It is common for critics of evolutionary theories to accuse the proponents of such theories of giving just-so stories—stories about the evolutionary path to F with the property that if they are true, then some feature F of an organism has an evolutionary explanation, but where there is little independent evidence (independent of the claim that evolution is the correct explanation for biological features in general) for the truth of the story. It is common to criticize this by noting that a moderately clever person could give just-so stories for almost about any imaginable feature and its lack.
I take the accusation that evolutionary theorists sometimes give just-so stories to be correct, but not damning. What is important is to be clear on what the just-so story accomplishes: It shows that for all we know the development of F is compatible with evolutionary theory. It is also important to be clear on what the story does not accomplish: It does not provide significant independent evidence for the truth of evolutionary theory. The just-so story is, thus, a defensive move by evolutionary theorists, and as such is perfectly respectable, but one must be clear that it is not an offensive move.
Similarly, theists when challenged why God would have permitted some particular evil E are apt to answer with a "pet theodicy"—a story about the benefits of E with the property that if it is true, then God had good reason to permit E, but where there is little independent evidence (independent of theism, that is) for the truth of the story. These stories are relevantly very much like just-so stories. Typically what triggers the just-so story is the question of why evolution would lead to an organism with some feature that appears to be bad for the organism (cf. this discussion of blushing), while what triggers the pet theodicy is the question of why God would allow the occurrence of something bad (perhaps in a different sense of "bad"—but I am highlighting the structural similarity) that does not appear to lead to the good.
Therefore, an anti-evolutionary theist who thinks that it is damning for evolutionary theorists to give just-so stories must eschew such pet theodicies, and an evolutionary atheist who thinks that the practice of giving pet theodicies is intellectually corrupt (e.g., because a moderately clever person can give a pet theodicy for just about any imaginable evil) must refrain from just-so stories. But both would be wrong. For as defensive moves, just-so stories and pet theodicies are perfectly reasonable. What gives them bad press is the perception, just or not, that their proponents are using the stories to provide significant independent positive evidence for their theories. As long as it is clear that this is not what is being done, there does not appear to be anything problematic going on.[note 1]