Seeing the Potential of Realism in Economics

Abstract

In this article, I clarify some of the key concepts and commitments of realist social ontology in economics. To do so, I make use of a recent critique of Lawson’s Reorienting Economics by Mohun and Veneziani. Their article provides a useful foil because responding to their critique allows us to emphasize that realism’s claims are more conditional and less controversial than one might otherwise anticipate. The basic claim is that ontology matters and that explicit recognition and consideration of ontological issues can be beneficial. However, developing a focus on ontology can create problems of interpretation among economists regarding what is being claimed and offered. I discuss some of these with reference to an adaption of Maki’s concept of economics imperialism and also with reference to Mary Morgan’s recent typology of experiments.

Keywords

realist social ontology economics contrast explanation nature’s experiments Lawson economics imperialism imperialism for economics

Introduction

Economics is a subject with great power and influence within the world. It is one we all have an interest in being informed about. What it means to engage in economics is currently in transition (Colander, Holt, and Rosser 2004). Behavioral economics, information-theoretic economics, experimental economics, neuro-economics, and others are proliferating, while what is commonly collectively referred to as neoclassical economics is on the wane (Davis 2006).¹ This transition has been given some impetus by the Global Financial Crisis and its aftermath, which resulted in a great deal of criticism of economics (e.g., Krugman 2009; Palley 2013). Specific questions were asked of the appropriateness of theory forms, such as the efficient market hypothesis, and this was linked by some to the common characteristics of many theories, such as rationality and perfect information (e.g., Stiglitz 2009). The focus on common characteristics has, then, also opened up debate concerning the range of appropriate approaches to the study of economics, as well as related issues such as the formulation, use, abuse, and limitations of models; the significance of complexity; and the limits to forecasting because of problems of uncertainty that may or may not be transposable into matters of probability and prediction (see, for example, Colander 2010).

Many of the issues are not new (for development and substance of some issues, see, for example, Davidson 1991; Milonakis and Fine 2009; Morgan 2012). However, all the issues necessarily include underlying argument concerning philosophy and methodology in economics, and one strand in this debate has been pursued in terms of realist social ontology. The main proponent of realist social ontology in economics has been Tony Lawson. The core commitment of realist social ontology is that one can develop a contingent account (an ontology) of the general characteristics of social reality including its economic domain, and this, then, has some bearing, though does not determine, how one can appropriately engage in economics. This focus on ontology has been developing for in excess of 20 years and, in terms of key concepts used in methodological discussion in economics, now exceeds reference to Kuhn (paradigms), Popper (falsification), and Lakatos (research programs) (Fullbrook 2009, 2).

In the following article, I clarify some of the key concepts and commitments of realist social ontology in economics. To do so, I make use of a recent critique of Lawson’s Reorienting Economics by Mohun and Veneziani (2012). Their article provides a useful foil because responding to their critique allows us to emphasize that realism’s claims are more conditional and less controversial than one might otherwise anticipate. The basic claim is that ontology matters and that explicit recognition and consideration of ontological issues can be beneficial. However, developing a focus on ontology can create problems of interpretation among economists regarding what is being claimed and offered. I discuss some of these with reference to an adaption of Maki’s concept of economics imperialism and also with reference to Mary Morgan’s recent typology of experiments.

The Context and Significance of Realist Social Ontology in Economics

As stated in the introduction, realist social ontology is an attempt to derive a set of general characteristics of social reality, including economics.² The general characteristics are derived from many sources of argument, ranging from an adapted Kantian transcendental question form in philosophy (what must be the case for x?), to immanent critiques of inter-theory disputes regarding some specific problem from which a further solution is proposed (e.g., the problem of individualism vs. collectivism), to inferential claims based on given and possibly disputed empirical evidence. As such, realist social ontology is not “news from nowhere.” The original argument in philosophy distinguishes it from a prioristic rationalism, empiricism, various forms of positivism, and also some varieties of social constructivism. It has subsequently developed as a form of social theory that is engaged with long-standing recognized problems in the social sciences. One of its core tenets is that one cannot reduce matters of ontology to matters of epistemology; the world cannot be reduced to our knowledge claims in its regard without committing an epistemic fallacy (where what is real is reduced to what is known). Knowledge is fallible, and this applies also to realist social ontology as theory. Though the intent is to develop a set of general characteristics, such an ontology cannot be foundational, merely a relevant domain of argument that should not be neglected. The underlying argument for realist social ontology in economics is that the domain has been neglected and that this has been significant for how economics has developed, a point we will return to later.

The general characterization of society developed as ontology is that social reality is an open dynamic system within which different entities can be distinguished—most prominently agents and structures, each with properties, powers, or capacities (e.g., Archer 1995; Lawson 2003, 2012). The innate capacities of persons and the acquired capacities of persons as agents based on their positioning or specified roles affect the nature of events. Similarly, structures of social relations, organizations, and institutional rules affect what is made possible, what is preferred or promoted, and what is denied, restricted, or limited within society.³ People draw on or act within existing structures, and moment-to-moment social reality is a series of events based on the sequenced interaction of agency and structure. As such, social reality is held to be causal (there is the effective causation of intention and the formal causation of context) but not deterministic. It is non-determinist for a variety of reasons: because persons are not merely over-socialized homogeneous agents, because choice is operative, and because complexity also applies—the multiplicity of causal connections and possibilities means that events are unlikely to be strictly regular in terms of a given single cause producing a given particular outcome, and furthermore, events cannot be reduced to intention, aspects are shaped by structure, and some aspects are likely to be unintended.

A key claim of realists is that there is a particular knowledge focus in economics on identifying regularity in the form of constant conjunctions or universal law-like associations of the form “whenever α then β.” The conjunction can be between a single α and β, or a fixed and well-defined group of α and then β or a fixed and well-defined range of αs to β (a definite probability set). For realist social ontology, all three versions of the constant conjunction are based on fixed relations (quantities, proportions, impacts, likelihoods). According to realists and based on the ontology, a knowledge focus on seeking constant conjunctions or universally applicable law-like associations of the type “whenever α then β” is then likely to be of limited use. The term “limited” is used because the reasons cover a range, and most of them are not simple refutations of a constant conjunction approach, but rather statements of the potential problems of emphasizing constant conjunctions as the key focus, development, or test of theory or empirical work in economics. For example,

In the most primitive sense, seeking constant conjunctions can focus work on events rather than the underlying powers or properties that then give rise to events through interactions. As such, the focus can fail to explore (rather than simply describe some small element of) causation and fail to capture the way reality has “depth” because not everything that is operative or can be operative is always expressed in events; such a social science focus would be incomplete.

Seeking constant conjunctions may also be a focus that is in tension with a general theory position; an approach may articulate dynamic open systems in terms of its general theory position and then focus down on exploring particular constant conjunctions. The question, then, becomes, “What does seeking constant conjunctions actually tell one about a dynamic open system?”

