independently of the fact that this constraint is currently binding or not. For this reason, means are scarce even when they are relatively abundant and we do not feel that our choices are constrained by real scarcity. Take the example of the problem of exploiting an exhaustible resource: even if the quantity currently available is enough to meet all our (current) requirements, and we do not feel constrained by real scarcity, it is still conditioning our behaviour and, therefore, the resource is scarce—for we all know, the use of a unit of the resource today will impede the use of the same unit tomorrow; that is, the resource is depletable. This is the underlying reason why it makes sense to consider what the optimal rate of extraction of the resource should be, as Hotelling (1931) does in his famous paper. However, this characteristic of real scarcity is missed in a static constrained optimization problem, which can only represent the influence of real scarcity on the chooser’s behaviour if the relevant constraints are binding. 4.2 A sketch of the influence of Robbins’s confusions on the later development of economic science The false impression that the influence of real scarcity on human behaviour is completely captured by the formulation of a static constrained maximization problem initiated a process which ultimately resulted in the emergence of economics imperialism and that can be summarized as follows.36 Note that this view implicitly equates the problem of making optimal use of scarce resources to maximizing subject to constraints (see Simon and Blume, 1994, p. 411). This association was tightened by the work of later leading economists who pointed out that the common logical structure of the diverse parts of economic theory was maximization under constraints imposed by scarcity (see Backhouse and Medema, 2009B, p. 491).37 The sanctioning of this linkage facilitated the spread of the stance that the essence of economics can be reduced to solving constrained optimization problems (see Dixit, 1990, p. 1),38 and the subsequent convergence between the methods employed at different leading research institutions (Backhouse and Medema, 2009B, p. 498). Once a shared framework of analysis that locates rational utility maximizing agents at its core is constituted (see Brenner, 1980, p. 180), ‘being an economist’ became a matter of adherence to certain techniques or methods (see Ahsman, 2012, p. 6), rather than the kind of topics studied. From this point on, only a very little step is needed to proclaim that economics is simply a research method with no proper subject matter.39 This position conveys the idea that economics can be applied without limits, which is currently a view that has become so widespread that it can be even found in textbooks: for example, in the provocative text by Mckenzie and Tullock (1978) or, more recently, in the excellent manual by Barreto (2009). 5. Conclusion The present article has contributed to the debate on the linkage between Robbins’s thinking and the development of economic imperialism, a topic over which no consensus has yet emerged in the literature. In this regard, the salient outcome of this investigation is to propose a different and more accurate view of this relationship. On the one hand, it has been shown that, if we take into account Robbins’s position regarding the capacities of economic analysis, he cannot be counted as a purveyor of this intellectual movement; even more, in the light of the pieces of textual evidence presented here, it does not seem implausible to argue that he may have even opposed the theoretical incursions of the economists into other fields of research. On the other hand, the investigation has also shown that the imperialistic conception of the science has been indirectly favoured by Robbins’s confusions in his characterization of the subject matter of economics, for they ultimately facilitated the view that the formulation of constrained optimization problems (method) suffices to ensure that a problem raised by the scarcity of resources takes place in reality (subject matter). The resulting thesis is that all these intricacies make the influence exerted by Robbins’s thinking on economics imperialism both subtle and far from easy to trace—hence the term ‘convoluted’ to qualify it. Thus, the proper evaluation and understanding of this influence requires taking into account all these subtleties; otherwise we run the risk of either overstating or belittling the real impact of Robbins’s work. The view put forward in this article has further implications for the recent history and methodology of economics that will be considered in subsequent investigations. For example, it can shed light on the role played by Robbins in the overall unfolding of the science during the past century. Being an eclectic economist, Robbins built his conception of economics mainly on the ideas of Austrian and Neoclassical economists, but he was knowledgeable about the works of the classical economists as well. Consequently, he was able to bring out the relationship between the classical debates on wealth and the (at that time) more recent views on the subject matter of economics (see Falgueras-Sorauren, 2017), which provided his discussion of this issue with a sound philosophical background. This fact, together with his (as we have seen, confusing) depiction of the problems raised by scarcity in terms of constrained optimization problems, may help explain why his characterization of the science served as the keystone