In defense of Brian Leiter’s naturalized jurisprudence

Brian Tamanaha (BT) critiques Brian Leiter’s (BL) naturalized jurisprudence on Balkinization. But the critique leaves me somewhat cold.

First, note that the BL paper that BT discusses looks to be a catalogue of ways that one can apply a broadly naturalistic approach to jurisprudential questions. BT only considers the first of these, a Quinean “follow the social science” approach, which BL ends up somewhat skeptical about.

That being said, there seem to be two plausible interpretations of BL’s first approach — I’m not sure which is correct, so I’ll discuss both, in the absence of clarification from BL (paging a professor Leiter … Brian Leiter… please pick up the red comment button or talk to a ticket agent…).

First, there’s the pick a theory that is consistent with the social scientific results approach. If we were to take this approach, I take it we’d proceed in some way like the following:

1. Identify the dominant social scientific theories about the phenomenon in question.
2. Determine what features of the concepts encompassed by the phenomenon in question are implied by those theories.
3. See if we can draw conclusions about what must be true, conceptually, if those social scientific theories are correct, or which conceptual accounts best organize the results of those theories.
4. Prefer conceptual accounts that are consistent with 3) to those that are inconsistent with it.

If that’s BL’s general approach, then BT’s critique seems to miss the mark. BT makes several critiques of BL’s approach, but they all amount to the claim that social scientists don’t need to be committed to a particular theoretical/conceptual framework to do their jobs.

Several thoughts on that. First, it seems just false to me. The theory-ladenness of observation suggests that conceptual choices about, e.g., what law is are going to pervade social scientific work about law anyway. That doesn’t help BL, for it might be reasonable to critique a project of “naturalized” conceptual analysis for circularity to the extent that it derives conceptual claims about law from social scientific results that themselves depend on contestable claims. (That is, if the social scientific results only make sense if law is X, it’s problematic to infer from them that law is X. It’s equally legitimate to infer that law is ~X and the social scientific results are, for that reason, flawed.) But note that this is the opposite problem from the one that BT identifies, which hinges on there not being conceptual claims underneath social science.

A second problem with BT’s critique is that it focuses on the presuppositions of the social scientific theories, not their results. That is, he takes BT’s argument to be: the attitudinal model of judicial decision making presupposes the following notions about law: A), B), C). A-C are only consistent with Raz’s hard positivism. Therefore, hard positivism. This would be a dubious mode of argument — not only because, as BT points out, A-C may be consistent with a bunch of stuff in addition to hard positivism, but also for the circularity reason noted above. But I don’t think it’s BL’s position. I take BL to be making a subtly different argument: the results and conclusions that have come out of the attitudinal model of judicial decision making show us certain truths about how the law works and how the players in the legal system understand their roles. Hard positivism provides the best account that makes sense of these facts about the world. (See also: xphil.) And I think this is a much stronger position.

So much for the first interpretation. On the second interpretation, what we do is identify the necessary presuppositions to the social scientific theory’s functioning as a social scientific theory.

There are passages in the BL paper that BT linked that do suggest that we’re supposed to be looking at the assumptions of successful social scientific theories rather than their results. I take the general approach to be as follows:

1. Consider the various social scientific theories at play. Which has explanatory power as a theory — that is, which has the features of a good scientific theory, like testability, simplicity, etc.?
2. Does that theory have its explanatory power in virtue of any facts about its content, e.g., its presuppositions about what law might be?
3. Then build the conceptual theory out of those presuppositions, for it is those presuppositions that advance our scientific attempt to understand the phenomenon.

On reflection, I think this might get closer to what BL’s trying to do than the previous version. It’s also a little closer to the meat of BT’s critique, but, once again, I think BT’s critique falls short of the mark.

The way in which step 2 works in this interpretation seems to be that one determines what must be true to give a theory empirical content. In the case of the attitudinal model, for example, in order for it to have any empirical content, as BL says, “there has to be, among other things, a clear demarcation between the ideological attitudes of judges (which are causally effective in determining the decisions) and the valid sources of law[.]” (Call the quoted claim D.) But it’s not just a direct translation of D from the attitudinal model to the concept of law. Rather, there’s the intermediate claim that the attitudinal model only works — is only explanatory — only has social scientific bite — if D is true. And that idea doesn’t rely on D’s actually being presupposed by the people who implement the attitudinal model. So, where BT says: “it is a supreme stretch to assert that a simplified model of judging makes any claim at all, implicitly or otherwise, about a “concept of law[,]” BL can just agree and say that, nonetheless, for the attitudinal model to work, the claim must be accepted.

While that might post circularity problems, as noted above, it does not pose the problem that BT identified, viz., demanding that social scientific models do more heavy lifting than they were entended to do.

Lastly, BT might be right (probably is right) that this approach can yield more than one conceptual theory of law — it might be that the presuppositions of the attitudinal model are consistent with soft as well as hard positivism. But that doesn’t call the methodology into question, just one application.


Leave a Comment