Seeking constant conjunctions may then also incrementally shape the way theory and models are constructed; as one focuses on constant conjunctions, one may be prone to conflate stability or relative repetition in events with definite fixed causal relations. This creates a presumption of closure in a system. However, a social system is never entirely closed in this sense since it has both internal and external potentials for change. It is in this weak sense that realist social theory claims that systems are open, the point is made to place a question mark against constant conjunctions rather than to suggest that systems are simply “open,” there are degrees of openness but also analytically a limit since a fully open system is by definition not a system.

Insofar as the conflation between stability and constant conjunctions occurs, then an adverse focus of investigation may be created since stable outcomes may disguise relative shifting or complexity in the underlying entities and mixes of causation. The claim here is that the reason for stability is not a given constant conjunction per se but rather the patterning in events provided by a set of more complex social relations that are expressed in interactions as events. The inference drawn in realist accounts is that if one focuses on seeking the conjunction, one is then also motivated to seek to isolate the conjunction and thus separate it from many of the aspects that are actually responsible for the stability and its potential variations and breakdown.⁴ Thereafter, in reconstructing around a focus on the conjunction, one is motivated to build theories or models that express the conjunction most perfectly and so can be tempted to not just simplify the real world of economy but replace some of its elements with assumptions that are essentially fictions (but are phrased as “abstractions”). Theory and model building then become less about focused simplification and more about, perhaps inadvertently, shedding potentially significant aspects of reality, particularly the contingency and potential instability of social reality. Some of the problems this creates become clear when significant change manifests (often as crisis). The constant conjunction focus may have resulted in models that desensitize the economist to, or put aside or marginalize, worst cases.

Once seeking constant conjunctions becomes a focus, this can lead in terms of the subsequent sociology of knowledge to the cumulative dominance of theories and models that focus on constant conjunctions. Models provide elegance and parsimony and insofar as they are expressed in mathematical form (are formalized) seem to conform to a natural science discourse. This provides a degree of legitimacy to them (they are “rigorous”), while the expression of simple law-like relations enables them to have a readily translatable policy facilitating function, giving them social power. They are also a form of knowledge that can be developed with limited actual empirical work since the central problems of models and model building often concern how to resolve problems of mathematical expression that arise from model building and, thus, how to resolve issues of mathematical proof or how to develop innovative new proofs.⁵ As such, theories and models that focus on constant conjunctions are ones that contain their own internal developmental concerns and can quickly proliferate based on those concerns. Moreover, since the main point of interest is the expression of the constant conjunction, real-world relevance may cease to be the primary issue. So, a focus on constant conjunctions can lead to both an adverse focus within theory and model building and a further proliferation of model building.

There are, of course, other historical reasons for the expansion of economics as it is currently practiced, but realist social ontology as a critique tends to focus on constant conjunctions in particular because the intended point is to highlight the tension between the perceived attractive characteristics of model building and purported limitations. From the point of view of realist social ontology, a focus on constant conjunctions tends to make it more difficult to do work that has a fuller and deeper appreciation of the economy as an aspect of social reality. The point being made is not so much that work on constant conjunctions cannot be done, since manifestly it has been done, and that it can have some value, impact, and insight. We are steeped in a world influenced by and shaped according to economic theory and model building—everywhere from central banks to derivatives markets. Rather, the point is that a focus on constant conjunctions within academic discourse is not always and everywhere the most effective approach and ought not to dominate.⁶ I make these initial points because they provide a slightly different context for considering Lawson’s work in particular than that provided previously by Mohun and Veneziani (2012).

Mohun and Veneziani on Lawson’s Reorienting Economics

Mohun and Veneziani’s argument specifically addresses Lawson’s Reorienting Economics. Its focus is on Lawson’s critique of mathematical-deductive reasoning and methods. According to Mohun and Veneziani, for Lawson, the dominance of this approach, effectively a focus on constant conjunctions, leads to model building whose formalism is inappropriate because it provides limited insight into a social reality characterized by open dynamic systems. They then sub-categorize Lawson’s position into a strong thesis and a weak thesis. In Mohun and Veneziani’s strong thesis, mathematical-deductive reasoning and methods are (almost) never appropriate, and in the weak thesis, it is only their excessive use that is inappropriate. On the basis of their reading of Reorienting Economics, Mohun and Veneziani focus on the strong thesis, which they take to be Lawson’s implicit position, his caveats notwithstanding. They, then, make several points to refute the strong thesis.

First, they claim that Lawson’s (2012) concept of closure is overly restrictive. The association of closure with constant conjunctions of the type “whenever α then β’” creates subsequent problems (pp. 130-31). If actual systems are open, it may still follow that some aspects of systems are for some period stable such that one can observe “whenever α then β” and so models that focus on such constant conjunctions may be appropriate. Moreover, incidents of failure of models that focus on “whenever α then β” are not necessarily indicative of the need to reject such models since this creates the possibility that one failure of an observed regularity dispenses with an otherwise relevant model (a classic critique of falsification as a tenet). Both these points seem sound insofar as one accepts that Lawson’s argument should be judged by the strong thesis.

However, if one were to read Lawson more sympathetically, based on the points in the previous section, then an alternative interpretation can be put forward. The weak thesis is not simply a set of caveats to disguise a preference for the strong thesis, and one might argue Mohun and Veneziani’s point may actually support Lawson on the basis of the weak thesis. In economics, the realist approach to constant conjunctions is a conditional and cumulative argument based on the multiple reasons why a focus on those constant conjunctions can be inappropriate. Clearly, there can be periods of significant relative regularity, association, and so forth between an α and β in reality. It is because this can be the case that the concept of law-like regularity has had such traction since the time of Hume. This is a particular attraction for economics as a social science because its subject domain is one that explores quantities and, in various ways, the measure of changing quantities of phenomena (e.g., price, consumption, inflation, unemployment, gross domestic product [GDP], etc.).

However, Lawson’s point is that the emphasis on constant conjunctions comes with various costs. The very problem of falsifiability only exists if one begins from a constant conjunction focus. What Lawson is arguing for is a form of social ontology in economics that can encompass both stability and change when it is then expressed in theory and empirical investigation. Falsifiability is not a problem for an open systems social ontology, it is simply a non-issue because one is not committed to constant conjunctions, but rather to exploring the multiple and contingent causation expressed in events based on the powers of entities. If one begins from models that focus on constant conjunctions, then the problem of falsifiability becomes relevant (how can one know the model is valid, at what point does it cease to be relevant?). The broader point that can be drawn from Lawson is that the constant conjunction focus can create a presupposition that constant conjunctions are primarily what are significant. This is not so with an open systems approach because it does not presuppose constant conjunctions but is not inconsistent with periods of relative stability for some aspects of social reality in given times and places. An open systems approach does not pre-judge reality in quite the sense a constant conjunction approach does—regularities may emerge, and it is in this sense that one cannot discount the potential of mathematical-deductive models and methods. They may express this regularity. The point is that they are limited because they do not easily encompass the absence of such regularity. Nicholas Stern, for example, makes this point in his recent and remarkably candid assessment of the consequences of the dominance of Integrated Assessment Models on both the estimation of climate change’s economic impacts and the subsequent policy issues (Stern 2013). The inference drawn by Stern is that they can be used but that perhaps there are better ways forward and also a need for more diversity in ways forward—they have dominated too much.⁷

