Central to Mirowski's account is his observation that scientific reasoning always requires the imposition of some conservation principle: the postulation that amid observed variation some entity remains fixed. In a given explanatory framework, a conservation principle serves two purposes: it allows for observation of change where change is measured against the constancy of the conserved entity, and it gives temporal definition to a system and thus allows for causal analysis. All branches of theoretical science, Mirowski says, require some sort of conservation principle. In classical physics (or in Mirowski's reconstruction, "proto-energetics"), the conserved entity is total energy; in evolutionary biology, it is "species"; and in economics, it is value.
In economics, the conserved entity need not be value, yet value is the most obvious choice since some notion of conserved value is a presupposition of most economic activity. The exchange of equivalents and the transformation of inputs into outputs presuppose a generalized sense of determinate price relations among commodities, relations implicitly tied to some underlying conserved quantity of value. But to note our everyday reliance on some social sense of a set of determinate price relations is but an empirical generalization. It does not logically follow that value (as opposed to some other possible invariant) should be employed as the conserved entity in economic analysis. Such a stipulation, Mirowski implies, requires additional justification - justification that Mirowski argues has been provided by the metaphors of various branches of science, in particular physical theory. Recourse to metaphor for this justificatory role, Mirowski maintains, is necessitated by the fact that in the real world there are no conserved entities:
the research program situated at each vertex (here Mirowski refers to biology, physics, and economics) derives legitimacy for its radically unjustifiable conservation principles from the homeomorphisms with the structures of explanation at the other vertexes [i.e., from the other disciplines]. This legitimation function is central to the success of each research program because the central syndetic principle at the heart of each is purely conventional, and thus, from a disinterested and detached point of view, simply false [1989, 116; emphasis in original].
Economics, Mirowski maintains, has relied specifically on physics for this legitimation function: "within the ambit of the dominant schools of Western economic thought - classical political economy, neoclassical economics, and Marxian economics the theory of value, the very core of the explanatory structure, has been dictated by the evolution of physical theory" [Mirowski 1989, 396].
Thus, two rationales for value theory can be identified: the traditional idea that since the economy operates through a price mechanism, analysis of the systematic properties of the economy requires an account of price formation; and Mirowski's view that value theory is a product of intellectual need - the need to hold some entity fixed in order to define the analytical framework and to identify change and causation. Implicitly, the conservation of value is postulated in both approaches. But in the former case, it is an empirical generalization based on the fact that some ...