« Can public policy support the institution of friendship? | Main | Volokh on Obama and Patriotism »

Thursday, October 04, 2007

An alternative theory of Kelsen's neglect - Americans lazy, provincial

Yesterday I asked why it is that Americans ignore the most influential philosopher of law of the previous century - Hans Kelsen. I suggested one reason was that "Americans look at law and legal systems empirically - as fundamentally involving questions of social facts," whereas for Kelsen the law was fundamentally non-empirical. Brian Leiter has responded to my post over at his Legal Philosophy Blog. What he says seems to confirm my diagnosis.

Leiter is surprised that I didn't mention "what is surely the main reason for lack of attention to Kelsen, namely, that it is pretty widely thought that Hart persuasively undermined the two most distinctive features of Kelsen's jurisprudence." But one of these two distinctive features turns out to be, as Leiter puts it, Kelsen's view "that the normative force of law was only explicable by reference to a non-natural transcendent fact, what Kelsen called the Grundnorm."

Americans side with Hart, Leiter argues, because he offered "an account of law and its apparent normativity in terms that were exclusively psychological and sociological - in terms of what legal officials do and their attitudes towards what they do - that render otiose the need to posit transcendent norms." Americans don't like Kelsen, in short, because he took a non-empirical approach to the law.

Leiter might be correct that Kelsen's resistance to empiricism is fatal. (My goal was describing why Americans don't talk about Kelsen, not whether they are right not to do so.) But his motivations appear to be just what I described in my earlier post.

By the way, I can certainly understand why a naturalist like Leiter doesn't think much of Kelsen. Leiter is a twenty-first century representative of just what Kelsen and other late Neo-Kantians were reacting against - namely naturalism and psychologism in philosophy.

But Leiter's post does suggest a problem with my diagnosis. He points out that many non-American Anglophone philosophers of law write on Kelsen. He mentions Raz, but in the UK there's also Hart himself, J.W. Harris, Tony Honore, Deryck Beyleveld, Roger Brownsword, Neil MacCormick, and J.G. Starke. What is more, those in the US writing on Kelsen (e.g. Andrei Marmor) often have an international connection. I was first exposed to Kelsen during the two years I lived in Berlin.

Since all Anglophone philosophy of law tends to have an empiricist bent, philosophical disagreement is not enough to explain the neglect of Kelsen in this country. To be sure, much that's written on Kelsen in the UK is highly critical. But so is much that's written on him elsewhere. The point is that it is written, and generally well informed. In this country, little is written and it usually gets Kelsen wrong.

Leiter's post is an example. His description of the basic norm as a "transcendent norm" is one Kelsen explicitly rejected (e.g. Introduction to the Problems of Legal Theory p. 25). Kelsen is a Neo-Kantian and the distinction between the transcendent and the transcendental is basic Kant.

The same thing might be true of Leiter's second "distinctive feature" of Kelsen's philosophy of law, also allegedly undermined by Hart, namely Kelsen's view (which Leiter associates with Austin) that "the nature of law is essentially tied to its use of sanctions." True, Austin understands the law in terms of the use, or probability, of sanctions. But this is just the sort of empiricism that Kelsen rejected. Kelsen understood the law in terms of conditions for appropriateness of sanctions. The question is normative. Whether sanctions actually occur or are indeed at all probable does not matter (Introduction p. 25). The unenforced law is no problem for Kelsen. Only lex imperfecta (that is the law with no provisions for enforcement) is.

Nor are power-conferring laws (e.g. the conditions for creating valid contracts or wills) a problem for him. Kelsen understands such laws as fragments of larger laws (If the contract is in writing, and if the other side breaches, then sanctions are permissible). He does not understand them, as Austin does, as themselves commands with nullity as a sanction. Hart tries to argue against Kelsen's fragment theory, but I don't find him that persuasive here.

Still, Leiter might be right that Kelsen's emphasis on sanctions (properly understood) is also responsible American neglect of his thought.

But the real question has less to do with whether Kelsen can be criticized. Even if he can, the question remains: Why do American philosophers of law do so little of the criticizing? After all, Austin can be criticized too, but philosophy of law courses in this country rarely omit him. 

Here's a new theory of why Kelsen is neglected. One reason is his difficult prose ("his unlovely writing style" as Leiter puts it). Reading Kelsen is hard (at times, he seems to make it gratuitously hard). The other is that American philosophers of law, especially those who work in law schools, aren't exposed to enough foreign work on Kelsen for them to feel any pressure to discuss or even understand him.