over which a new unified method was erected in the 20 years or so following the publication of his Essay, and which lately became the dominant one in economics. In particular, the implications of the analysis developed in this (and my previous) article may deepen our understanding of several issues related to this evolution of the science; to mention just a few of them: (i) How can it be true that the Robbins definition suggested that rigorous mathematical methods could be at the heart of economics, despite Robbins himself not being enthusiastic about mathematical economics? (see Backhouse, 2010, p. 101); (ii) What are the bases for his expulsion of ethical and psychological considerations from economic discourse? This manoeuvre resulted in his methodological prescription for excluding the study of the reasons for changes in tastes and technology from the field of economics—a recommendation that permeated the new economic method and was viewed as one of the sources of its alleged ‘narrowness’ (Klappholz and Agassi, 1959); (iii) What was the impact of Robbins’s views on the subsequent evolution of some schools of thought? In particular, the break it introduced between the ‘old neoclassical’ writers (e.g. Knight) and their ‘modern’ counterparts, and the differences that appeared between this later group and the Austrian camp. All in all, this study on the influence exerted by Robbins’s writings on the emergence of economics imperialism may help us gain insight into some central premises that are generally overlooked but, at the same time, are key to obtaining a correct understanding of his conception of the economic science. Since Robbins’s ideas—with their pros and cons—lurk implicit in many of the ways economics is currently understood and conducted, the task of unveiling and discussing them is essential for the progress of the science—exactly as Robbins himself did with the ideas existing at his time. † This paper was presented at the 19th ESHET Conference held in Rome in May 2015; the 9th Jornadas de la Asociación Ibérica de Historia del Pensamiento Económico held in Valencia in December 2015; and the 2016 Meeting of the Scottish Economics Society held in Perth in April 2016. Financial support from the following institutions is acknowledged: the Junta de Andalucía, SEJ-122, the Ministerio de Educación y Ciencia ECO2011-29355 and ECO2014-53767P. The author would like to thank Ricardo Crespo for his comments on a previous version of this article, as well as the two anonymous referees for helpful and constructive criticism during the revision process. All errors remain my own. Footnotes 1 To cite some examples of articles on such ‘non-traditional’ topics see, for example, Hammermesh and Soss (1974), Azzi and Ehrenberg (1975), Fair (1978) and, more recently, Donohue and Levitt (2001). 2 Mäki (2009, p. 352) distinguishes between economics imperialism and economic imperialism, and in this article I adhere to his classification. 3 See Lodewijks (1994, pp. 88–91) for a review of this debate. Obviously, Polanyi’s conception of economics was neither the only nor the most relevant one that conflicted with Robbins’s at that moment in time. As I have explained in greater detail elsewhere (see Falgueras-Sorauren, 2017, p. 86), Robbins’s targets were the ‘materialistic’ definitions of the science—e.g. Marshall or Cannan’s definition—which equated ‘wealth’ with ‘material utilities’ and, to a much lesser extent, those definitions which located the subject matter of economics in some key features of economic activities such as the presence of prices, money or exchange—e.g. those advanced by Pigou or Davenport. As a referee correctly pointed out to me, by the early 1930s, with the Bolshevik revolution and the rise of totalitarian governments in Europe, economics applicable to non-market economies might be more interesting than previously thought—which might help us understand why Robbins considered that the definition of economics should be capable of explaining the economic phenomena under different institutional settings. 4 Burling (1962), Edel (1969) or LeClair (1962) are exemplars of the way economic anthropology was conceived by this second group of authors. Dalton (1961) is a good example of the ideas defended by its opponents. 5 Khalih (1996) expressed the same view in different terms when he said that ‘Robbins’ definition of the scope of economics came to provide the backbone to the argument of the universality of economics’ (p. 17); ‘[Robbins] certainly has turned on the green light for orthodox economists to export the neoclassical kit into areas which have been traditionally part of other disciplines’ (p. 22). 6 This nickname of economics, which is mentioned by Mäki (2002, p. 3), among others, reflects the high status accorded to this science by many (if not all) advocates of economics imperialism. For example, for Brenner (1980) economics is a maturing science (in fact, the only one), because it is the only social science that has a unified paradigm capable of explaining a wide range of human behavior. Buckley and Casson (1993, p. 1041) considered that economics stands apart from the other social sciences because it is more advanced than other subjects in its theoretical development. They also argued that economics can provide at least some of the core principles to be used in the process of unification of the social sciences (p. 1051). Hirshleifer (1994, p. 3) provocatively manifested his faith in the superiority of economics by predicting that, as economists began to study those modes of economic activity which involve conflict instead of cooperation among the agents involved, they ‘will encounter a number of native tribes—historians, sociologists, psychologists, philosophers, etc.