Fehr, C., Pluralism and Sex: More Than a Pragmatic Issue, Philosophy of Science, Vol. 68, No. 3, Supplement: Proceedings of the 2000 Biennial Meeting of the Philosophy of Science Association. Part I: Contributed Papers (Sep., 2001), pp. S237-S250

The evolution of sexual reproduction is a case of explanatory pluralism, meaning that there is more than one explanation for this phenomenon.
(..)
There are more than 20 kinds of theories that answer the question "Why sex?" Some of them focus on meiosis, some on out crossing, and some on the combination of these two processes.
(..)
Red Queen explanation: if a host reproduces sexually it has the opportunity to produce offspring with novel defense mechanisms that the parasites are not already adapted to.
(..)
DNA Repair explanation: if both DNA strands are damaged, this can be repaired during the process of meiosis.
(..)
By domain, I mean those phenomena, that scientists group together as something to be explained. (..) Heterogeneous domains are constitutive of explanatory pluralism. (..) The case of gene regulation is a hierarchical heterogeneous domain.
(..)
Orthogonal domain: if some phenomena are members of different subdomains.
(..)
I argue that in homogeneous and hierarchically heterogeneous domains, pluralism can be decreased using van Fraassen's pragmatic machinery.
(..)
But there is a kind of domain that does not fit this analysis, where in order to understand a phenomenon one must construct a complex question that has more than one answer. This is the case of orthogonally related domains. (..) example: explanations for the length of a person - genetic (genes) and developmental (food)
(..)
Several of the presuppositions of van Fraassen's pragmatics of explanation are violated if we partition orthogonally related domains. (..) By including two variables in the contrast class we are reintroducing an ambiguity that the contrast class was designed to ferret out.
(..)
The different subdomains in orthogonally related domains are interrelated in such a way that we cannot use the machinery of the topic and the contrast class to disambiguate the question in such a way as to decrease pluralism. In some populations, when we ask "Why meiosis?" it will sometimes be explained in conjunction with out crossing and sometimes in conjunction with inbreeding, but in each of these explanations, meiosis is the same thing.