February 10, 2008
Dickson on the Supposed "Conventionality" of the Rule of Recognition
I have just finished reading a characteristically careful and intelligent article by Julie Dickson (Oxford) on "Is the Rule of Recognition Really a Conventional Rule?" Oxford Journal of Legal Studies 27 (Autumn 2007): 373-402. Together with Leslie Green's important piece on "Positivism and Conventionalism," Canadian Journal of Jurisprudence 12 (1999): 35-52, it should put to rest the idea that the rule of recognition is a conventional rule in the precise sense that "the reasons for accepting and following it must include the fact that others follow it too" (Dickson, p. 375) (I'll refer to this as the Conventionality Thesis). The Rule of Recognition is most certainly a customary (or conventional) rule, in the sense of being constituted by a "shared common practice" of officials, but that is not enough for the Conventionality Thesis about which so much ink has been spilled.
Dickson shows, with a detailed textual analysis, that in the original edition of The Concept of Law, Hart plainly did not endorse the Conventionality Thesis, but that in the "Postscript" he seems to endorse the Conventionality Thesis (as a response to Dworkin's argument in "The Model of Rules II"), but, in fact, makes ambiguous remarks even in that context. Her hypothesis (plausible, I think) is that Hart was misled by what he perceived to be the "lifeline" Dworkin had thrown him in criticizing the practice theory of rules, namely, that some rules can be Hartian social rules insofar as they satisfy the Conventionality Thesis. But Hart, of course, did not need to go that route in order to response to Dworkin's false dilemma (the supposed choice between concurrent and conventional practices).
Ultimately, as Dickson writes, "the question remains whether there are good reasons to adopt...a conventionalist explanation of the rule of recognition" (388). She doesn't directly answer that question, instead examining Marmor's recent account of the rule of recognition as a "constitutive" conventional rule. She argues, persuausively to my mind, that Marmor's account is not one that, in fact, vindicates the Conventionality Thesis. Here is Dickson's useful summary of her core claims from pp. 398-399:
[T]he crucial question in all of this is whether the practice in common amongst judges of recognizing certain things as valud law is reason-giving. In examining Hart's and Marmor's views, I have sought to make clear a distinction which, in my view, is sometimes under-attended to in discussions of the rule of recognition, namely the distinction between common judicial practice as an existence condition of that rule, and as playing an identifying role with regard to it on the one hand, and that practice playing a reason-giving role with regard to it on the other....In explaining the character of the rule of recongition, including the role of common judicial practice with regard to that rule, Hart was undertaking the explanatory task of giving an account of some of law's essential properties. He was not attempting to justify or explain what is valuable about the social institution of law, and he was not attempting to give an account of the reasons why and the conditions under which the rules of a legal system including the rule of recognition ought to be accepted and adhered to.
This seems to me to be exactly right, and it supports the worry that Hart took a "wrong turn" in the "Postscript" in appearing (albeit not entirely clearly) to endorse the Conventionality Thesis.
So from Dickson and Green we have a challenge to all those (Coleman, Postema, Marmor, Shapiro, et al.) who either endorse the Conventionality Thesis or think that positivism needs to explain why the rule of recognition is reason-giving. Why isn't it enough for positivism to simply note the social fact that, where there is law, the rule of recognition is in fact taken to be binding? Does anyone know of a good response to this challenge? I do not.
As usual, Julie Dickson has it just right.
Posted by: Les Green | Feb 11, 2008 4:44:18 PM
I have not read Julie's paper yet, but I would like to clarify my view which has of course evolved over time. I simply do not believe in rules of recognition at this point but that is neither here nor there. In Negative and Positive Positivism and in subsequent articles I advanced the view that the rule of recognition is best thought of as a conventional rule, and at one point I was more explicit in claiming that it was a conventional rule in the sense of being a Lewis-like convention. I held that view in part to help explain the normative force of the rule. The rule after all was supposed to be reason giving for those whose conduct it governed. I was ultimately persuaded by Les Green and first by Scott Shapiro that this approach would not work since the rule of recognition is not a Lewis-like convention. Fine. But ever since then I have argued that at the foundation of law for positivists is a conventional social practice -- not a convention in the sense that Dickson identifies. I toyed with the idea that the social practice is best understood as a joint cooperative activity, but I am not persuaded by that view either. But if there is a rule of recognition or a conventional social practice it is supposed to be normative for those it guides. Now at one point I thought that it meant that the rule is supposed to impose duties on the officials or create genuine reasons for action, but both of those claims seem too strong to me at this point. Mistaken in fact. A practice can be normative in a number of ways, for example, by framing and shaping deliberation or by ruling out options or precluding those governed by the rule to consider certain future options. Some like Bratman refer to this as 'normative pressure' and refer to 'framework reasons' to identify the kinds of reasons certain norms create. (Framework reasons of course are not really reasons.) This all seems right to me. So I continue to believe that if there is a rule of recognition or if there is a social practice at the foundation of law, it is not enough that we say that those engaged in it think or believe that it is binding on them. We need an account of it that explains what kind of normative force it has and how it is that it has the normative force that it does.