This brings us to Mohun and Veneziani’s second point. They claim that Lawson’s (2012) position (expressed as the strong thesis) is based primarily on a characterization of econometrics and that it may be less applicable elsewhere (pp. 131-133). As Boland notes in a rather different context, “economics is dominated by model builders—but not all models involve econometrics” (Boland 2010, 533). Boland’s central point is that economists (as model builders in general) are not actually interested in causation and prefer to restrict themselves to “whether their model can be used to mathematically determine the values of endogenous variables” (Boland 2010, 536). Here, there is surely some significant overlap between Mohun and Veneziani and Lawson since each would argue that economists ought to be more interested in causation as a social process in a way that this approach to modeling does not capture (whatever other merits there may be in this approach). This is actually implicit in Mohun and Veneziani’s (2012) subsequent point of criticism. The criticism follows Hodgson to make the point that a formal model may actually serve some use since it may identify a regularity as a potential causal mechanism that, then, can be integrated to form part of an understanding of a dynamic open system (Mohun and Veneziani 2012, 132). The inference is that Lawson’s critique of mathematical-deductive modeling rejects this. However, the validity of the point is predicated on the strong thesis. Again, if one places Lawson’s work within the weak thesis, the argument looks rather different. It is surely because there are limits to the usefulness of modeling that one is motivated to then place the insights in terms of a contribution to the understanding of causal mechanisms in a dynamic open system. Hodgson, Lawson, and many others (e.g., Dow 2007; Milonakis and Fine 2009) may differ over some aspects of each of their critiques of the mainstream in economics—but all are committed to the need for a critique of that mainstream because of its underlying forms (variously termed its ontology, metaphysic, methodological elements, technical apparatus, etc.). Thereafter, the issue is one of the burden of proof, that is, what does the model actually add to one’s understanding of a dynamic open system?

Mohun and Veneziani (2012) treat this proof issue with care. The point they want to emphasize is that reference to econometrics creates too much of a focus on the issue of prediction in economics, and thus on instances of explanatory failure of only some types of models that are constructed for predictive purposes. They make the alternative point that there are many insights, including from mainstream economics, that are made through mathematical-deductive reasoning and formal models and that perhaps those insights would not have been made without the formal deductive reasoning. They provide 10 illustrations. For example:

3. The conditions [required to establish] the uniqueness of equilibrium are extremely strong. In general, therefore, equilibrium cannot be used for predictive purposes and comparative statics analysis cannot be performed. (Mohun and Veneziani 2012, 133)

This seems a curious example for various reasons. It is a negation, and its insight is not about reality but about the absence of relevance for reality of a formal product of deductive reasoning. Its significance only arises if one assumes that a formal model of equilibrium is worth producing. The negation that then follows is a formal recognition of a problem of formalism. This, of course, has a history, as Mohun and Veneziani are clearly aware, Walras placed a question mark against proofs of the uniqueness of equilibrium over a century ago. This did not result in Walras abandoning general equilibrium, and for later economists, the problem became a mathematical puzzle to be pursued. Insofar as there have been attempts to overcome the problem, they have been about demonstration proofs to resolve a problem of construction for and of the model. These are not attempts to focus directly on a real-world issue (nothing is being proved about the real world in a positive sense—quite the reverse insofar as fixed point theorems are adopted from pure mathematics, one must assume the existence of an equilibrium and then build the model from this basis, that is, the equilibrium does not emerge naturally from the mathematics; see Velupillai 2007). The example, therefore, actually illustrates precisely the problem of how models can gain their own internal dynamics that make issues of reality less of a focus, and so supports a basic point (point 5) we made in the first section. Furthermore, the very fact that the point regarding statics is still current and worth stating in the twenty-first century, despite long-standing and devastating critiques of general and partial equilibrium and of statics (e.g., Kaldor 1972; Sraffa 1926), speaks volumes regarding the issue of the dominance of particular problematics in mainstream economics.⁸ Consider also Mohun and Veneziani’s (2012) final illustration:

10. Asymmetric information between agents may destroy markets. (p. 133)

This claim has been part of the general rollback of models that assume perfect information and a pure rational calculation in mainstream economics (for the classic articulation, see Muth 1961). But the information-theoretic and behavioral revolution had to fight hard to get such insight accepted and published in the 1970s, as one of the main proponents and subsequent Nobel Prize–Winner, George Akerlof (2003) notes. Moreover, there is a great deal of critique concerning what the actual gain is from these developments and others, such as new institutionalism (compare, for example, Sassower 2010 and McCloskey 2010); and one can ask the relatively basic question of just how counterintuitive and, thus, genuinely insightful the specific claim (“10.”) is. Is there anything beyond common sense in the argument that participants with different information in a market may be operating according to different expectations and even (if one extends what information is) different informal rules, and this may be detrimental to that market, either through deliberate subversion and exploitation or through unintended consequences? It does not require set theory, symbolic logic, or any form of involved algebra, or even tightly specified and consistently stated axioms and assumptions to come to the same conclusion. Nor would it require a syllogism. Any observation of a real market anywhere in the world would bring to the fore the potential for such an outcome as the outlier of contingency and instability in a complex system that is open and dynamic. The fact that the claim has been heralded as a great leap forward for economics serves mainly to indicate how economic theorization can become divorced from readily recognizable aspects of reality once simplification serves the function of tractability. Again this speaks volumes for the particular problematics of the mainstream. The example does not, however, unambiguously and persuasively illustrate that deductive reasoning in a formal context has proved of some additional value (except as an antidote to a problem created by a previous formal model of perfect information).

The Potential of Realism in Economics?

What I want to suggest here is that the greater force of Mohun and Veneziani’s critique is predicated on the claim that Lawson is committed to their strong thesis. I would suggest the weak thesis seems more appropriate and consistent with Lawson’s overall argument and with realist social ontology as an approach in general. I do not think, however, that Mohun and Veneziani are necessarily traducing Lawson. They explicitly state that Lawson argues carefully and that each claim is made in a conditional way. Their reading, however, requires them to have interpreted the conditional form of the argument as camouflage rather than actually integral to the total argument. This reading is possible because they have begun from statements, in terms of the logic of argument, about mathematical-deductive reasoning in connection with models and formalism. They are, therefore, asking a question that cuts across the original intent of Lawson’s approach and of the general tenor of realist social ontology. Their critique seems to be structured around, “Are mathematical-deductive reasoning and formalism possible?” The general tenor of the realist position, however, might better be contextualized through, “Are mathematical-deductive reasoning and formalism always and everywhere appropriate, and are they the best economists have to offer?” It is only from the former that an argument concerning consistency and compatibility can be made in the way that Mohun and Veneziani pursue. Yet this is to make the argument for ontology, and the array of underlying problem in economics of constant conjunctions, residual. However, the realist argument is that ontology needs to be explicitly considered precisely to address the problems that Mohun and Veneziani are addressing—different ways to explore dynamic open systems in economics.

