A lotta Hart, but is there a “there” there?

Scott Shapiro has an excellent article about the debate between Ronald Dworkin and HLA Hart over (to waffle a bit) the nature of legal systems. I have a question about one of Dworkin’s objections (as reconstructed by Shapiro) to Hart’s legal positivism.

Very briefly, Hart says that a standard of conduct is a rule of law (in some or another legal system) just if it meets the criteria specified in that system’s rule of recognition. A rule of recognition is a social rule specifying the criteria of legal validity.

For Hart, a rule exists if a quorum of society (in the rule of recognition’s case, the vast bulk of those learned in the law) takes what he calls an internal point of view towards a particular standard of conduct. That is: if they accept it, are disposed to guide their behavior by it, to criticize those who fail to follow it, and not to protest against such criticism when it is legitimately leveled against them.

A social rule is a standard of conduct towards which a quorum of society takes an internal point of view for reasons other than that it is a law of the society. (Obviously, one cannot take the internal point of view towards a rule specifying the criteria of legal validity because the rule is a law of the society; a law cannot specify the criteria of its own validity.)

Finally, for a rule to specify the criteria of legal validity is for it to identify the properties some other rule must possess in order to be part of the legal system. Those who accept a rule of recognition will be disposed to accept any rule that meets the criteria that rule of recognition specifies (e.g. having been enacted by Parliament, etc.).

Hart posits that all legal systems have a rule of recognition because he thinks this explains certain features that all legal systems possess. It explains why some rules, but not others, count as rules of law. (It does this better than Natural Law Theory, because it allows for the possibility of morally bad law.) It also explains why judges enforce those rules in the way they do. (It does this better than JL Austin’s Command Theory of Law. Austin’s positivism held that a legal systems consists in commands backed by the threats of a sovereign who is habitually obeyed. But judges enforce rules not out of habit, or fear of reprisals, but out of a sense of felt-obligation. That is, because they accept a rule that specifies which standards of conduct ought to be enforced.)

Enter (what Shapiro calls) Dworkin’s “objection from theoretical disagreements.” Dworkin thinks that if legal systems have a rule of recognition then it is utterly mysterious why legal theorists disagree with each other on the proper interpretive methodology. Interpretive methodologies each tell us how we are to ascertain what the law says. Originalism – which states that the content of the law is fixed by the meaning of the text at the time of the law’s enactment – is one interpretive methodology, purposivism – the content of the law is fixed by the law’s purposes – is another, and pragmatism – that the content of the law is whatever it is (all things, including legitimacy, considered) best for it to be – is a third.

These are very different methodologies, and they are all presented as specifying the criteria of legal validity. But, if Hart is right, the criteria of legal validity are specified by a standard of conduct that a given swathe of society (legal professionals) just so happens to accept (for non-legal reasons). But here are legal professionals (the most learned of them, the ones who should know the most about the nature of law, and the ones with least incentive to be disingenuous*) disagreeing about the criteria of legal validity. They are, in other words, disagreeing about what, if Hart’s right, they, themselves already accept. As Shapiro puts it:

[I]t is common ground between [all Hartian] legal positivists that the [criteria of legal validity] are determined by convention. How can they account for disagreements about the legal bindingness of certain facts whose bindingness, by hypothesis, requires the existence of agreement on their bindingness?

The rule of recognition thus may explain some features that all legal systems possess, but it leaves another – theoretical disagreement on the criteria of legal validity (or the “grounds of law”) – completely unexplained.

I think there might be a there lurking in Dworkin’s objection, but it isn’t what Shapiro identifies. Here’s Shapiro’s version, made explicit:

1. If there are rules of recognition then they are determined by convention among legal professionals.
2. If the rule of recognition is determined by convention among legal professionals then legal professionals agree on what the rule of recognition is.
3. Legal professionals do not agree on what the rule of recognition is.
4. There are no rules of recognition. [1, 2, 3]

Shapiro presents (2) as a conceptual truth: that the criteria for legal validity are supposed to be “determined by convention” means that “by hypothesis” the rule of recognition “requires the existence of agreement” about what it is. But, far from being a conceptual truth, (2) is false. Hart acknowledges that, in order for the rule of recognition to exist, it must be accepted by legal professionals (for non-legal reasons). But someone can accept something without knowing its content (or, a fortiori, agreeing on what it is). Acceptance is manifest in what one is disposed to do, not what one knows one is disposed to do. As I’ve already had occasion to observe, GE Moore makes an analogous point regarding understanding:

[T]he question whether we understand [some proposition’s] meaning (which we all certainly do) [is] entirely different [from the] question whether we know what it means, in the sense that we are able to give a correct analysis of its meaning. The question what is the correct analysis of the proposition meant on any occasion … is, it seems to me, a profoundly difficult question…”

One can believe something without knowing what it is one believes, just like one can accept something without knowing what it is one accepts. Like the analysis of what it is one believes, the analysis of what it is one accepts is (it seems to me) a profoundly difficult question. Can’t Hartians say it’s the question that legal theorists are striving to answer, and thereby explain the persistence of theoretical disagreement?

Maybe not. Perhaps the objection can still be pressed that legal theorists aren’t doing the kinds of things that are well-suited to ascertaining the content of a rule of recognition. Some of what they do – trying to show that their methodological proposals yield paradigmatically (legally) correct results – is apt to that purpose. But other theoretical contests, such as those over the moral justifiability of different interpretive approaches, seem inapposite to the question whether those approaches are (as a matter of fact) accepted. And, anyway, the attitude of legal theorists is not that of anthropological researchers aiming to uncover (even an important) social truth, but of philosophers looking to justify a practice.

I’m presently agnostic on whether considerations like those just sketched are a problem for Hartian positivism. (Food, I threaten, for a future post.) But I’m certain they are different from the clearer-cut objection with which we started, and which fails.

* All these descriptions are contestable – Critical Legal Studies contested them vigorously – but I accept them for the sake of argument. (Shapiro acknowledges that they can be contested, but implies it would be a disreputable thing to do.)