In 'Functional Explanations in Sociobiology' Horan proposes a characterization of functional explanation and then explores what sort of research could establish the correctness of such an explanation in sociobiology. These are both important topics, and while I am sympathetic to many features of Horan's analysis, I find her account of functional explanations and the way in which they can be justified both problematic.
Horan characterizes functional explanations as "forward looking" (p. 8) and differentiates them from evolutionary explanations, which attempt to account for an organism's traits in terms of the contributions they made to the organism's ancestors. In differentiating functional explanations from evolutionary explanations, Horan parts company with philosophers like Wright (1976) and Wimsatt (1972), who attempted to make teleological functional explanations plausible by construing them as evolutionary explanations. Horan's argument against using evolutionary explanations to explicate functional statements is, in part, that traits or behaviors may now serve different functions than those for which they were selected. I agree with the main thrust of Horan's differentiation between functional accounts and evolutionary explanations: functional statements direct us to the current contributions a trait makes to fitness, not to its past contributions. My concern, however, is with continuing to think of functional statements as explanations. In its modern usage, the term explanation refers to an account of how something came to be and, at least in the covering law account of explanation, would enable us to predict how things would be. Even those who depart from the covering law tradition, such as Salmon (1984), treat explanation as accounting for how something came to be. While Horan tries to develop the notion of a consequence explanation as an alternative to causal explanations, her own language suggests that something is askew. She tells us "consequence laws tell us that causes are brought about ... by the fact that they have certain effects" (p. 10, emphasis mine). Bringing something about would appear to be a causal process, and it would seem that this could only be done by events prior to or contemporaneous with the event being brought about.
In fact, much of what Horan has to say about explanations in sociobiology could easily be read in terms of evolutionary explanations, for at least one goal of sociobiology is to account for behavioral traits in terms of a standard evolutionary framework. In fact, Horan frequently seems to be concerned more with evolutionary explanations of functions than with functional explanations as she has characterized them. Her discussion of adaptive failures and her attempt to defend consequence laws from the charge of falsity seem much better construed as attempts to justify evolutionary explanations of the current existence of functional traits. While this may seem to provide the most natural reading of many passages in Horan's paper, it is clear that she intends consequence laws and functional facts to refer to the future. This proposal not only reverses of the normal view that present phenomena can only be explained by current or past phenomena, but creates as serious a problem as the one Horan raises against equating functionality with evolutionary explanations. Unfortunately, we must reckon with the possibility of disasters such as thermonuclear war which could occur before current members of many species could produce