The basic strategy of Mohun and Veneziani is to show that mathematical-deductive reasoning is potentially compatible with dynamic open systems (and so Lawson is inconsistent). But the implication is, then, that the ontology matters—the standard remains dynamic open systems, since it is this position that Mohun and Veneziani argue one can be integrating the work into. If dynamic opens systems remain the standard, then argument in and regarding an ontology that articulates this position must surely be one of significance. And again, if one reads Lawson through the weak thesis, this point is supported rather than contradicted—ontology matters. One can consider this, for example, in terms of Margaret Schabas’s (2009) work on the concept of the “economy” and what this suggests about the nature of mainstream economics. Schabas argues that mainstream economics does not seem to have an adequate recognition of and consistent set of concepts of emergent properties, unintended consequences of activity, and of economic phenomena as institutional facts (following Searle). Her underlying point is that economists have not defined their subject in terms of “the economy” and that issues concerning the term highlight “ambiguity and confusion” in economic discourse regarding what an economy is and how it should be appropriately conceptualized and explored. These are all basic issues that realist social ontology seeks to bring to the fore under the heading of ontology. The potential in doing so is that the consistency and implications of theory and method are placed in a wider context of the underlying implications for the nature of social reality.

The potential, however, is one that invites a degree of unease in economics including among heterodox economists, and this may contextualize Mohun and Veneziani’s concerns. One can consider this unease in terms of an adaption of Maki’s (2009) approach to economics imperialism. Maki makes the point that economics imperialism (the expansion of economics theorization to territories currently occupied by other disciplines) involves several forms of expansion, one of which may be ontological unification:

Ontological unification is a matter of re-describing large classes of apparently independent phenomena as forms or manifestations of a common system of entities, causes and mechanisms. It is based on the representational capacities of theories in depicting such underlying systems. Explanations are construed as descriptions of the order of things, or goings on, in the world. Theories are regarded as purportedly true pictures of the simplest mechanisms and processes of the world’s working; phenomena are regarded as manifestations thereof. (Maki 2009, 364)

For its adherents, the expansion of economics to new territories is a signal of its success as a scientific research program. To others, it is a sign of the obtrusive social power of economics as a discipline able to extend (and sometimes impose) inappropriate approaches because of the generality and essential emptiness of those approaches (e.g., the exploration of behavior beyond the traditional focus on markets and in terms of rational choice in situations of trade-offs). The uneasiness regarding realist social ontology in economics is perhaps rooted in a reversal of this critical stance.⁹ Realist social ontology may be being read as imperialism for economics rather than economics imperialism (it is all about ontology, and the realist ontology is imposed as a standard to which economists must conform).¹⁰ Concomitantly, Mohun and Veneziani make the point that the degree of generality of ontology, and the focus on philosophical-social theory analysis leaves the further substantive elements of specific economic theory, as well as the socio-historical emergence of economic theories within capitalism, relatively under-emphasized.

Again, this line of critique is not without plausibility. Many economists will recognize it (and Fine, for example, has consistently highlighted it). A reasonable realist response, however, would not reject the critique but rather treat it as indicative of an adverse potential to be avoided through critical engagement. Realist ontology is not an object to be imposed but a theory to be constructively engaged. To claim that ontology matters is not to assert that it is the only thing that matters or that a given ontology is simply definitive (there are, not least, also other varieties of realist social ontology).¹¹ However, Lawson’s general approach may be considered well defended and accepted in different ways to different degrees—recall that Mohun and Veneziani’s position uses dynamic open systems as a standard. Thereafter, its acceptance ought not to be (and has not been as Mohun and Veneziani’s account indicates) uncritical. Irrespective of that line of critique, nothing about Lawson’s position entails the argument that the significance of ontology is reducible to the validity of a single theory of ontology. This would be a conflation of a domain of argument with an argument within a domain. Moreover, it would entail a contradiction in terms of realist social ontology, since a consistent argument for ontology is anti-foundational and fallibilist in regard of any theory of ontology, and, as such, a theory of ontology must also be conditional and corrigible. The main claim of realist social ontology in regard of economics is that ontological argument and analysis should not be neglected and not that a realist social ontology should be passively accepted. In this regard, the main critical claim of realist social ontology in economics is that the ontological implications of constant conjunctions within the development of theory and empirical investigation ought not to be ignored. This does not, then, invalidate a further focus on substantive issues of theory and does not entail that there are no other historical reasons worth exploring that account for the state of economics—in the latter case, an entailment would be an inconsistency since economics is not knowledge without context, but rather a product of multiple processes like any other aspect of social reality (for political economy, see Brown 2007; Fleetwood 2012; for the link between ontology and technical apparatus, see Fine 2006).¹²

Again, realist social ontology seems more palatable when phrased in a weak sense in terms of limits. It may resonate with an adaption of Maki’s concept of economics imperialism (and perhaps neglect many of Fine and Milonakis’s 2009 concerns in their approach to economic imperialism) and invite concerns based on ontological unification, but the unease this may invoke ought to be tempered by a separation of claims—the warrantability of realist social ontology and, recalling the concerns of Schabas, the relevance of ontological argument as one way to constructively consider the nature of economics. This is a consideration that tends to lend credence to the point that the weak thesis form of Lawson’s argument is not camouflage but rather a consistent consequence of the ontology.

Yet one could still assume that Lawson’s approach is vulnerable to a tacit strong thesis reading of the kind pursued by Mohun and Veneziani. This is for two reasons. First, it is one thing to acknowledge limits and to state that there may be uses for mathematical-deductive reasoning, model building, and formalism, but an acknowledgment is not an argument for the positive. By not periodically providing a positive demonstration of the uses and then specific limits based on those uses, Lawson’s position does become vulnerable to skeptical inferences regarding the acknowledgment. This is particularly so since Lawson is a trained mathematician (rather than originally an economist) and teaches econometrics—and so surely has the skills to provide the demonstration. Lawson, of course, might respond that the positive claims and weight of output are all on the side of model building, and so forth, and so it is not up to him to provide this work; it is up to others to demonstrate its value. Still, one might consider this a strategic dilemma, since there may be gains to providing the positive case, in terms of avoiding interpretations of the kind Mohun and Veneziani pursue. Problems of inference in terms of the meaning frame of concepts have been an ongoing issue for realist social ontology in general, as Kaidesoja (2013) has noted in various contexts.