—who, in the various intellectually primitive ways, have preceded us in reconnoitring the dark side of human activity. Once we economists get involved, quite properly we’ll of course be brushing aside these a-theoretical aborigines’ (emphasis added)—and in the footnote he adds: ‘when these researchers do good work, they are doing economics!’ Demsetz (1997, p. 2) defended the primacy of economics on the grounds of its ability to ‘explain phenomena within its traditional boundaries better than the other social sciences have explained phenomena within their respective boundaries’. Lazear (2000, p. 99) began his well-known paper on economics imperialism by stating that ‘By almost any market test, economics is the premier social science’. That this is the image of the science projected by the economics imperialists is corroborated by other authors, e.g. Thurow (1977), Heilbroner (1991) or, more recently, Fourcade et al. (2014). Finally, Klamer (2008, p. 230) suggested that this is the dominant view of the science among graduate students of economics. 7 With regard to this point, Robbins (1949, p. 103) stated that ‘It is quite possible for a man to be a most brilliant analytical economist and yet to show the most childish ignorance both of administrative and political possibility’. This is a statement difficult to reconcile with the imperialists’ conception of the science. 8 This exhortation on the part of Robbins to ‘look outside’ should be interpreted along the lines of his remarks on the participation of the political debate mentioned above, and not as an invitation to engage in imperialistic research. It is related to Robbins’s defense against the accusations that some parts of his Essay were an admonition to economists to refrain from all interest or activity outside its own subject (see Essay, p. viii; or Robbins, 1971, p. 149). In different places Robbins tried to make it clear that this view was a misreading of his texts in that he had never opposed economists engaging in discussions of social philosophy (Essay, pp. viii–ix; or Robbins, 1938, p. 345; 1971, p. 149). The point is that, to Robbins, these ‘discussions of social philosophy’ had nothing to do with the kind of investigation done by economic imperialists. In fact, he grouped those discussions under the heading of ‘Political Economy’ (Robbins, 1938, p. 346, fn.), which ‘is a discussion of principles of public policy in the economic field … and while it makes appeal to the findings of economic science, it also involves assumptions which, in the nature of things, lie outside positive science and which are essentially normative in character’ (Robbins, 1976, pp. 2–3). That is to say, Robbins’s final advice has to be read as an invitation to economists to go beyond our ‘value-free’ scientific research and enter into ‘value-laden’ debates on practical policy issues (Robbins, 1963, p. 7)—see Crespo’s (1998) explanations on the distinction between ‘Economics’ and ‘Political Economy’ in Robbins’s thought. 9 Similar ideas are repeated by Robbins in his Essay (p. ix). 10 In other passages Robbins’s reservations are transmuted into total disbelief: ‘by itself Economics affords no solution to any of the important problems of life’ (Essay, p. ix). 11 For example, the applicability of the economic method to applied policy issues is recurrently advocated by Lazear (2000, pp. 102, 104, 107–8, 118, 120, 125). In addition, many opponents of economics imperialism argue that it is changing practice in areas to which it has been applied—see, for example, Osterloh and Frost (2009) for the case of management practice. 12 For example, Oliveira and Suprinyak’s (2016) analysis of Robbins’s opposition to the conclusions of the committee of the Economic Advisory Council reveals that he was careful to distinguish between his purely economic arguments and those which also included value judgments and fell within his discipline of political economics. It is worth mentioning that, despite this initial clash with Keynes, during the Second World War Robbins worked in close collaboration with him. As a consequence, Robbins changed some of his initial views on this dispute and on the ideas expressed by Keynes in his General Theory—which shows the breath of Robbins’s economic vision (see Howson, 2011, pp. 184ff., 310, 408, 419–20, 666, 1072; or Robbins, 1971, pp. 153–55, 188). 13 For example, Robbins (Essay, p. 132) admitted that movements of population are responsive to money incentives and, consequently, conceded a very restrictive role to an economic theory of the equilibrium of population, but he explicitly denied the capability of this theory to predict changes in the population (Essay, p. 133). Subsequently, he expressed a similar disbelief in the capability of the science to predict changes in technique (Essay, p. 134). Finally, and regarding the political systems, although Robbins conceded that the legal framework is also responsive to economic factors and interests—and, in this sense, (some part of) it falls within the scope of the science—he explicitly denied that economics can be used to predict changes in it (Essay, pp. 134–35). Exactly the opposite views are expressed by Frey (1999, p. 11) in his defense of economics imperialism, to whom ‘economics seems to be better equipped to explain changes in human behavior, while sociology seems to be better equipped to explain historically existing levels’ (emphasis in the original). 