Most of my own research on these matters has focused on an entirely different issue: namely, how to explain the consistency of the claim that only social facts can contribute to legal content with the claim that the content of law calls for a moral semantics (see my OJLS Hart Lecture), but insofar as I still worry about the social practice at the foundation of law, I persist in the claim that the practice constitutes a norm for those governed by it. It does not follow from the fact that the practice is normative that it imposes a duty on officials or that it provides substantive or genuine reasons for action. The rule or practice can be normative in a number of ways, as I mentioned above. We need an account of the way the norm operates to guide the conduct of officials and an account of how it is that a rule constituted by the behavior and attitudes of officials can be a norm in that sense.
Posted by: Jules Coleman | Feb 13, 2008 11:31:18 PM
"We need an account of the way the norm operates to guide the conduct of officials and an account of how it is that a rule constituted by the behavior and attitudes of officials can be a norm in that sense." Why for any of Hart's purposes in The Concept of Law do we need such an account?
Posted by: Brian | Feb 14, 2008 3:00:01 PM
It seems to me that Hart's goal was to provide an account of what it is for a community's norm-governed or -regulated practice, and to distinguish such practice from mere convergence in behavior motivated by fear of punishment. He was *not*, as far as I can tell, trying to explain how a practice (whether conventional or some other kind) generates or furnishes reasons for action. The former is a problem for an observer of legal systems who is trying to sharpen his perception of the relevant phenomena, whereas the latter is a problem for participants in legal systems. Starting in Dworkin's articles in the 1970s, there seems to have been a systematic conflation of these two problems. And the use of the term "normativity" has not helped as it can be used to describe both problems. That, at least, is the impression I have of this literature.
Posted by: Kevin Toh | Feb 14, 2008 11:35:59 PM
Sorry. The first sentence in my last post should have been: "It seems to me that Hart's goal was to provide an account of what it is for a community's practice to be norm-governed or -regulated, and to...."
I should have been less hurried and slapdash in writing my first post for this forum.
Posted by: Kevin Toh | Feb 15, 2008 11:22:10 AM
Thanks, Kevin, I inferred that was your meaning. And I share your view.
Posted by: Brian Leiter | Feb 15, 2008 11:46:38 AM
I think this interesting discussion raises three questions. First, is there a Rule of Recognition? Second, can we accurately describe it? Hart answered the first question in the affirmative and he gave an account of it (ie, he described the Rule of Recognition). In answering both of these questions, Hart was doing what he termed “descriptive sociology.” If Brian is correct in his recent claim (see BL, “Explaining Theoretical Disagreement”) that “there is massive and pervasive agreement about the law throughout the system” (p. 14), then the existence of the Rule of Recognition is the best explanation for the phenomenon of agreement. To this point, the questions are empirical only: “is there a Rule of Recognition and what is its content?” As I read him, Jules has gone beyond this question in asking after the normativity of the Rule of Recognition (this is the third question). Hart did not ask this question. And, I agree with the import of Brian’s question, to wit, that Hart did not need to answer this question for any of his purposes in Concept. It is against this background that I find this comment by Jules arresting: “I simply do not believe in rules of recognition at this point.” If Jules means to say that he denies the very existence of a Rule of Recognition, then (it seems) he is denying the truth of Hart’s claim that there is a Rule of Recognition. And if there is, in fact, no Rule of Recognition, then the normativity of law cannot be explained with reference to it.
Posted by: Dennis Patterson | Feb 15, 2008 12:21:06 PM
I agree with Kevin, and have sometimes thought we need a moratorium on the use of the expression "the normativity of law"--without more, it only engenders confusion about what problem(s) we are trying to solve.