Second, Mohun and Veneziani focus on Lawson’s Reorienting Economics. This work is more concerned with (though by no means is restricted to) issues posed in terms of mathematical-deductive reasoning, with no positive demonstration provided. If one takes Reorienting Economics and more recent work in isolation, then one might infer that Lawson was tacitly committed to the strong thesis. However, to do so would require one to translate the development of his work into a repudiation of its own context of development. Nothing in Lawson’s later work is phrased as a rejection of his former commitments, arguments, acknowledgments, and statements. Reorienting Economics is perhaps more appropriately read as an elaboration where the previous work is taken as a given. In Economics & Reality, Lawson (1997) engaged in an immanent critique of economics theory and method. Reorienting Economics continues this but also does more to flesh out a general social ontology for economics. There is a transition here, but one might read this differently to Mohun and Veneziani. The transition is toward doing more to develop alternative possible approaches to knowledge acquisition in economics. This can be considered consistent with the weak thesis. If mathematical-deductive reasoning, modeling, and formalism are not always and everywhere appropriate and are not the best economics has to offer, then the need for alternatives necessarily follows. To pursue them is not indicative of a commitment to a strong thesis position but rather one possible constructive way to add diversity in the face of the recognized problem of dominance. The focus, then, is about expressing potential.

The Realist Dilemma, Economics Imperialism, and Contrast Explanation

Realist social ontology is not intended as a substitute for schools of economic thought and the particular theory that adherents of schools provide. There is no realist economic theory in this sense because there is no realist school of economics. The focus on ontology emerged as a critical discourse of economics. Its focus has been on the ontological underpinnings of various schools and methodologies. As such, it has claimed an “under-laboring” function. This is another way of saying that ontology matters and that ontological argument should not be neglected—the warrant for this claim is that all knowledge claims have an implicit ontology (how the world is), and it is preferable that the ontology be explicit since this requires the implications of the ontology to be clearly considered and defended. One might argue that Mohun and Veneziani are turning this point back on realist social ontology, and in a certain sense, this is constructive because it requires some clarification of the terms of the argument along the lines we have pursued and this, as a by-product, illustrates our previous point concerning engagement.

However, the unease that seems to motivate Mohun and Veneziani’s position still remains. It remains because, over time, there is a tendency as a position matures for the terms of engagement to shift. A critical discourse typically begins as an immanent critique. Lawson, as noted, provides an immanent critique of the knowledge focus of economics in Economics & Reality. Concomitantly, he develops an argument that seeks to highlight the shared ontological commitments of heterodox economics and some of the potential inconsistencies within heterodoxy. He claims heterodox economics tends to share a dynamic open systems approach, but with particular research agenda and foci for different schools of thought—classical and post Keynesians, Marxists, feminists, old institutionalists, and so forth (see, for example, Lawson 2006). The point is an invitation to engage issues of ontology within the particular research agendas and focus of schools and to consider the difference between the fundamental commitments of these schools and mainstream economists. Ontology is offered as a domain of argument in terms of which heterodox economists can recognize the common basis to their diversity—providing a medium through which a productive pluralism can be pursued as a research community. However, the issue of a constant conjunction approach and of the use of formal models necessarily arises because of the claims made for the specific ontological position of realism. In what sense is a constant conjunction approach compatible with the concerns of heterodoxy? From this perspective, the context of Mohun and Veneziani’s line of critique is explicable. There remains an unease that realist ontology is an imperialism for economics, and a challenge seems to have been issued to defend mathematical-deductive reasoning. However, as we have suggested, an invitation to engage is not imperialism per se, and it cuts across the intent of realist social ontology to defend mathematical-deductive approaches on the basis of whether they are possible rather than whether they are always and everywhere appropriate and ought to dominate.

Yet Mohun and Veneziani’s response might still be considered motivated in a reasonable way by a reading of the transition from Economics & Reality to Reorienting Economics. If one considers Lawson’s work as a whole, then it can be read as cumulative, but there is necessarily going to be a transition in concerns. One cannot simply continue to repeat the same argument in the same context through the same immanent critique. As such, the later work does more to develop the general ontology of dynamic open systems and also begins to develop alternative approaches to knowledge acquisition for economics. In doing both, however, the weight of material in Reorienting Economics compounds the existing problems of inference. First, by developing more focus on the realist account of ontology (aspects of dynamic open systems), it feeds unease regarding imperialism since the approach for economics becomes more substantive and thus more easily challenged in terms of the claim to be under-laboring, despite that it is still not articulated as a substitute for particular schools of thought. Second, as approaches to knowledge acquisition are explored, the terms of those approaches invite the counter-challenge: “How do these stand in terms of mathematical-deductive reasoning, formal modeling, and so forth?”

The dilemma for Lawson and anyone working on realist social ontology in economics is that one cannot avoid substantively developing the position but that, in doing so, the development invites critique because of that transition. The main proposed approach to knowledge acquisition in Reorienting Economics is contrast explanation (Lawson 2003, 86-109). Contrast explanation is posed in the first instance as a question form following a particular orientation to observation. One observes an unexpected or unusual outcome—and this invites the question “why x rather than y?” It is an observed contrast that motivates investigation, and this shapes the approach to what is the significant aspect of some phenomena in terms of subsequent research. Mohun and Veneziani (2012) respond to this approach by extending the general line of argument regarding the possibility of mathematical-deductive reasoning. They make the point that contrast explanation can be conducted along formal lines and, thus, is not necessarily an alternative to mathematical-deductive reasoning (Mohun and Veneziani 2012, 138-42).

Again, this is a perfectly reasonable point in terms of the logic of argument they pursue. But the intention of introducing contrast explanation does not hinge on whether, in fact, one can reconstruct a contrast explanation question form within some expression of formalism. Rather the point of contrast explanation seems to be to orient research and investigation on what the researcher deems unexpected or surprising. The contrast is one of anticipation rather than of fact per se. As such, the observation may be of a relative stability—one is surprised that x persisted rather than a change to y occurred. Alternatively, the observation may be of some sudden break—one is struck by a sudden change in outcomes or events (including quantified expressions of social phenomena). Clearly, then, there can be no presumption of a specific designation of what an x or a y is and what methods will be applied to the investigation of x and y. There is also implicitly a strong reliance on the cumulative expertise of the observer as an economist since it is this that gives significance to what events or outcomes are attended to and how they are attended to.

For Mohun and Veneziani, Lawson’s argument for contrast explanation is broad, subjective, and under-specified. In these terms, it is, therefore, deficient as an approach. However, rather than deficient, one might more sympathetically describe it as intentionally open to match the variety of possibilities of an open environment. As such, it is not empty merely because it is open. It is broad because it is no more or less than a generalized expression of a then specifiable form, analogous perhaps to Searle’s definition of a constitutive rule. It is subjective because how it is specified relies on context and on the way the economist attends to that context. There are multiple ways a contrast explanation may be elaborated, and the framework the economist is working with necessarily becomes a matter of significance because it orients inquiry (e.g., Morgan, J. 2013). So, Lawson’s intent is clearly informed here by his realist social ontology. He intends contrast explanation to be taken up as a way of focusing research on the variability of events or the variability underlying otherwise stable events. Contrast explanation seems to be no more or less than a means to encourage a link between focusing research and explicitly considering the significance of ontological frameworks.