14 Furthermore, in other work, Robbins clearly stated the differences in nature between the analysis made by the economists and the psychologists: ‘But, it may be asked, when we descend from these austere regions of pure theory and investigate the disposal of a particular scarce goods, in particular situations, are we not in fact making psychological investigation? … If we are attempting to ascertain the conditions of demand for fish in a particular market at a particular time we are indeed investigating psychological data. But we are not investigating them from the point of view of the psychologist. He wishes, presumably, to know why these things exist and to what law of psychic equilibrium or psychic genetics they conform. We, on the contrary, wish to know simply that they exist in order to discover, in our own field according to the laws of our science, what are the implications of such an existence’ (Robbins, 1934, p. 99, emphasis added). 15 In an earlier paper, Michael and Becker (1973) offered a different, and complementary, reason for his rejection of these kinds of explanations: economists lack the appropriate theory to explain the formation of tastes and neither can they rely on a well-developed theory from any other discipline; therefore, they must resort to ‘ad hoc’ theorizing when explaining differences in tastes. 16 Thus, the notion of real scarcity corresponds to Robbins’s characterization of scarcity as the conjunction of ‘limited means and alternative uses for the means’. It has its counterpart in the following fragment of Robbins’s definition (Essay, p. 16): ‘which studies human behavior’ [when influenced by] ‘scarce means which have alternative uses’. However, formal scarcity corresponds to his characterization of scarcity as the presence of ‘limited means and multiple ends’, which, in turn, is expressed in the remaining part of Robbins’s definition, that is, ‘as a relationship of ends and (scarce) means’ (see Falgueras-Sorauren, 2017, p. 93). 17 Robbins (1926, pp. 224–25) defended the utility of this theoretical device to understand basic economic problems. 18 On this way of conceiving the economic science, see Kizner (1976, pp. 51–70). 19 There are different places in which Robbins equated ‘ends’ with this category of ‘ultimate ends’ in a more or less explicit way (e.g. Robbins, 1927, p. 174; 1963, p. 7). 20 To put this idea differently, although in a formal analysis of human behaviour it is legitimate to classify the elements at hand as means or ends, as Parsons (1934, p. 525) recognized, this practice should not lead to the conclusion that the role assigned in the theory to a given element is in the nature of things. To do justice to Robbins, it must be recognized that there are passages where he recognized the existence of intermediate goals in human action (see Essay, pp. 15n, 31). The problem is that he did not seem to realize that this affected his conception of the relationship between ethics and economics. 21 See Robbins (1939, pp. 116–17) and Robbins (1947, pp. 5, 10), respectively. 22 The discussion of the hierarchy of ends is a complex and problematic subject with a long history in economics. Nevertheless, the details of the many contributions to this topic do not need to be worked out to see how it is relevant to the question here considered. 23 In addition to the text quoted by Sudgen, this manifests in different places of the Essay. Among the conditions under which the division of Robinson’s time between the production of income and the enjoyment of leisure has an economic aspect, Robbins listed the existence of a ranking of ends as a necessary element for an economic problem to appear (Essay, pp. 13–14). Subsequently, Robbins also argued that this assumption is applicable whenever the conditions that give rise to economic phenomena are present (Essay, p. 80). 24 In this respect, Samuelson and Zeckhauser (1988, p. 8) suggested that doing nothing or maintaining the status quo is an alternative that is present in (almost) all real-world decision problems. 25 Robbins implicitly associates the ‘existence of complete preference ranking’ with ‘being perfectly rational’ in different passages of his writings (Essay, pp. 90–94; 1934, p. 100). 26 According to my previous explanations, these theoretical assumptions are, at least, the following ones: a clear-cut ends-means distinction, and the existence of multiple and hierarchically ordered goals. 27 This association is also at the heart of the subtle way in which Robbins first introduces choice in the description of the subject matter of the science and, then occasionally transmutes choice into its actual subject matter—see Falgueras-Sorauren (2017, p. 94). 28 In fact, Robbins incurs a twofold reductionism: he equates real scarcity with the formulation of a static constrained maximization problem, then he equates this type of mathematical problem with the problem of choice. Hence, he is implicitly equating choice with choice under the influence of scarcity. This produces the impression that all kinds of choices can be studied with constrained maximization problems. This position is correctly criticized by Polanyi (1957, p. 246), who asserts that alternatives can appear because of moral reasons or because of the presence of abundant means—so that choice cannot be reduced to economic choice. 