Like Dennis, I am struck by Jules's recent doubts on whether there are any recognition rules in legal systems. (I take that this is not, in contrast to, say, Fred Schauer's scepticism on that point, the byproduct of some special view of what it takes to have a customary rule?) Unlike Dennis, however, I'm not sure what to make of the thought that, "if there is in fact, no Rule of Recognition, then the normativity of law cannot be explained with reference to it." No doubt. But this this also presuppose that, if there are recognition rules, then they might provide a candidate explanation of "the normativity of law"? How would they do that?
Posted by: Les Green | Feb 15, 2008 1:07:49 PM
I agree that Hart had no interest in explaining how a practice (the rule of recognition, particularly) can generate reasons for action, but I wouldn't go so far as to claim that it is "enough for positivism to simply note the social fact that, where there is law, the rule of recognition is in fact taken to be binding" (Leiter). I think there is a middle way between the observer's and the participant's problem (Toh); one could try to explain (as it seems to me professor Coleman does at times) how a social practice could generate reasons for action so as to motivate participants to take part in it. In such a case, those reasons might explain 'what holds the practice together' (to use an expression from professor Patterson's 'Explicating the Internal Point of View'). Noting that the 'rule of recognition is taken to be binding' doesn't seem to explain much. I think Hart was trying to explain how the practice that underlies the rule of recognition is possible, but his appeal to the participants' attitude falls short of a satisfactory explanation, at least for someone interested in law as a social phenomenon (as he supposedly was).
Since such an explanation needn't end up showing 'what is valuable about the social institution of law' (Dickson), it seems to me it would be true to Hart's descriptive-explanatory aims.
Posted by: E Riffo | Feb 15, 2008 5:30:12 PM
Here's my thinking on the comments. I don't think we add anything of real importance in calling it a Rule of Recognition unless that did some explanatory work that couldn't be done without it. In that sense I don't believe in the Rule of Recognition. Of course I believe in recognition rules, but like Raz, I do not believe in the Master Rule of Recognition. When i did believe in it, I believed in two other things I no longer believe in. I believed it had to be reason-giving in the full sense of reasons; and I believed that it was a coordination convention. Later I believed in it being some other kind of a social rule, e.g. as in a Shared Cooperative Activity. In each case I continued to hold on to the view that it was reason giving in the full sense. That is what I have abandoned.
And in abandoning that I have abandoned the coordination convention and SCA stories. So I have given up the idea that there is a Rule of Recognition, not the idea that there are recognition rules. I have given up the view that it must be a reason giving rule; and of course i have given up various explanations of the kind of reason giving rule it is. Finally, whereas in Neg and Pos Positivism I was trying to explain what Hart is up to, I have not been doing that for a long time. I am not trying to explain what Hart did or what he took himself to be doing. I thought the question was whether legal positivism is committed to a Rule of Recognition, and whether if it is, such a rule is a convention.
So if this is what I have given up, what do I hold. I am fundamentally concerned with a different but related question. What are the determinants of legal content and how do those determinans fix the content of law. I The Dworkinian believes that moral facts necessarily contribute to legal content (through interpretation). the Razian believes that only social facts contribute to legal content (through the sources thesis). i believe that only social facts can determine the determinants of legal content (which can include moral facts). Some may recognize this as the true nature of the dispute among Dworkin, exclusive and inclusive legal positivists. The second question is what descriptions of legal content are warranted; the third question is 'what warrants those descriptions.'
The question we are exploring in these comments is tangential but important to these questions. And summarily and perhaps unhelpfully, my answer is this. The social facts that fix the determinants of legal content is a practice among officials. That practice not only fixes the determinants of legal content, but also must be understood as guiding the conduct of those engaged in the practice in regards to the determinants of conduct. So it is a norm. I believe that we cannot do without the claim that there are social facts at the foundation of law, and that those social facts alone fix something: in my view that fix the sources of legal content and not legal content itself. Now it is a further question that is not part of legal positivism, but part of my theory as such that those facts that fix the determinants of legal content constitute a practice among officials; Read Hart as claiming they constitute a R of R. In any case, in my view, that practice is normative because it must also play the role of regulating officials with regard to determining the content of law: what sources they have grounds to appeal to and which not; and so on. And so for me, I think I have to give an account of what kind of norm it is and how it guides the conduct of those it regulates. My earlier comments are designed to show that while I have given up the view that it guides by providing full fledged reasons, it can do so by providing what Bratman calls framework reasons, etc. I don't want to get into a debate here ( I am saving it for my book with OUP), but I am not convinced that it is is enough to say merely that the officials take the practice to be normative for them. You need to explain what that amounts to: There is law when there is a practice that is normative for them (not in the sense of imposing duties or providing full reasons) but in the sense of guiding their conduct in certain ways.... etc. And its normativity in that sense is not a funciton of their beileving that it is normative or in their taking it as normative; its normativity depends on it being the kind of practice that it is; and so we need an account of the kind of practice that the relevant social facts about officials constitutes. I for one do not see Dworkin as having conflated two issues as much as his having drawn our attention to the fact that we cannot separate them. None of this is incompatible with adopting an explanatroy -descriptive methodology and none of it commits one to a view that jurisprudence is an account of the value of law; and so on. At least that's my take; and note, I am not saying that this is an explication of Hart, nor am I saying that my view represents what positivism as such is committed to. Like Raz, no longer find it especially helpful to think about postivism as such. I think Kevin Toh has done a good job of explaining what Hart was up to in most ways. And I think Les and Julie have done good jobs undermining conventionalism in the narrow sense. I am just trying to work through my own understanding about the determinants of legal content and the normative structure of the social practice that in my view is the foundation of law.