Since contrast explanation is about orienting or sensitizing the observer as an economist rather than specifying a definite method, it is unsurprising that the general form can be reconstructed in a formal manner. To do so does not, then, introduce or highlight an inconsistency because the focus of the point of critique is at the level of method rather than how the economist orients investigation. Contrast explanation is being put forward as an alternative holistic way to initiate investigation. It does not have the necessary form or content to preclude a reconstruction along formal lines in terms of how the investigation is then stated or pursued. Whether one wanted to state or pursue investigation along lines of formalism would then be a further issue, and this would then invite a consideration of the significance of ontological frameworks—what has been the value of the deductive form or the constructed model? This is not an a priori issue of whether in fact one can construct a contrast explanation along such lines but an a posteriori issue of what was gained by doing so and what the potential costs have been—this again would draw one back to the potential limits of a constant conjunction focus, this time with a link to specific empirical dispute.

Contrast Explanation and Mary Morgan’s Nature’s Experiments

Taking all the points made so far into account, there can be no simple refutation of Mohun and Veneziani’s case. The response can only be that the issues can be looked at differently. When considered sympathetically, realist social ontology is more consistent and reasonable than one might infer. It is the potential in explicitly considering ontology that is the enduring point. One might consider this in terms of Mary Morgan’s (2013) recent article on forms of experiment in the social sciences. Morgan develops a series of concepts or categories of experiment that go beyond the way the term is usually understood in realist discourse.¹³ In particular, Morgan distinguishes between nature’s or society’s experiments and natural or social experiments. Natural or social experiments are situations where the (social) scientist tries to retro-fit events into the traditional form of field or randomized trial experiments within society. Nature’s or society’s experiments, meanwhile, are ones where the conditions themselves, rather than the agency of the (social) scientist, throw up the potential for investigation because they have some characteristic that is “as if” experimental intervention has occurred within an otherwise open environment:

Nature’s experiments may be defined as natural situations holding the characteristics of experiments . . . [But one must consider also that] First of all nature is in a constant state of flux and experimentation, so why are some events singled out as Nature’s experiments? Second, it is the scientist who, by practical experience, chooses certain places, situations, or events in the field, and conceives of them as offering Nature’s experiments. Both points are equally valid for the social world. (Morgan, M. 2013, 343)

Clearly, there is a strong resonance with Lawson’s thinking in regard of contrast explanation. One is seeking some point of purchase within an otherwise open environment—some focus of significance based on the characteristics that might be observed. Furthermore, what is attended to hinges also on the way the social scientist is sensitized. Morgan, in particular, identifies four possible circumstances where an “as if” might provide the potential for an investigation, of which the third is held to be the most likely/common:

III. Events that stand out in some way as unusual and yet take place in very stable situations.

Many, perhaps most, events in the natural world and social world occur with lots of other events happening around them, but some individual events happen in short time periods in specific places where it is reasonable to suppose that the environmental features are very stable, and the other causal factors (that might normally vary over space or time) are also rather stable. (Morgan, M. 2013, 345-46)

The phrasing here brings to mind Lawson’s (2003) first example and, thus, initial development of contrast explanation—variations in crop yields within a field (p. 88). The situation is one where some surprising change occurs within what seems to be an otherwise stable situation for comparison. So again, one can point to a resonance between Lawson’s and Morgan’s thinking. One must, of course, then acknowledge that resonance between contrast explanation and Morgan’s concept of nature’s or society’s experiments is not a knockdown argument for contrast explanation. It is, however, an indication that Lawson’s thinking is not aberrational. He is thinking about a fundamental problem in a way that others also are. That thinking is not simply empty because it is initially general and acknowledges subjectivity—rather it is oriented to the open nature of the environment in which investigation occurs and is also oriented to expressing his own approach to ontology (which may or may not be an adequate one, since this is a different issue than whether, in fact, considering the problem in terms of ontology is appropriate).

Note also that Morgan’s typology of nature’s or society’s experiments concern different ways in which what occurs may reproduce some of the conditions associated with controlled experiment. As such, circumstance is providing a situation analogous to experiment. Recall that Mohun and Veneziani’s first criticism of what they take to be Lawson’s position was that if actual systems are open, it may still follow that some aspects of systems are for some period stable such that one can observe “whenever α then β,” and so models that focus on such constant conjunctions may be appropriate. One might then argue that Morgan’s approach to nature’s and society’s experiments also supports Mohun and Veneziani’s claim. To a certain degree, it does—it becomes a contestable empirical issue as to whether a focus on conjunction provides some valuable insight in the given case. This, however, does not then undermine Lawson’s position. Rather it reaffirms that it is important to consider the context in which any apparent relative regularity might arise. As Morgan also notes,

Most of the sites that might offer Nature’s or Society’s experiments for study do not fulfill these [the typology’s intrinsic conditions] ideal requirements. But that most fall short is equally true of other kinds of experiments—those in the laboratory or field also fall short of their ideals in practical ways to a greater or lesser degree. Being less than perfect in any of these kinds of experiments may create problems for making valid inferences about them but does not seriously eat into the validity of the category outlined here of Society’s or Nature’s experiments. (Morgan, M. 2013, 347)

Here, one is acknowledging, among other things, that insofar as “most fall short” reality is not reducible to the regularities that may arise, and as such, the matter of where it is appropriate to use models and approaches that focus on such regularities must be one that is sensitized to the possibility that they are inapplicable. Thus, any warrant for the use of such approaches requires an economist whose expertise extends to understanding the limits of their relevance and the grounds of those limits, and this is an ontological matter. So, insofar as the possibility of nature’s or society’s experiments resonates with contrast explanation, it does so, logically speaking, in a way that places the possibility of a constant conjunction focus within the context of ontology and places ontology as a significant issue for the way the economist is sensitized as a researcher. As such, Morgan’s position provides some credence for Mohun and Veneziani’s claim that a constant conjunction focus may be relevant but does so without then supporting the way ontology becomes a residual issue in Mohun and Veneziani’s critique. The point, then, also lends credence to the way that Lawson seeks to pursue contrast explanation as a way to address the grounds of both relative stability and irregularity.

Conclusion

The global financial crisis has provided some impetus for dialogue and debate in regard of how economics should develop as a social science. In this article, we have made the case that realist social ontology has more to offer than one might infer from Mohun and Veneziani’s position. Ontology matters, and a focus on ontology does not commit one to forms of exclusion that suppress debate. On the contrary, it opens up a significant domain of argument but does so based on recognition of limits (as this article itself demonstrates). As such, the realist position is one of potential. It is worth reiterating a point made in the introduction here. As Fullbrook (2009) notes, use of the term “ontology” in economics now exceeds reference to Kuhn (paradigms), Popper (falsification), and Lakatos (research programs) (p. 2). Ontology has become a significant critical discourse in economics. As Mohun and Veneziani’s critique also indicates, it has not thereby become uncontroversial—an aspect of normal science. Whether it does become uncontroversial may well hinge on how realists respond to the concerns of skeptics like Mohun and Veneziani and how constructive dialogue is pursued.¹⁴

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

1

Though there is also debate concerning what exactly neoclassical economics is (e.g., Lawson 2013) and what it means to move beyond it, for example, whether the information-theoretic turn constitutes, a new, an improved, and/or the current most adequate way of developing economics (see, for example, Fine and Milonakis 2009; Fullbrook 2009; ).

