Dan Priel (Osgoode) has posted a draft of this paper on SSRN. I assume it's an early draft, since there's lots of rhetorical excess and overreaching, and associated carelessness, but there is a core idea that is interesting and prima facie plausible and deserves attention and development.
The idea is this: the "classical" positivists like Hobbes and especially Bentham "saw theorizing about law, just like theorizing about morals, as part of theorizing about nature and about human nature in particular" (4), and in this respect, they have much more in common with both contemporary and classical natural law theorists. By contrast, contemporary legal positivists (Hart and progeny) do not have developed views of nature and human nature which inform their defense of positivism as a theory of law. Priel sometimes overstates this, since I assume we know, by now, that Hart operated against a background set of metaphysical assumptions (a kind of moral anti-realism, non-cognitivism about moral semantics, some views about human nature apparent in his conception of the 'minimum content of natural law' and so on), but it is certainly right that the core defense of the positivist theory is that it explicates what the "ordinary" man familiar with the "modern municipal legal system" knows about law, however inchoately: in that regard, it is a legacy of the approach to philosophy prevalent in England at the time Hart wrote. Priel is sometimes weirdly dismissive of this project (it is mere "empirical observation or a linguistic stipulation" , and he reduces it, wrongly, to a mere concern with questions of 'legal validity'), but he's certainly right it's rather different than Bentham's approach (Priel, 7-9), which assumes the truth of psychological hedonism, and then proceeds to evaluate rules, legal or 'moral,' on the basis of how well they maximize pleasure or pain. The separability thesis, on this way of looking at things, is essential for the normative project of opening up the law to utilitarian criticsm. This isn't quite how Priel puts thing, but I think this is the jist of the idea. There's some unclarity about the precise connection between the theory of human motiviation and the positivism in Priel's presentation, but I think he's right to understand Bentham's legal positivism as a complement to the theory of human nature (unfortutanely, a false theory of human nature, which might, of course, give someone pause about this kind of approach). So Priel's paper invites us to ask: what might a theory of law look like that tries to locate it within a broader picture of what the world, and people, are really like? Natural law theory proceeds that way, so too classical positivism (as Priel argues), and so, too, one might add Scandinavian Realism in some measure, though without a theory of human nature, per se.
I'll leave to others to correct some of the rhetorical excess and careless of the current draft of the paper, and just focus, for obvious reasons, on the surprising discussion of my views at 21-24--surprising, in part, because Priel is often sympatico with my views in his other work, but here makes such a hash of what I've written that I'm not sure what to think. It's just false (indeed, obviously false) that I try "to ground legal positivism on metaphysical grounds similar to those of Hobbes and Bentham" (21). Hobbes and Bentham have views about human nature; I do not suggest that a view about human nature supports legal positivism, but rather that a view about correct philosophical methodology (methodological naturalism) does, as an a posterioi matter.
Priel, however, expresses the bizarre view that methodological naturalism in my sense "is either uninteresting or false" (22). Certainly proponents of a priori methods in philosophy don't think the naturalist challenge is "uninteresting," though they would agree it is false! Priel thinks that scientific method involves "drawing inferences only if they follow logically from the data or if they are supported by statistically significant evidence" (22). We've learned the lesson over the last forty years that methods that work in the various sciences are not unitary, but no one, of course, thinks they are exhausted by this characterization, which puts no constraint on what counts as data, the methods by which we ascertain epistemically relevant data, and what inferences are warranted from such data. The discussion that follows is such a mess I'm not sure where to begin. But when Priel arrives at the conclusion that "there is no purely 'philosophical' argument, becuase there is no pure 'philosophical' domain," he sounds, alas, like my methodological naturalist. But Priel thinks this shows naturalism is substantive not methodological, even though no substantive thesis about what there is or what we know is on offer. I hope Priel will simply purge these pages from the final version of the essay, since they are confused and irrelevant to the interesting, 'core' idea, discussed above.