Posted by: Jules Coleman | Feb 16, 2008 8:33:52 AM
Thanks, Jules, for your clarificatory comments. My difficulty with what you say I think reduces to the following. You want to say and argue for something like:
(*) Practice P determines or fixes (directly or indirectly, partially or wholly) the content of the legal norm L.
Now, as Hart’s distinction between internal and external statements shows us, statements like (*) are systematically ambiguous. It could be a statement in the mouth of an observer who is saying that the existence of a legal norm in a community consists of the existence of a particular practice in that community. Or it could be a statement in the mouth of a participant in a legal system who is saying that the existence of a particular practice gives him reasons to act according to a norm. If you mean the former, I have no problem with it, and I doubt that anyone would disagree. But if you mean the latter, and I am thinking that you do, then I don’t see why a legal positivist or anyone should adopt such a view and try to justify it. As Hart says, people can accept legal norms for a variety of reasons -- not just out of moral convictions, and not just because there is a practice in their community of adhering to those norms. And I am not saying these things merely to point out what Hart says. The distinction between internal and external statements and the systematic ambiguity of normative expressions is something that has been widely noticed by deontic logicians as well as legal philosophers like Hart. And what Hart says about the variety of reasons that motivate people’s acceptance of laws seems right. So, I don’t quite see why you want to take on the task of trying to show why people’s acceptance of laws is justified (in the full or some anemic point-of-view sense) by the existence of a practice (whether conventional or some other kind) among their fellows of adhering to those laws.
Posted by: Kevin Toh | Feb 16, 2008 10:11:59 AM
No more than Jules am I interested in patrolling the boundaries of 'legal positivism'--or any other 'ism', for that matter. But I am always anxious about the possibility that I've simply misunderstood my elders and betters. So without trying to entice him to tell us how the next instalment ends, I am interested in this idea:
"There is law when there is a practice that is normative for [the relevant officials] (not in the sense of imposing duties or providing full reasons) but in the sense of guiding their conduct in certain ways."
Of course, no one should be satisfied by the bald declaration that "officials take the practice to be normative for them." We still need an explanation of what it is for them to take an existing practice in that way. On one view, there is law only if officials are bound, in accordance with their own customary practices, to treat certain things as sources. If one thinks that, then an adequate account of recognition rules (whether one or more such rule is not at issue here) does need to explain how it is that officials can intelligibly see themselves as having duties to apply law (where there is law to apply). I read Jules's last remarks as now suggesting something much weaker: so long as these practices guide official conduct *in some way or other*, we have what we need. Is that right?
Posted by: Les Green | Feb 16, 2008 11:16:19 AM
I think there is a way to explain what it is for officials “to take the practice to be normative for them." Sociologists might explain this by refering to some kind of class structure; psychologist might prefer psychological structures, Dworkinians moral reasons. Judges would, as a first guess, give their specific reasons for each individual case. Since each individual case is disputed it is easy to challenge this answer and insist that a general explanation must either refer to social facts or moral reasons. Neither answer would however satisfy the judge as neither answer would account for the fact that the judge thinks her verdict is a correct judicial decision. This can however be explained by pointing out the fact that a different judge (as an ideal type) would have reached the same verdict. This is a jurisprudential explanation in the precise sense that it describes the practice while taking an internal point of view.
Posted by: Detlef von Daniels | Feb 27, 2008 11:34:26 AM
The comments to this entry are closed.