2

For examples of those who work on realist social ontology in economics, see Fleetwood (1999) and Lewis (2004), and for a broader range .

3

Including the creation of emergent powers, see Archer (1995) and Elder-Vass (2010), and for specific properties and artifacts of significance to economics such as money, see .

4

This may have a variety of consequences. For example, one may default to a form of social atomism or become embroiled in a potential contradiction between methodological individualism and the recognition of emergent properties.

5

And may also promote the form of decontextualized and alienated research that consists in little more than accessing a database, such as those provided by the Office for National Statistics in the United Kingdom.

6

This kind of claim, however, provokes the counter-question by those well versed in philosophy: what criterion can be used to know that it would be appropriate to apply a constant conjunction approach? The immediate response offered by Lawson and others working in his Cambridge tradition is that one is locating a constant conjunction approach as one manifest in a given method and one is refusing it the more dominant position of orienting the inquiry as a form of methodology or contextualizing concept of appropriate social ontology (metaphysics, etc.). Whether, then, a constant conjunction method is appropriate becomes a matter of the context-specific use of the given method (e.g., a branch of econometric application). This is something that the economist develops a feel for based on context and expertise and also based on a community of scholars’ critical appraisal over time of the explanatory success of the use of the method (there is a contingent empirical testing that accumulates to a context for judgmental rationality a posteriori). As such, according to the argument, there is no definite formal epistemological criterion to be applied, rather there is an argument for the refusal of such an a priori in favor of a continuing consideration regarding ongoing judgment of what is appropriate, and this flows from the open systems ontology (so there is also no—and seemingly can be no—formal criteria regarding the point of focus of investigation to develop explanations). The implication is that realism of this variety also has a relativistic subtext and elasticity based on the need for critical agreement and judgment without strict criteria. One recognizes that reality makes a difference and that one can fail to explain something or can succeed in explaining something more adequately. While this is a reasonable argument, it is also dissatisfying from the point of view of epistemology, as realist epistemologists such as Alston have noted (). It creates issues not only in terms of Chisholm’s (1966, 1982, 1996) general work on criteria but also ontology. See later discussion of contrast explanation and then also Mary Morgan’s work.

7

A similar admission can be found regarding the other major field in which economics has had significant and adverse influence in recent years: finance. According to Chapter 4 “Why Did the IMF Fail to Give Clear Warning?” “Perhaps more worrisome was the overreliance by many economists on models as the only valid tool to analyze economic circumstances that are too complex for modelling” (, 18).

8

The point being made here is that formalism is being used to dispute formalism. Sraffa, as notes, and as Mohun and Veneziani are clearly aware, also uses a formal model (having moved on from his critique of Marshall in the 1926 paper) to both dispute mainstream capital theory and build an alternative picture of a simplified economy.

9

The concept of causation is also one that can create problems here, and this is something that has not received a great deal of attention in the economics discourse regarding philosophical aspects of the problem. However, Greco and Groff (2012), Groff (2008), and have explored the issue in ways that are within the same ontological framing.

10

For this line of argument, see, for example, . If one considers this specifically in terms of Maki’s concerns with ontological unification, then clearly Lawson’s version of social ontology does face challenges. It shares common origins with Bhaskar’s critical realism and its various forms and has also substantive content—and thus claims that are significant for the nature of domain-specific theorizations. Insofar as it is, then, intruding into disciplines that are occupied by existent theorizations, it challenges those theorizations based on specific critique and also the notion of immanent critique of their emergent problems. This necessarily creates resistance, sociological but also substantive, by those who reject ontology or this ontology. Again, however, much of the concern here misreads the intent of Lawson’s work, though there is not the space here to develop this.

11

For example, one might also consider Gilbert’s (1992) work on social facts and Sheehy’s (2007) work on the reality of social groups. I thank an anonymous referee for these references and would also add the work of .

12

For critique, see .

13

The term is typically used to differentiate ontology and epistemology in terms of material reality—laboratory experiment creates a controlled environment in which regularities are induced/observed—but the regularities are expressions of powers of things that would not be observed in quite the same way beyond the laboratory—otherwise laboratory experiment would not be required. Insofar as it is required, experiment is conditional and fallible, and its insights corrigible, so reality cannot be reduced to knowledge claims—otherwise further experiment would be redundant. Thereafter, experiment is referred to as part of the argument for the circumscribed naturalism of society—laboratory experiment has less insight and practical purpose with regard to society and is typically morally objectionable in terms of consequences for people—though clearly some forms of experiment are still undertaken.

14

One might, for example, place this in the context of Wittgenstein’s critique of GE Moore’s conflation of “know that” and demonstrated “certainty.” “Know that,” “certain that,” and the grounds of doubt are distinguishable issues and highly contingent. When considering potential, one must also consider the standard of doubt applied—one may cease to pursue a course of investigation or action based on a given standard of evidence, but it is important to consider what else among our beliefs would meet the same standard.

Author Biography

Jamie Morgan works in the Faculty of Business and Law, Leeds Metropolitan University. He co-edits, with Edward Fullbrook, Real World Economics Review, the world’s largest open source subscription economics journal, and he is the coordinator of the Association for Heterodox Economics. He has published widely in political economy, economics, philosophy, sociology, and international studies and is particularly interested in financialization.

References

Akerlof

George

. 2003. “Writing ‘The Market for ‘Lemons’’: A Personal and Interpretive Essay.” http://www.nobelprize.org/nobel_prizes/economic-sciences/laureates/2001/akerlof-article.html.

Alston

William

. 1996. A Realist Conception of Truth. Ithaca: Cornell University Press.

Archer

Margaret

. 1995. Realist Social Theory: The Morphogenetic Approach. Cambridge, UK: Cambridge University Press.

Boland

Lawrence

. 2010. “Cartwright on ‘Economics.’” Philosophy of the Social Sciences 40 (3): 530-38.

Boylan

Thomas

O’Gorman

Paschal

. 1999. “Critical Realism and Economics: A Causal Holist Critique.” In Critical Realism in Economics: Development and Debate, edited by Fleetwood

Steve

, 137-50. London: Routledge.

Brown

Andrew

. 2007. “Reorienting Critical Realism: A System-Wide Perspective on the Capitalist Economy.” Journal of Economic Methodology 14 (4): 499-519.

Chisholm

Roderick

. 1966. Theory of Knowledge. Englewood Cliffs: Prentice.

Chisholm

Roderick

. 1982. The Foundations of Knowing. Minneapolis: University of Minnesota Press.