29 The following discussion is just one instantiation of the more general problem of the relationship between idealized mathematical models and their intended real-world target, a topic which has received much attention within the Philosophy of Economics during the past 20 years, but on which consensus is still lacking. In any case, the following lines of argument do not attempt to provide a general solution to the problem, but simply attempt to illustrate some of the difficulties lurking in Robbins’s characterization of the basic economic problem. 30 This point is at the heart of the disagreement between Cannan (1932, pp. 426–27) and Robbins (Essay, p. 11), but their failure to distinguish the concepts of real and formal scarcity impede agreement, although both authors offer (partially) correct arguments. Cannan replied to Robbins’s criticism that his definition of the science cannot account for decisions regarding the division of time with the following counterexample: ‘we may have, for example, to decide whether to refuse to work overtime when an urgent order has suddenly come in and thereby offend our employer and damage our industrial prospects, or break an appointment with the young woman we are courting and damage our matrimonial prospects’. Hence, Robbins is right in stating that decisions regarding the use of time have an economic aspect, because they involve the use of a scarce resource and, therefore, are affected by real scarcity. Nevertheless, Cannan is also right in claiming that the former does not mean that all these problems fall within the scope of the science, in that its method does not provide practical guidance to their solution. For example, Cannan’s example cannot be fruitfully analysed with the economic method because it does not provide additional insight into its nature or shed light on its solution. 31 Note that, when a particular person makes use of a scarce resource, the quantity is not available to other members of the society—see my explanations in Falgueras-Sorauren (2017, p. 92), as well as Hazlitt (1979), who (especially in Chapters 2 and 3) provided many examples of how scarcity can be relevant at social levels. 32 In fact, this is the case that underlies the justification of distributive policies. Moreover, in different passages, Robbins describes the economic problem in terms of ‘social objectives’ in conflict (see Robbins 1981, p. 2; 1953, p. 110), which is an unconscious recognition that economic problems of scarcity may appear at the social level. 33 Hence, the Robbins definition implied a reduction of the methods required to properly analyze the economic problems that emerge at different social levels, as Toboso (1994) explains. This reductionism was strongly criticized by Knight, who considered that, whereas individual economic problems can be properly analysed in terms of the strict economic science, the analysis of social economic problems requires a broader approach because they imply discussing the way the economic activity should be organised—on this issue, see Hart (2014, pp. 299–301). 34 Obviously, this is an ‘opportunity cost’: the cost of foregone alternatives. It should be noted that the argument of the text does not conflict with Buchanan’s (1964, 1973) approach to the subject. Here, I am referring to the conditions under which an ‘opportunity cost’ appears, which can be objectively determined, for they depend on the existence of alternative uses of a limited resource. To avoid confusion, this issue should be kept separate from the topics discussed by Buchanan. For example: Is it always possible to calculate the size of this opportunity cost? Is the opportunity cost subjective or objective in nature? Note that the answer to the former question determines the relevance of the latter, but not the contrary: if a given use of a means has no alternatives, nothing is given up when the person acts or chooses, so no opportunity cost appears. In this case, the second group of questions becomes irrelevant. 35 Note that this does not imply that the opportunity cost disappears: each time we use a scarce means, it cannot be used for another purpose. Take the simple example of (monetary) wealth: the opportunity cost of spending money is very low for a very rich person. As he can dispose of a great amount of this scarce resource, the quantity of alternatives that he has to renounce when he spends 1 Euro is nearly insignificant. In this case his monetary constraint is not binding. However, this does not change the fact that each time he spends 1 Euro on a thing or activity, he cannot spend the same Euro on something different—that is the reason why he can fall into ruin! 36 To be sure, I am not claiming that Robbins was the only one responsible for the process explained in the text. Clearly, history is more complex, and Robbins’s work exhibited a great degree of complementarity with the work of other influential authors, especially Hicks ( 2001) and Samuelson (1983). Nevertheless, exploring this issue would require (at least) another paper, and the analysis developed suffices for the aims of this investigation. 37 These authors mainly refer to the work of Koopmans, for this author explicitly mentioned and discussed Robbins’s views. 