One more comment. Leiter claims that "Kelsen's huge influence in the non-Anglophone world...has much more to do with his contributions to constitutional and international law, than to legal philosophy."

In fact, it is Kelsen's philosophy of law that these foreigners are most interested in. In February of 2006, I was at a gathering of Kelsen scholars assembled by Stanley Paulson at Wash. U. Only Paulson and I were from the U.S. (One of the benefits of Americans' neglect of Kelsen is that I can count as a Kelsen scholar simply by having written one article!) The rest were from France (Charles Leben), Sweden (Ake Frandberg), Australia (Iain Stewart), Poland (Jan Wolenski), Israel, and Japan. Each person gave a presentation concerning the state of Kelsen studies in his country. The conversation was mostly about his philosophy of law.

As further evidence, look to the end of my article on Kelsen, where I list the books written on him between 1983 and 2003. I can find over 20 that are clearly on Kelsen's pure theory of law, and only a few that are arguably on his views on international law or constitutional theory.

Finally, Kelsen's views on constitutional and international law (and particularly the latter) are intertwined with his philosophy of law. You can't understand what Kelsen says about international law without understanding his doctrine of the unity of law. This doctrine has two elements: that all valid laws (including international law and the laws of the various domestic legal systems) form a single system and that valid laws may not conflict. (Introduction pp. 111-22.) Kelsen comes to this conclusion on the basis of his Neo-Kantianism and his theory of the basic norm. (By the way, the doctrine - like much of Kelsen - is far more plausible than it sounds at first.)

Posted by Michael S. Green on October 4, 2007 at 08:21 PM in Legal Theory | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference An alternative theory of Kelsen's neglect - Americans lazy, provincial:

» Michael Green and Hans Kelsen Redux from Brian Leiter's Legal Philosophy Blog
Happily, Michael Green made some revisions to the original version of his putative response to my post answering his question why Kelsen is relatively neglected by Anglophone philosophers of law (Green thinks he is particularly neglected by American ph... [Read More]

Tracked on Oct 8, 2007 8:35:16 AM


Jeremy - I think 1 and 2 especially are very important. Concerning 1, I sometimes hear it said that Kelsen's philosophy of law is only applicable to civil law. That's not at all true. But it may be more congenial to those used to the way law is taught in civil law countries. 2 is important because it helps explain why Kelsen is taken seriously in the UK but not here.

My only real objection is with characterizing Kelsen as somehow continental (as opposed to analytic). To the extent that the distinction is useful at all, he is very easily put on the analytic side. Consider, for example, his interest in deontic logic.

Posted by: Michael Steven Green | Oct 6, 2007 5:15:36 PM


Perhaps you might compose for us one of your delightful limericks on this discussion!

Posted by: Patrick S. O'Donnell | Oct 6, 2007 5:09:57 PM

Thank you, Michael, for an extremely lively discussion. I am currently working on an essay on the reception of Kelsen's works in the United States and so this and related threads have been a gold mine for me.

I am skeptical of claims that the American legal academy rejected Kelsen on the merits. I don't think he was ever really understood here, and my initial thoughts in response to Brian Leiter's comments were thus quite similar to yours. In any case, I don't think phenomena such as this can be understood purely on the level of the history of ideas. As others have suggested, one ought to take sociological perspectives into account, and here I would just note the following, and since, as I said, I am currently at work on this subject, I would be interested in hearing what others think about this:

1. Kelsen's mode of thinking about the law was especially ill-suited to the American legal academy, based as it is on the case method. Even anglophone philosophy of law abounds with concrete examples, hypotheticals and applications. In comparison, Kelsen's writing remains at a high level of abstraction throughout. From a pedagogical perspective, he must have seemed a very odd bird to the few law students who were exposed to him.

2. I think it is highly significant that legal education in the U.S. is about post-graduate professional training and therefore does not focus on foundational questions such as "what is law" in the way legal education does abroad. U.S. law students do not come to law school ready to discuss philosophy and, having already completed a course of study, they are much less willing than are undergraduates to immerse themselves in philosophically challenging material.

3. Not only do very few law students or law professors have the training to make much headway with Kelsen, because of the long-standing divide between the continental and the analytic traditions, there are very few American philosophers who are comfortable with the neo-Kantian background necessary to an appreciation of Kelsen's project. I took two jurisprudence courses during law school. Kelsen was on the syllabus for one of them. But the week before we were to read Kelsen, the professor told us not to bother as, he determined, Kelsen wouldn't make any sense to us. He was probably right about that, although the same could have been said about a number of other works that remained on the syllabus.