Chisholm

Roderick

. 1996. A Realistic Theory of Categories: An Essay on Ontology. Cambridge, UK: Cambridge University Press.

10.

Colander

David

. 2010. “The Economics Profession, the Financial Crisis, and Method.” Journal of Economic Methodology 17 (4): 419-27.

11.

Colander

David

Holt

Richard

Rosser

Barkley

Jr.

2004. “The Changing Face of Mainstream Economics.” Review of Political Economy 16 (4): 485-99.

12.

Davidson

Paul

. 1991. “Is Probability Theory Relevant for Uncertainty? A Post-Keynesian Perspective.” Journal of Economic Perspectives 5 (1): 129-43.

13.

Davis

John

. 2006. “The Turn in Economics: Neoclassical Dominance to Mainstream Pluralism.” Journal of Institutional Economics 2 (1): 1-20.

14.

Dow

Sheila

. 2007. “Variety of Methodological Approach in Economics.” Journal of Economic Surveys 21 (3): 447-65.

15.

Elder-Vass

David

. 2010. The Causal Power of Social Structures. Cambridge, UK: Cambridge University Press.

16.

Fine

Ben

. 2006. “Critical Realism and Heterodoxy.” Unpublished Mimeo, Response to Lawson’s “the Nature of Heterodox Economics.” http://eprints.soas.ac.uk/7024/.

17.

Fine

Ben

Milonakis

Dimitris

. 2009. From Economics Imperialism to Freakonomics. London: Routledge.

18.

Fleetwood

Steve

, ed. 1999. Critical Realism in Economics. London: Routledge.

19.

Fleetwood

Steve

. 2012. “From Political Economy to Economics and Beyond.” Historical Materialism 20 (3): 61-80.

20.

Fullbrook

Edward

, ed. 2009. Ontology and Economics. London: Routledge.

21.

Gilbert

Margaret

. 1992. On Social Facts. Princeton: Princeton University Press.

22.

Goldman

Alvin

. 1999. Knowledge in a Social World. Oxford: Oxford University Press.

23.

Greco

John

Groff

Ruth

eds. 2012. Powers and Capacities in Philosophy. London: Routledge.

24.

Groff

Ruth

, ed. 2008. Revitalizing Causality. London: Routledge.

25.

Hodgson

Geoff

. 2004. ‘On the problem of formalism in economics’ Post-Autistic Economics Review 28:3-11.

26.

Independent Evaluation Office. 2011. IMF Performance in the Run-Up to the Financial and Economic Crisis: IMF Surveillance in 2004-2007. Washington, DC: IEO-IMF.

27.

Kaidesoja

Tuuka

. 2013. Naturalizing Critical Realist Social Ontology. London: Routledge.

28.

Kaldor

Nicholas

. 1972. “The Irrelevance of Equilibrium Economics.” The Economic Journal 82 (328): 1237-55.

29.

Krugman

Paul

. 2009. “How Did Economists Get It so Wrong?” The New York Times–Magazine, September 2 http://www.nytimes.com/2009/09/06/magazine/06Economic-t.html?pagewanted=all&_r=0.

30.

Kurki

Milja

. 2008. Causation in International Relations. Cambridge, UK: Cambridge University Press.

31.

Lawson

Tony

. 1997. Economics & Reality. London: Routledge.

32.

Lawson

Tony

. 2003. Reorienting Economics. London: Routledge.

33.

Lawson

Tony

. 2006. “The Nature of Heterodox Economics.” Cambridge Journal of Economics 30 (2): 483-507.

34.

Lawson

Tony

. 2009. “Applied Economics, Contrast Explanation and Asymmetric Information.” Cambridge Journal of Economics 33 (3): 405-19.

35.

Lawson

Tony

. 2012. “Ontology and the Study of Social Reality: Emergence, Organization, Community, Power, Social Relations, Corporations, Artefacts and Money.” Cambridge Journal of Economics 36 (2): 345-85.

36.

Lawson

Tony

. 2013. “What Is this ‘School’ Called Neoclassical Economics?” Cambridge Journal of Economics 37 (5): 947-83.

37.

Lewis

Paul

, ed. 2004. Transforming Economics. London: Routledge.

38.

Maki

Uskali

, ed. 2001. The Economic World View. Cambridge, UK: Cambridge University: Press.

39.

Maki

Uskali

. 2009. “Economics Imperialism: Concepts and Constraints.” Philosophy of the Social Sciences 39 (3): 351-80.

40.

McCloskey

Deirdre

. 2010. Bourgeois Dignity: Why Economics Can’t Explain the Modern World. Chicago: University of Chicago Press.

41.

Milonakis

Dimitris

Fine

Ben

. 2009. From Political Economy to Economics. London: Routledge.

42.

Mohun

Simon

Veneziani

Roberto

. 2012. “Reorienting Economics?” Philosophy of the Social Sciences 42 (1): 126-45.

43.

Morgan

Jamie

. 2013. “Forward-Looking Contrast Explanation, Illustrated Using the Great Moderation.” Cambridge Journal of Economics 37 (4): 791-818.

44.

Morgan

Mary

. 2012. The World in the Model. Cambridge, UK: Cambridge University: Press.

45.

Morgan

Mary

. 2013. “Nature’s Experiments and Natural Experiments in the Social Sciences.” Philosophy of the Social Sciences 43 (3): 341-57.

46.

Muth

John

. 1961. “Rational Expectations and the Theory of Price Movements.” Econometrica 29 (3): 315-35.

47.

O’Boyle

Brian

McDonough

Terrence

. 2010. “Critical Realism, Marxism and the Critique of Neoclassical Economics.” Capital & Class 35 (1): 3-22.

48.

Palley

Thomas

. 2013. “Gattopardo Economics: The Crisis and the Mainstream Response of Change that Keeps Things the Same.” IMK Working Paper, Dusseldorf, April.

49.

Sassower

Raphael

. 2010. “Is Homo Economics Extinct?” Philosophy of the Social Sciences 40 (4): 603-15.

50.

Schabas

Margaret

. 2009. “Constructing ‘the Economy.’” Philosophy of the Social Sciences 39 (1): 3-19.

51.

Sheehy

Paul

. 2007. The Reality of Social Groups. Aldershot: Ashgate.

52.

Sraffa

Piero

. 1926. “The Law of Returns under Competitive Conditions.” The Economic Journal 36 (144): 535-50.

53.

Stern

Nicholas

. 2013. “The Structure of Economic Modelling of the Potential Impacts of Climate Change: Grafting Gross Underestimation of Risk onto Already Narrow Science Models.” Journal of Economic Literature 51 (3): 838-59.

54.

Stiglitz

Joseph E.

2009. Report of the Commission of Experts of the President of the United Nations General Assembly on Reforms of the International Monetary and Financial System. New York: United Nations.

55.

Velupillai

K. Vela

. 2007. “Variations on the Theme of Conning In Mathematical Economics.” Journal of Economic Surveys 21 (3): 466-505.