ABSTRACT
Pluralism applies at a range of levels, which are sometimes confused in the literature.1 These levels are: understanding of reality (ontology), theory of knowledge (epistemology), approach to the practice of economics (methodology), and theory. At the ontological level, a distinction is drawn between understanding the economy as an open or closed system. If the latter, then behavior is law-like and at least in principle it is possible to identify these laws. This is the presumption of traditional economics, which seeks confirmation for theory in empirical evidence of event regularities. Deviations from law-like behavior are assumed reactions to “shocks”; and since these events are assumed random – a strong knowledge presumption – these shocks are part of the closed system. For many economists, the economy is understood to evolve, subject to forces which evolve; thus there is no intrinsic closure (Lawson 1997). As a reminder, it only takes one condition for closure not to be met for the system to be open (Chick and Dow 2005). If the economy is an open system (or even if it is a closed system, but our cognitive limitations preclude such knowledge) consequences exist for how we build knowledge. An open system evolves, and evolves in a variety of ways and in a variety of contexts (otherwise it would be possible to establish laws of evolution and we would be back to a closed system). Thus, with an open system we expect a variety of understandings of the nature of reality, and therefore a variety of approaches to building knowledge about it. Even if we all start with a general open-system understanding of reality, we very quickly develop different approaches to building knowledge about it. The general theory of knowledge therefore supports pluralism at the level of methodology. Once we move away from closed systems, there is no longer any demonstrably superior methodology. Thomas Kuhn’s (1970a, 1970b) concept of paradigm is useful here. Kuhn’s study of the history of astronomy demonstrated that knowledge develops within communities which have shared understandings of reality, a shared approach to knowledge, and thus shared meanings of terms, techniques, and theories. Knowledge progresses, but only by the criteria of a particular paradigm. Purveyors of the paradigm argue for the relative merits of their approach to knowledge, and communicate it by teaching from textbooks full of exemplars; in this way, through persuasion, paradigms grow in strength. Debate may exist with other paradigms, and indeed knowledge may progress as a result of this interchange (again by the criteria of a particular paradigm). But there is no basis for authoritative adjudication between paradigms. Knowledge builds within the dominant paradigm, parrying challenges to its understanding of reality and ignoring contrary evidence, until it becomes untenable. A revolutionary episode then installs an alternative paradigm, with its own understanding of reality, meaning of terms, techniques, and theories. But once again, no independent set of criteria exists to compare the two paradigms (because the subject matter is open and so cannot generate one best way of establishing truth). The paradigms are not directly comparable, although some scope for communication may exist. The paradigms are not completely incommensurate; nor are they completely commensurate.2 Thus, it is best to allow a
range of paradigms with a plurality of alternatives. As in biology, species survivability increases with an increase in plurality: given environmental change, the current strain declines and another more suited to the new environment will emerge. Given a variety of opinion, a range of paradigms allows knowledge to develop in parallel according to different approaches. This range of paradigms is, however, bound to be limited. Knowledge progresses within communities. So if meanings differ between paradigms, than individualistic paradigms inevitably fail through communication failure. Pluralism is sometimes misunderstood as “anything goes” – a “pure” form of pluralism. But if indeed any economist could claim anything as fact and any theory tenable, that is the end of knowledge. Rather, in order to be workable, pluralism must be structured around a limited number of approaches (Dow 2004a). This is not a matter of signing up to manifestos, just a matter of practically functioning within the society of economics. Economists, in addition to choosing how to do economics, must decide which conferences to attend, which journals to subscribe to, etc., according to whose approach is closest to one’s own and therefore where communication is likely to be most productive. Categories (as in paradigms) are a helpful shorthand, but the reality rarely involves strict boundaries. Indeed, many great economists are great because they creatively intermix ideas from more than one paradigm. This is how pluralism is most commonly understood at the epistemological and methodological levels: accepting that an open-system subject matter can be understood in a variety of ways, and thus it is beneficial to have a range of approaches to building knowledge. This is toleration beyond the level of common academic courtesy and ethics, compelling though such considerations are (McCloskey 1994, Screpanti 1997). It follows from the nature of the subject matter that there is no universal best way of establishing knowledge, so that it is better to allow a range of flourishing approaches in order to have more reliable knowledge overall. However, this is not a prescription for economists to follow more than one approach to knowledge. That way madness lies – it is incoherent to simultaneously sustain competing understandings of reality and meanings of terms. Any synthesizing of different approaches to knowledge, to be coherent, must be a new approach to knowledge (Groenewegen 2007). Pluralism is often discussed at other levels, notably the levels of method and of theory. If pluralism is denied at the methodological level, a presumption exists that there is one best way of approaching knowledge. This best way might consist of a range of methods, yielding a range of theories, as in traditional economics. But a monist position at the methodological level presumes the subject matter yields laws to account for human behavior, which implies a mathematical method is sufficient, and in principle one theory should emerge to represent these laws. This thinking is implicit in traditional economics, implying one best model to represent reality. The growing plurality which some have identified in the methods and theories of traditional economics is incongruent with the underlying monist methodological approach – it doesn’t make sense. Indeed some econo-
mists pursuing different methods and theories have the explicit goal of bringing them within the traditional deductivist framework.3
