I came across a seminar given by Prof Ricardo Crespo (IAE Universidad Austral) in the IAS building last night organised by the Centre for Humanities Engaging Science and Society (CHESS) on Causality, Teleology and Explanation in Social Sciences.
The abstract ran:
This paper argues that four analytical levels may be found in social sciences, including economics –namely, a) a statistical descriptive level, b) a causal explanatory level, c) a teleological explicative level and d) a prescriptive teleological level. Typically, social sciences only consider levels a) and b). The exclusion of level c) may lead to viewing behaviors that do not respect theories such as the rational choice theory or the expected utility theory –theories which adopt “instrumental rationality”—as “anomalies”. Including level c) entails considering “practical rationality” and makes those anomalies reasonable. The paper adopts Aristotle’s causality notion and teleology as a theoretical framework. The first section introduces these notions, while the second section explores contemporary conceptions of causality and teleology. The third section applies the former theories to the analysis of social sciences –specifically, economics—and looks at Carl Menger’s classification of economic disciplines.
There is real challenge in giving seminar presentations to interdisciplinary audiences. I think that Ricardo, a charming and amusing Argentinian philosopher of economics, was torn between a more concrete presentation on the philosophy of economics - crucially, how a normative notion of final cause changed the kind of accounts available - and a thorough account of the relation between his Aristotle-influenced account and modern philosophical discussion of normativity, teleology and functions. So, as sometimes happens, the talk fell a little between stools.
But I wondered how such a view might work. For simplicity, one might concentrate not on the four levels of the abstract but the contrast emphasised in discussion between efficient causes and final causes with the latter connected to talk of teleology and functions. Given that passing mention was made to philosophers who were reductionists or eliminativists about functions, reducing them to merely efficient causes, what might be claimed by stressing the importance of teleology and final causes in a discipline such as economics?
My diagnostic thought ran something like this. Imagine that someone only ever offered accounts of phenomena in efficient causal terms, explanations, we might say. That might capture the full explanatory ambitions possible for worldly happenings. But still, there would be something missing from the account from the perspective of someone who also thought of the world in normative terms, of what the right or correct or ideal thing to happen was. Given that I am an IAS Emergence fellow, I could say that the normative emerged from the efficient causal history of happenings. To someone who also wanted such normative understanding, mere efficient causality is only part of the story. But I am not sure that that is news to anyone, that we need a philosophy presentation to say that. (I could be wrong about economic theory, of course, but I doubt it given the talk I have overheard in pubs of rational choices and rational choice theory.)
But post Darwin, there is a further possible disagreement. If one thinks that reductionists about – apparent, we should now say - teleology aim to show how the pattern of intelligibility of normative notions can be fully captured or explicated in causal efficient terms then the addition of normative notions to an efficient causal story is not an addition of kind after all. The addition is merely a neat shorthand for notions fully capturable in the former account. So in insisting on teleological notions one might be announcing a disagreement with the possibilities of such reductionism. One might be saying that any teleological addition is irreducible. (If so, though, I would expect the argument to focus on just this point. Why, eg., is Milikan wrong to claim that logic will become an empirical science?)
Note two ways of reading the reductionist ambition. It might aim merely to show how efficient causal processes can track normative one, independently understood. Or it might aim to explain the latter concepts. I take Milikan to be aiming at the latter, more ambitious project. She wants to show that the very teleological concepts are really disguised efficient causality concepts. A more modest reading would return us to the position outlined a little earlier: we can render unto Caesar teleological concepts whilst showing how they add nothing to the history of happenings in the efficient causal history-of-happenings' own terms.