4. Finally, I think Kelsen arrived in the United States at a time when the key elements of the professionalization process in the U.S. legal academy had fallen into place and were in the process of consolidation. While the case method was now central to legal pedagogy, Legal Realism rejected the Scientism that undergirded that pedagogy. Kelsen's project of a science of law -- the distinction between his "science" and Langdell's "science" was surely lost in translation -- must have seemed retrograde to his colleagues in the U.S. legal academy.

Posted by: Jeremy Telman | Oct 6, 2007 4:59:06 PM


Don't you think there are quite a few people who ignore or dismiss, or perhaps better: simply don't read both Marx and Freud in the academy today? American academics can have notorious blindspots. It was not long ago, for example, that a student who wanted to study Sartre probably had to go to French Studies, and Simone de Beauvoir, to French Studies or perhaps Women's Studies, or that Heidegger was missing from the agenda of predominantly ordinary language and analytic Anglo-Saxon philosophy, or that phenonomenology had little respectability in many philosophy departments. And professional "anglophone" philosophy has been scandalously slow to recognize the significance of Indian and Chinese philosophy, or Islamic philosophy for that matter. Today, despite occasional progress on some fronts, many if not most of these philosophers will treat questions in the philosophy of religion, or on personal identity, or virtue ethics, about the emotions, or topics in philosophy of mind, for example, absolutely ignorant of how such questions have been treated in (non-Anglophone) long-standing philosophical traditions in other parts of the world. How many American philosophers are well acquainted with the writings of historical and contemporary Spanish philosophers? Canon formation is well studied in Literary Studies and professional philosophers might take a cue from their academic colleagues and think more about their canonical traditions and questions of canon formation in general within their discipline (and to quote exceptions to the generalization does not change the rule). In the age of globalization, professional philosophy remains in many respects rather provincial and parochial, which certainly allows one to seriously entertain the possibility that Kelsen might be (comparatively speaking) ignored or forgotten in Anglophone jurisprudence.

Posted by: Patrick S. O'Donnell | Oct 6, 2007 9:08:38 AM

Bruce and Tim - I try to answer the question of why Kelsen should be read in a new post at


Also - can 50 million Frenchmen (and Austrians, Germans, Italians, Latin Americans...) really be that wrong? Because Leiter thnks that Hart so decisively refuted Kelsen, he is forced to attribute their abiding interest in Kelsen to 1) their being not "well-informed about Anglophone jurisprudence," 2) their low "level of philosophical sophistication," or 3) "issues of national pride and loyalty (sometimes an issue, one senses, in some German-language jurisprudential work when it comes to Kelsen)."

Couldn't it be instead that there is something we are missing?

Posted by: Michael Steven Green | Oct 6, 2007 7:30:28 AM

My previous post didn't get approved somehow.

Anyhow I wanted to say this. Is it possible that the fact that anglophones ignore Kelsen, despite access to him, some evidence that his enterprise wasn't quite all that?

That is, academic work like that of Karl Marx, Sigmund Freud, Albert Einstein Joseph Schumpeter or Werner Hesieburg can't be ignored even if you wanted to. Its too powerful and ground-shifting to ignore. Even by a lazy or provincial American academic.

But even though I like Kelsen, of what I've read, he ends up being more like a poet or musician whose work is enjoyable, but in some sense optional.

This may be a tough standard, but surely there's some truth to this.

Posted by: Tim Wu | Oct 6, 2007 4:04:42 AM

Isn't there a more obvious answer?

Don't you think that the fact that Americans have felt free to ignore Kelsen throws doubt on the power of Kelsen's whole enterprise? That is, if Kelsen's work was really so powerful, Americans or anglophones would be unable to ignore it, in the sense that, for example, its impossible for Americans to ignore the work of Karl Marx, Sigmund Freud, or even an architect like La Corbusier.

Perhaps this is a tough standard; but I'm suggesting that Kelsen ends up being more like an interesting musician or poet who you can read if you want to. This being said even though I have enjoyed and learned from what I have read of Kelsen's.

Posted by: Tim Wu | Oct 6, 2007 3:57:22 AM

I have a post up at:

Posted by: Lawrence Solum | Oct 5, 2007 1:54:06 PM

"Michael, this is weirdly non-responsive, I'm not sure you want to have this in the blogosphere. If you want to leave it up, I'll get to it tomorrow or Saturday. I hope you will rethink this."

As we used to say back in 6th grade, "Ooooooooooo!"

