Also, Blackburn’s dismissal ignores the fact that Anscombe never says that no ethical concepts can have application in a non-theistic context, but only that concepts like duty, obligation, and others that presuppose the existence of moral law per se (as opposed to moral virtue) cannot have it, since moral law, so the argument goes, has no coherent meaning if the existence of a moral lawgiver is denied. So to show that the sorts of moral judgments Blackburn wants to affirm really can coherently be affirmed within an atheistic framework, he has to show that their apparent coherence doesn’t in fact depend on an illicit transition from their defensibility in terms of concepts like virtue, flourishing, social utility, etc. (which do not require a lawgiver) to the conclusion that they have also thereby been shown to count as genuine duties or obligations. That is to say, in order to make his case, Blackburn has carefully to disentangle, and address separately, those aspects of the atheist’s moral judgments that Anscombe might admit can be rationally justified without appeal to a lawgiver, and those that she would insist cannot be so justified. It will not do merely to note that secularists can coherently apply “moral” categories, for the term “moral” is ambiguous between these two distinct elements which need to be distinguished.
He also discusses Blackburn's more specific complaints about 'natural' as a moral term. Arguably Blackburn is on stronger ground here, but as Feser notes, the argument presupposes that 'using something contrary to its natural function' just is 'using something for something other than its natural function'; and given the theory of practical reason behind natural law theory it is not so clear that we can always make this identification. The sense of 'natural' here is the sense of natural in which human beings are naturally rational animals; anyone who admits that immorality is irrational has conceded the general point, unless they hold either that rationality is not natural to us or that it is irrelevant to the rest of our natural functions. And it is clear that reason allows for a distinction between doing something other than a given conclusion of practical reason and doing something contrary to a given conclusion of practical reason; e.g., for the former there might several possibilities, and therefore several different rational paths. It is certainly true that the specifics are difficult to get precise on (this has always been explicitly pointed out by natural law theorists), and there will certainly be room for argument about them; but Blackburn's argument doesn't seem promising as a refutation of Anscombe's general line of argument.
(Blackburn's walking analogy, by the way, is a much weaker parallel than its verbal formulation suggests, even abstracting from issues of practical reason. When I bicycle, it isn't clear that I am using my legs for a different function than when I am walking. The parallel would not be to cases like contraception, etc., but to artificially aided insemination -- the reasons Catholic might have for caution on artifical insemination are rather different from their reasons for rejecting artificial contraception, which is, for instance, why Catholic philosophy is more accommodating of artificial insemination than it is of artificial contraception.)