Leiter against Dworkin, the continuing series

I love reading Brian Leiter's articles. He does a great job combining clarity with erudition, making clear where his arguments fit in with those that have been advanced by others while avoiding the temptation to romp over a field of defeated straw men. (While his readings of Dworkin aren't exactly "generous," he's hardly guilty of distorting the substance of Dworkin's arguments, even if his conclusions about their merit are rather vituperative^*.) His newest paper, available here, doesn't disappoint: I think it's his clearest and most persuasive articulation yet of why he thinks Dworkin's argument-from-disagreement--as glossed by Scott Shapiro in his wonderful The Hart-Dworkin Debate: A Short Guide For The Perplexed--fails. But I'm still not entirely convinced that theoretical disagreement can be pushed to the side.

In a nutshell, Leiter's argument is that 1, we have no good reason to treat disagreement about criteria of legal validity as the central problem for theories of general jurisprudence, since the massive agreement that characterizes legal systems is of far greater import; 2, to the extent we care about understanding such disagreement, the two natural positivist explanations (there is no truth of the matter about some legal disputes, and thus claims to the contrary are either mistaken or disingenuous) are good enough. Positivism, then, remains superior to Dworkinian interpretivism on the basis of the standard criteria (simplicity, consilience, conservatism) of theory selection. (But go read it. It's very good. Shapiro's, too, if you haven't.) Now, I'm not convinced that interpretivism is the way to go (nor do I think Shapiro's suggested solution works, either). I do think, however, that disagreement about the criteria of legal validity cannot be satisfactorily explained away by an error theory of the sort Leiter supposes.

To some degree this may be just a matter of emphasis. I care about constitutionalism, first and foremost; I look to general jurisprudence only insofar as it relates to this, and think it a mistake to just identify the legal system with the state apparatus. Since I don't really worry about, e.g., tort law or contracts, the fact of massive agreement on these topics isn't so salient. What concerns me is fundamental disagreement over the criteria for constitutionally legitimate activity, about the powers of and limits on the state, and here I just don't think you can consider the topic "marginal." I suppose that one could respond with the claim that, even here, the vast majority (however measured) of issues really are settled (even if perhaps in a decidedly suboptimal way); that's the difference between the USA and, say, Somalia or Iraq. But I just don't think this is good enough, at least not in the United States (which may be uniquely problematic in this respect). At the very least, the class of officials on whose practices constitutional criteria of validity supervene include a number executive and legislative branch actors, besides the usual Article 3 suspects ... and if you don't see deep disagreement over basic questions of constitutional authority there, I don't know what to say.

So the reason we cannot, as Leiter urges, be content with simply saying that constitutional validity norms supervene on convergent constitutional practice and that where such convergence fails, the norms run out, is that this is just not good enough for even a naturalized project of constitutional theory. What I mean by this is straightforward, and ultimately owes much to Leiter's own writings: we want to understand how it is that societies structure political institutions and procedures so that the emergent states are both empowered and constrained in credible, public ways. And to do this, we need a concept of constitutional norms that will tell us exactly how these norms supervene on the practice, what degree of convergence is required for truth-talk of them to be apt (such that disagreement reflects epistemic rather than metaphysical failure), and perhaps even how they reflect or incorporate extra-constitutional norms.

To clarify: I am not saying that a naturalized constitutionalism requires us to deny constitutional indeterminacy; that would be lunacy (or a solid reductio). But it does require us to explain the supervenience relation in a way that fits into this project--or, more pessimistically, to show us the limits of the constitutionalist project by highlighting the precarious nature of any such norms. Theoretical disagreement has to be a central problem for any theory of constitutionalism because the project is about creating structures that will withstand it.

In some ways, this might be seen as a naturalist analog of Raz's authority argument. But rather than claiming that our intuitions about legality show that laws must be individuated such that they could, in theory, guide conduct, the claim is that understanding the (social-scientific!) project of constitutionalism requires a theory of the meaning of consitutional norms, and their relation to the practices on which they supervene, rich enough to help us see when and to what degree they can be institutionalized and sustained^#. Even if many of the actors within the practice will--for predictable, Leiter-noted reasons--consistently overstate constitutional determinacy, we need to understand when and why there is or isn't a truth of the matter.

* -- "In any case, what the preceding shows, rather clearly, is that when Dworkin declares “there is no positive evidence of any kind that when…judges seem to be disagreeing about the law they are really keeping their fingers crossed” (as the Disingenuity Account would have it), what he really ought to have said is, “There may be lots of evidence, but I have made no effort to consider any of it.”" -- Leiter, p.39. Ouch.

# -- This might lead towards Mark Greenberg's "rational relation" doctrine. Or it might not! Stay tuned!