In all seriousness, as someone not familiar with Kelsen at all, I look forward to a substantive response from Brian; also a response from Michael to Brian's point that (as I understand it) reading Kelsen would not be worth my while; I should just read Hart et al. instead.

Posted by: Bruce Boyden | Oct 5, 2007 1:11:14 PM

It seems I've been put in my place, to wit: 'Certainly those who read Prawfsblog, who mostly haven't a clue about the philosophical issues, may find your "response" congenial; but those who know something are going to be rather surprised.' ---Brian Leiter

Posted by: Patrick S. O'Donnell | Oct 5, 2007 11:04:10 AM

I should add there is a big dose of Kelsen in the work of J.W. "Jim" Harris, the British legal theorist, particularly "Law & Legal Science," in which he attempts to reduce to basic principles the "science" by which participants in the legal system determine what the law is (they are: exclusion, subsumption, derogation, and non-contradiction). There is a neat book of essays dedicated to Harris (including writings by Stanley Paulson, Brian Bix, Richard Epstein, Stephen Munzer and others) called Properties of Law.

Posted by: Jeff Lipshaw | Oct 5, 2007 10:05:58 AM

To add some support to Michael's point (and this is something I discussed in my paper), by page 25 of Kelsen's Introduction to the Problems of Legal Theory (somewhere in my move from Tulane, I lost my copy - this is very upsetting), we are deep into the difficult distinction between "ought" as a Kantian transcendent idea, and into Kelsen's audacious move to adding the legal "ought" as an a priori (or transcendental) category of the way we order perceptual data (the Kantian categories of unity, plurality, totality, causality, etc.) I don't see that as much a problem of writing style as writing subject. In other words, if you liked slogging through the Transcendental Analytic of the Critique of Pure Reason, you'll love this. Given that metaphysics for American lawyers are by and large anathema (see my little comment in the Hofstra Law Review on this), Q.E.D.

On the other hand, we are almost at page 100 in The Concept of Law before we touch the Rule of Recognition, and it is an inductive, naturalistic derivative from the existence of primary rules and their interaction with social norms. Even then, Hart doesn't ask the "all the way down?" question about the Rule of Recognition, I suspect much to the relief of his American audience.

A different way to look at it is the fascination that Continental European thinkers have with paradox (e.g. Luhmann), a wholly non-empirical approach that Leiter and others view as "armchair conceptualizing". I'm sympathetic to that view, even if like an addict I can't stop doing it. I mean, where does it all get you? Just a greater need for another fix of metaphysics. (I have this image that I am in rehab and the treatment is that I have to do statistical analysis every time I utter the words "infinite regress.") One of the things I admire about Larry Solum's virtue jurisprudence is that it takes this philosophically based pragmatism that recognizes, almost inductively, an immanent norm of virtue, without the bother of trying to resolve the antinomies. But maybe if you like armchair conceptualizing and wallowing in antinomies, you like Kelsen. I know I do.

Posted by: Jeff Lipshaw | Oct 5, 2007 9:48:40 AM

Patrick - I haven't read Luntley's book on Wittgenstein, but after reading your comments it looks like I should. And yes I agree with the analogies you draw. We shouldn't be surprised that they are there. Kelsen was in Vienna at exactly the same time as Wittgenstein. Indeed, it is possible that they knew one another. But even if they didn't, there was a general movement toward Neo-Kantianism and away from naturalism and psychologism at the time.

Posted by: Michael Steven Green | Oct 5, 2007 9:38:49 AM

Michael (if I may),

I just read your 2003 paper on Kelsen (and Frege) and think the overarching argument is on target. In short, and please correct me if I'm wrong, what Kelsen was up to with legal science was analogous to a Fregean-like belief in the need to establish and demarcate rules of inference deemed necessary (transcendentally so) as conditions for the possibility of judgment, i.e., logical rules of inference "whose justification commits us to no more than is necessary to make thought and talk possible" (Michael Luntley), or the logical grammar of intentionality.