38 Furthermore, the fact that the different elements of these optimization problems can be read in terms of some key concepts of economics (e.g. the Lagrangian multiplier reflects the opportunity cost or shadow prices of the use of resources: see Dixit [1990, pp. 40–54] or Silberberg [1990, pp. 204–6 and 313–15]) reinforces the process explained in the text. It allows the economist to directly read an empirical content into his analytical tools, which facilitates concluding—via backwards reasoning—that each time a given phenomenon has been framed in terms of a constrained optimization problem, the economic dimension of an empirical problem has been captured—see Klonschinski (2014) for an example of this way of thinking. In other words, the economist gains the impression that all that is needed to analyse real economic problems is to frame the situation at hand in terms of a constrained optimization problem. 39 This view has recently found its way into textbooks—see, for example, Silberberg (1990, p. 2) or Heyne (1997, pp. 4–6). Bibliography Ahsman, S. 2012. Economics and political economy today: introduction to the symposium on Fine and Milonakis, Historical Materialism , vol. 20, no. 3, 3– 8 Google Scholar CrossRef Search ADS Azzi, C. and Ehrenberg, R. G. 1975. Household allocation of time and church attendance, Journal of Political Economy , vol. 83, no. 1, 27– 56 Google Scholar CrossRef Search ADS Backhouse, R. E. 1985. A History of Modern Economic Analysis , New York, Basil Blackwell Backhouse, R. E. 2010. The Puzzle of Modern Economics , Cambridge, Cambridge University Press Google Scholar CrossRef Search ADS Backhouse, R. E. and Medema, S. G. 2009A. Defining economics: Robbins’ essay in theory and practice, Economica , vol. 76, 805– 20 Backhouse, R. E. and Medema, S. G. 2009B. Robbins’s essay and the axiomatization of economics, Journal of the History of Economic Thought , vol. 31, no. 4, 485– 99 Google Scholar CrossRef Search ADS Barreto, H. 2009. Intermediate Microeconomics with Microsoft Excel , Cambridge, Cambridge University Press Google Scholar CrossRef Search ADS Barro, R. J. 2003. Nothing Is Sacred: Economics Ideas for the New Millennium , Cambridge, MA, MIT Press Becker, G. S. 1976. The Economic Approach to Human Behavior , Chicago, University of Chicago Press Becker, G. S. and Stigler, G. J. 1977. De Gustibus Non Est Disputandum, American Economic Review , vol. 67, no. 2, 76– 90 Brenner, R. 1980. Economics-an imperialist science?, Journal of Legal Studies , vol. 9, no. 1, 179– 88 Google Scholar CrossRef Search ADS Brunner, K. 1987. The perception of man and the conception of society: two approaches to understanding society, Economic Inquiry , vol. 25, no. 3, 367– 88 Google Scholar CrossRef Search ADS Buchanan, J. M. 1964. What should economists do?, Southern Economic Journal , vol. 30, no. 3, 213– 22 Google Scholar CrossRef Search ADS Buchanan, J. M. 1973. Introduction: L.S.E cost theory in retrospect, pp. 3– 16 in Buchanan, J. M. and Thirlby, G. M. (eds.), L.S.E Essays on Cost , London, Weidenfeld and Nicholson Buckley, P. J. and Casson, M. 1993. Economics as an imperialist social science, Human Relations , vol. 46, no. 9, 1035– 52 Google Scholar CrossRef Search ADS Burling, R. 1962. Maximization theories and the study of economic anthropology, American Anthropologist , vol. 64, no. 4, 802– 21 Google Scholar CrossRef Search ADS Cannan, E. 1928. Wealth: A Brief Explanation of the Causes of Economic Welfare , 3rd ed., London, P. S. King and Son Cannan, E. 1932. Review of Robbins’ An Essay on the Nature and Significance of Economic Science, Economic Journal , vol. 42, no. 167, 424– 27 Google Scholar CrossRef Search ADS Colander, D. 2009. What is ‘it’ that Robbins was defining?, Journal of the History of Economic Thought , vol. 31, no. 4, 437– 48 Google Scholar CrossRef Search ADS Crespo, R. F. 1998. The rebirth of political economy and its concept according to Lionel Robbins, pp. 233– 48 in Koslowski, P. and Shenk, R., Yearbook for Philosophy of the Hannover Institute of Philosophical Research , Wien, Passagen-Verlag Crespo, R. F. 2007. Practical comparability and ends in economics, Journal of Economic Methodology , vol. 14, no. 3, 371– 93 Google Scholar CrossRef Search ADS Crespo, R. F. 2013. Two concepts of economics and maximization, Cambridge Journal of Economics , vol. 37, no. 4, 759– 74 Google Scholar CrossRef Search ADS Dalton, G. 1961. Economic theory and primitive society, American Anthropologist , vol. 63, no. 1, 1– 25 Google Scholar CrossRef Search ADS Demsetz, H. 1997. The primacy of economics: an explanation of the comparative success of economics in the social sciences, Economic Inquiry , vol. 35, no. 1, 1– 11 Google Scholar CrossRef Search ADS Dixit, A. K. 1990. Optimization in Economic Theory , 2nd ed., Oxford, Oxford University Press Donohue, J. J. and Levitt, S. D. 2001. The impact of legalized abortion on crime, Quarterly Journal of Economics , vol. 116, no. 2, 379– 420 Google Scholar CrossRef Search ADS Edel, M. 1969. Economic analysis in an anthropological setting: some methodological considerations, American Anthropologist , vol. 71, no. 3, 421– 33 Google Scholar CrossRef Search ADS Fair, R. C. 1978. A theory of extramarital affairs, Journal of Political Economy , vol. 86, no. 1, 45– 61 Google Scholar CrossRef Search ADS Falgueras-Sorauren, I. 2017. A new view on Robbins’s definition of economics: the notions of real and formal scarcity, Cuadernos de Economía-Spanish Journal of Economics and Finance , vol. 40, no. 113, 83– 99 Google Scholar CrossRef Search ADS Fine, B. and Milonakis, D. 2009. From Economics Imperialism to Freakonomics: The Shifting Boundaries Between Economics and other Social Sciences , London and New York, Routledge Fourcade, M., Ollion, E. and Yann A. 2014. ‘ The Superiority of Economics’ , Maxpo Discussion Paper 14/3. Frey, B. S. 1999. Economics as a Science of Human