(sent to conlawprof)
1. Holmes does not have to sign on to the program of sociological jurisprudence for him to believe that the ultimate foundation and legitimacy in law comes from the new "policy science." The issue here is twofold: (a) what judges are supposed to do; and (b) what are the grounds of law. Holmes' rejection of sociological jurisprudence as a program concerns (a), not (b). That should come as no surprise. Neither judges nor the judiciary as an institution could perform it (even if the science could -- which we all know it can't). Indeed, the actual program of sociological jurisprudence performed by judges would have offend Holmes' strong sense of pragmatism.
2. But that does not mean that Holmes rejected "policy science" for (b). In fact, he did subscribe to the philosophic view that law's ultimate foundation could be had in "correct policy," (which, after all, is the intellectual event which created sociological jurisprudence in the first place). As to Holmes general endorsement of the idea that (1) law is policy; and (2) policy can be an empirically correct science, see G. Edward White, Justice Oliver Wendell Holmes, Law and the Inner Self, (New York: Oxford University Press, 1993), 148-155, 168-171, 181, 192.