Incidentally, a naturalized theory of meaning will assess our ordinary ascriptions of content in terms of belief and desire in terms of whether or not they fit into a mature physcial theory which shows how things, events, and processes instantiate causal patterns, while our folk theory of ascriptions of content in terms of belief and desire is bounded by a rationality principle which accounts for this content by normative patterns that "must make best overall rational sense of the subject's behaviour" (Luntley). Naturalists, empiricists, what have you, fail to realize, again in Luntley's words, that "the difference between causal and normative patterns is a qualitative [i.e., a transcendental] difference. No amount of tinkering with descriptions of causal properties of things can provide the normativity required to make sense of the normative properties of belief." As Luntley explains, this leaves the naturalist the possibility of either rejecting our pre-theoretical or "folk" descriptions of ourselves that we assume to be true (take for granted), or acknowledging the failure of the naturalization project. For many, the preference for the first option leads them to eliminativism (a la the Churchlands) of one sort or another. Kant, neo-Kantians and Wittgenstein all proffer alternatives to the naturalization of meaning (as Luntley says regarding the latter: 'Wittgentstein's rule-following argument is a transcendental argument to the effect that it is a condition for the possibility of meaning that we accept an account of experience, and of what is available within experience, that refuses to bend to the demands of standard forms of naturalization'). It would seem to help us make sense of what Kelsen was up to in legal theory if we appreciate these competing philosophical conceptions, although the description of his theory as a "legal science" perhaps gets in the way of an appreciation of the role of transcendental necessity in making sense of normativity qua normativity. Again, please correct me if I'm wrong, as I lack formal training in this stuff and so I'm making amateur attempts to understand it (in other words, enough said as I'm probably way in over my head).

Posted by: Patrick S. O'Donnell | Oct 5, 2007 8:47:20 AM

I've enjoyed this discussion. For what it's worth, I find your reply to Leiter persuasive.

Perhaps this is already well appreciated, but I find it interesting that Kelsen’s Grundnorm sits atop a normative hierarchy in a system of positive norms that is virtually identical to (the scientific/demonstrative approach of) rational systematization and explanation in philosophy insofar as, in Nicholas Rescher’s words, “philosophy cannot provide a rational explanation for everything, rationalizing all of its claims ‘all the way down.’ Sooner or later the process of explanation and rationalization must—to all appearances—come to a halt in unexplained explainers.” Linear logical explanation or rationalization must come to a stop at some point, an endpoint, some “inexplicable ultimate” if it is to avoid an infinite regress. And yet an “unexplained explainer” also serves, as it were, as a starting point in a system that begins from axiomatic principles and proceeds to make deductions therefrom (the axioms being foundational or the ‘ground’ of the system). Whether as the inexplicable endpoint of a systematic linear explanation or as the axiomatic starting point of systemic deduction, Kelsen’s basic norm functions in an identical or similar manner. And although a source of what some take to be an odd feature of Kelsen’s theory, namely, that this norm appears to be both the foundation and apex of a structure or system, this does not appear to be so much a paradox if we keep in mind that while norms may be logically derived from the Grundnorm, as in an axiomatic system in philosophy, their legal validation proceeds in the opposite direction (as with a Constitution), the validity of lower norms being “determined by tests that look simply to whether in fact the norm in question was produced or posited in accordance with the criteria stipulated by some higher norm.” (David Dyzenhaus). Of course Kelsen was not simply transposing philosophical systematization as such if only because, as Dyzenhaus notes, “Kelsen often emphasized that the normative structure governs not so much the interpretation of norms, but their creation.” (And hence the following: ‘Kelsen held that constitutional adjudication is as highly a political and legislative an act as an original act of legislation, but he also insisted that it is determined by norms in a way that such original legislation is not.’)

I’m no expert on Kelsen (or anything else for that matter), but I think Dyzenhaus often provides us with much insight into the virtues and problems (better: ambiguities and contradictions) of Kelsen’s “pure theory of law” in his Legality and Legitimacy: Carl Schmitt, Hans Kelsen and Hermann Heller in Weimar, 1997. And Neil MacCormick also seems to provide us with a nuanced if not sophisticated picture of Kelsen in his own legal theorizing which, while deeply appreciative of Kelsen’s work, does not hesitate to be critical of it as well (cf. his Rhetoric and the Rule of Law: A Theory of Legal Reasoning, 2005).

OK, back to being a lazy and provincial American rooting for the Cubs to beat the Diamondbacks (the evening began well enough, the Indians having routed the Yankees).

Posted by: Patrick S. O'Donnell | Oct 4, 2007 10:38:02 PM

I don't know Kelsen well enough to have a strong opinion as to the worth of his work so I won't address that. But, he is not completely ignored by American philosophers of law, either. To take two examples merely becasue they happen to be close to my desk, he is discussed in David Lyons' well-known book _Ethics and the Rule of Law_ and in Philip Soper's _The Ethics of Deference_. Now, I can't say whether Lyons or Soper get Kelsen right, and maybe they don't give him as much attention as they should, but they don't ignore him. I'm sure that other examples would not be too hard to find.

Posted by: Matt | Oct 4, 2007 9:23:56 PM

Post a comment