Behavior: Towards a New Social Science Paradigm , Boston, Kluwer Google Scholar CrossRef Search ADS Giocoli, N. 2003. Modeling Rational Agents: From Interwar Economics to Early Modern Economic Theory , Cheltenham, Northampton, Edward Elgar Hammermesh, D. S. and Soss, N. M. 1974. An economic theory of suicide, Journal of Political Economy , vol. 82, no. 1, 83– 98 Google Scholar CrossRef Search ADS Hands, W. 2009. Effective tension in Robbins’ economic methodology, Economica , vol. 76, 831– 44 Google Scholar CrossRef Search ADS Harcourt, G. C. 1979. The social science imperialists, Politics , vol. 14, no. 2, 243– 51 Google Scholar CrossRef Search ADS Hart, J. 2014. Frank Knight’s ‘categories’ and the definition of economics, Journal of Economic Methodology , vol. 21, no. 3, 290– 307 Google Scholar CrossRef Search ADS Hausman, D. M. 1992. The Inexact and Separate Science of Economics , Cambridge, Cambridge University Press Google Scholar CrossRef Search ADS Hazlitt, H. 1979. Economics in One Lesson , New York, Crown Publishing Heilbroner, R. 1991. Economics as universal science, Social Research , vol. 58, no. 3, 457– 74 Heyne, P. 1997. The Economic Way of Thinking , 8th ed., Upper Saddle River, NJ, Prentice Hall Hicks, J. 2001. Value and Capital , Oxford, Clarendon Hirshleifer, J. 1985. The expanding domain of economics, American Economic Review , vol. 75, no. 6, 53– 68 Hirshleifer, J. 1994. The dark side of the force, Economic Inquiry , vol. 32, no. 1, 1– 10 Google Scholar CrossRef Search ADS Hirshleifer, J. 2002. Introduction, pp. 9– 11 in Grossbard-Shechtman, S. and Clague, C. (eds.), The Expansion of Economics: Toward A More Inclusive Social Science , London and New York, M. E. Sharpe Hodgson, G. M. 2008. Prospects for economic sociology, Philosophy of the Social Sciences , vol. 38, no. 1, 133– 49 Google Scholar CrossRef Search ADS Hotelling, H. 1931. The economics of exhaustible resources, Journal of Political Economy , vol. 39, no. 2, 137– 75 Google Scholar CrossRef Search ADS Howson, S. 2004. The origins of Lionel Robbins’s Essay on the Nature and Significance of Economic Science, History of Political Economy , vol. 36, no. 3, 413– 43 Google Scholar CrossRef Search ADS Howson, S. 2011. Lionel Robbins , Cambridge, Cambridge University Press Google Scholar CrossRef Search ADS Ierulli, K. and Tommasi, M. 1995. Introduction, pp. 1– 12 in Ierulli, K. and Tommassi, M. (eds.), The New Economics of Human Behavior , Cambridge, Cambridge University Press Google Scholar CrossRef Search ADS Kaufmann, F. 1933. On the subject matter and method of economic science, Economica , no. 42, 381– 401 Khalih, E. L. 1996. What is economic action? From Marshall and Robbins to Polanyi and Becker, Journal of the History of Economic Thought , vol. 18, no. 1, 13– 36 Google Scholar CrossRef Search ADS Killingsworth, M. R. 1988. Labor Supply , Cambridge, Cambridge University Press Kizner, I. M. 1976. The Economic Point of View , Kansas City, Sheed and Ward Kizner, I. M. 2000. The Driving Force of the Market: Essays in Austrian Economics , London, Routledge, accessed 17 April 2015, ProQuest ebrary Google Scholar CrossRef Search ADS Klamer, A. 2008. Does this have to be our future?, pp. 227– 33 in Colander, D., The Making of an Economist Redux , Princeton, NJ, Princeton University Press Klappholz, K. and Agassi, J. 1959. Methodological prescriptions in economics, Economica , vol. 26, no. 101, 60– 74 Google Scholar CrossRef Search ADS Klonschinski, A. 2014. ‘Economic Imperialism’ in health care resource allocation— how can equity considerations be incorporated into economic evaluation?, Journal of Economic Methodology , vol. 21, no. 2, 158– 74 Google Scholar CrossRef Search ADS Lazear, E. P. 2000. Economic imperialism, Journal of Economics , vol. 115, no. 1, 99– 146 LeClair, E. E. 1962. Economic theory and economic anthropology, American Anthropologist , vol. 64, no. 6, 1179–1 203 Google Scholar CrossRef Search ADS Loasby, B. J. 1991. The Austrian school, pp. 40– 70 in Mair, D. and Miller, A. G. (eds.), A Modern Guide to Economic Thought: An Introduction to Comparative Schools of Thought in Economics , Aldershot, Edward Elgar Lodewijks, J. 1994. Anthropologists and economists: conflict or cooperation?, Journal of Economic Methodology , vol. 1, no. 1, 81– 104 Google Scholar CrossRef Search ADS Mäki, U. 2002. The dismal queen of the social sciences, pp. 3– 32 in Mäki, U. (ed.), Fact and Fiction in Economics: Models, Realism and Social Construction , Cambridge, Cambridge University Press Google Scholar CrossRef Search ADS Mäki, U. 2009. Economics imperialism: concept and constraints, Philosophy of the Social Sciences , vol. 39, no. 3, 351– 80 Google Scholar CrossRef Search ADS Marciano, A. 2009. Buchanan’s catallactic critique of Robbins’ definition of economics, Journal of Economic Methodology , vol. 16, no. 2, 125– 38 Google Scholar CrossRef Search ADS Mckenzie, R. B. and Tullock, G. 1978. The New World of Economics , rev. ed., Homewood, IL, Richard D. Irwin Medema, S. G. 2015. From dismal to dominance? Law and economics and the values of the imperial science, historically contemplated, pp. 69– 88 in Hatzis, A. H. and Mercuro, N. (eds.), Law and Economics. Philosophical Issues and Fundamental Questions , Abingdon and New York, Routledge Menger, C. 1994. Principles of Economics (English translation by James Dingwall and Bert F. Hoselitz), Grove City, Libertarian Press Michael, R. T. and Becker, G. S. 1973. On the new theory of consumer behavior, Swedish Journal of Economics , vol. 75, no. 4, 378– 96 Google Scholar CrossRef Search ADS O’Brien, D. P. 1988. Lionel Robbins , Houndmills, Macmillan Google Scholar CrossRef Search ADS O’Brien, D. P. 1990. Lionel Robbins and the Austrian connection, pp. 155– 84 in Cadwell, B. J. (ed.), Carl Menger and His Legacy in Economics, History of Political Economy Annual Supplement to vol. 22 , Duke U. Press, Durhan and London. Oliveira, T. D. and Suprinyak, C. E. 2016. The Economist Quae Political Economist: Lionel Robbins and the Economic Advisory Council, UFMG/Cedeplar Working Paper no. 535, Belo Horizonte Osterloh, M. and Frost, J. 2009. Bad for practice— good for practice: from economic imperialism to multidisciplinary mapping, Journal of International Business Ethics , vol. 2, no. 1, 36– 45 Parsons, T. 1934. Some reflections on ‘The Nature and Significance of Economics’, Quarterly Journal of Economics , vol. 48, no. 3, 511– 45 Google Scholar CrossRef Search ADS Polanyi, K. 1957. The economy as instituted process, pp. 243– 69 in Polanyi, K., Arensberg C. M. and Pearson, H. W. (eds.), Trade and Market in Early Empires , New York, Free Press Prachowny, M. F. J. 1994. The Goals of Macroeconomic Policy , London and New York, Routledge Radnitzky, G. 1992. The economic approach. its achievements and its potential, pp. 1– 68 in Radnitzky, G. (ed.), Universal Economics: Assessing the Achievements of the Economic Approach , New York, Paragon House Rivett, K. 1955. The definition of economics, Economic Record , vol. 31, no. 61, 215– 31 Google Scholar CrossRef Search ADS Robbins, L. 1926. Review of Gustav Cassel’s Fundamental Thoughts in Economics, Economica , no. 17, 223– 25. Robbins, L. 1927. Mr. Hawtrey on the scope of economics, Economica , no. 20, 172– 78 Robbins, L. 1934. Remarks on the relationship between economics and psychology, Manchester School , vol. 5, no. 2, 89– 101 Google Scholar CrossRef Search ADS Robbins, L. 1938. Live and dead issues in the methodology of economics, Economica , vol. 5, no. 19, 342– 52 Google Scholar CrossRef Search ADS Robbins, L. 1939. The Economic Causes of War , London, Jonathan Cape Robbins, L. 1947. The Economic Problem in Peace and War , London, Macmillan Robbins, L. 1949. The economist in the twentieth century: an oration delivered on the 53rd anniversary of the foundation of the London School of Economics, Economica , vol. 16, no. 62, 93– 105 Google Scholar CrossRef Search ADS Robbins, L. 1953. Robertson on utility and scope, Economica , vol. 20, no. 78, 99– 111 Google Scholar CrossRef Search ADS Robbins, L. 1962. An Essay on the Nature and Significance of Economic Science , 2nd ed., London, Macmillan Robbins, L. 1963. Politics and Economics: Papers in Political Economy , London, Macmillan Robbins, L. 1971. Authobiography of an Economist , London, Macmillan Google Scholar CrossRef Search ADS Robbins, L. 1976. Political Economy: Past and Present. A Review of Leading Theories of Economic Policy , New York, Columbia University Press Google Scholar CrossRef Search ADS Robbins, L. 1979. On Latsis’s method and appraisal in economics: a review essay, Journal of Economic Literature , vol. 17, no. 3, 996– 1004 Robbins, L. 1981. Economics and political economy, American Economic Review Papers and Proceedings , vol. 71, no. 2, 1– 10 Robbins, L. 1998. A History of Economic Thought: The LSE Lectures , in Medema S. G. and Samuels, W. J., Princeton, NJ, Princeton University Press Samuelson, P. A. 1983. Foundations of Economic Analysis , enlarged ed., Cambridge, MA, Harvard University Press Samuelson, W. and Zeckhauser, R. 1988. Status quo bias in decision making, Journal of Risk and Uncertainty , vol. 1, no. 1, 7– 59 Google Scholar CrossRef Search ADS Scoon, R. 1943. Professor Robbins’ definition of economics, Journal of Political Economy , vol. 51, no. 4, 310– 21 Google Scholar CrossRef Search ADS Sen, A. K. 1989. Economic methodology: heterogeneity and relevance, Social Research , vol. 56, no. 2, 299– 329 Siegers, J. J. 1992. Interdisciplinary economics, De Economist , vol. 140, no. 4, 531– 47 Google Scholar CrossRef Search ADS Silberberg, E. 1990. The Structure of Economics: A Mathematical Analysis , 2nd ed., New York, McGraw-Hill Simon, C. P. and Blume, L. 1994. Mathematics for Economists , New York and London, W. W. Norton & Company Souter, R. W. 1933. ‘The Nature and Significance of Economic Science’ in recent discussion, Quarterly Journal of Economics , vol. 47, no. 3, 377– 413 Google Scholar CrossRef Search ADS Stigler, G. J. 1984. Economics—the imperial science?, Scandinavian Journal of Economics , vol. 86, no. 3, 301– 13 Google Scholar CrossRef Search ADS Sudgen, R. 2009. Can economics be founded on ‘Indisputable Facts of Experience’? Lionel Robbins and the pioneers of neoclassical economics, Economica , vol. 76, 857– 72 Google Scholar CrossRef Search ADS Thurow, L. C. 1977. Economics 1977, Daedalus , vol. 106, no. 4 (vol. 2), 79– 94 Toboso, F. 1994. The role the static maximization approach plays in neoclassical analyses, Journal of Interdisciplinary Economics , vol. 5, no. 2, 121– 33 Google Scholar CrossRef Search ADS Údehn, L. 2001. Methodological Individualism: Background, History and Meaning , London and New York, Routledge Wicksteed, P. A. 1946. The Common Sense of Political Economy , in Robbins, L., London, Routledge © The Author(s) 2018. Published by Oxford University Press on behalf of the Cambridge Political Economy Society. All rights reserved.
Cambridge Journal of Economics – Oxford University Press
Published: Feb 24, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera