Spatializing the description of method in international law scholarship

Imagine the various methodologies in international law scholarship represented on a three dimensional Cartesian coordinate grid, with the following axes:

Three Explanatory Axes

Horizontal axis:
The sovereignty continuum.

This is the line along which are located positions on state sovereignty, with sovereignty for its own sake at the left extreme, democratic sovereignty midway along the left side, and the right extreme occupied by global parliamentary governance, with liberal internationalism somewhere along the right side. In the middle would be situated sovereign state multilateralism - the balanced midpoint admitting both of sovereign states and multilateral cooperation. I've written a lot on how to set out positions along this line, in my Squaring the Circle review of Anne-Marie Slaughter's A New World Order book, for example, or in my U of Georgia review essay on Goldsmith-Posner's Limits of International Law book.

Vertical axis:
The descriptivism versus prescriptivism/normativism continuum.

This vertical axis describes the range of positions between pure empirical descriptivism at one extreme (we'll put descriptivism on the lower half of the vertical line) gradually ascending into other descriptivist positions, such as rational choice theory (and perhaps various international relations power paradigms) somewhere along that line (it is not clear exactly that they form a graduated line, in fact), and normative claims, at the other extreme, on the upper half of the line.

By prescriptivist or normativist positions, I mean international law methodologies claiming that the proper subject matter of international law scholarship set forth normative moral claims that should (and perhaps must, overtly or covertly) undergird international law, both positive and customary. The prescriptivist claim at its strongest is that a purely descriptivist approach cannot make sense out of international law because the materials of international law only make sense understood and refracted through a moral prism, without which it is not comprehensible. And the highest task of international law scholarship is to connect that moral vision with law.

The normativist positions have been those of the leading theorists of international law of the last generation in the United States - Louis Henkin, Thomas Franck, Henry Steiner and, in the next generation, Harold Koh. These normativist positions have been associated with what I suggest is actually a separate proposition, viz., a certain answer to the sovereignty question which, in the case of all these normativists, is liberal internationalism. There is a strong tendency in the literature to conflate a normative approach to methodology with a certain normative result with regards to the sovereignty issue, viz., liberal internationalism. Part of the point of this spatializing exercise is to separate out those propositions; to admit of the possibility, for example, that one might be a normativist - but endorse a different normative answer to the question of sovereignty, eg, democratic sovereignty. I would locate myself in that position.

The descriptivist positions, by contrast, have been those of the rising generation of theorists, and have been conflated, in their turn, with another answer to the sovereignty question, viz., sovereign state positions. These scholars include, of course, Eric Posner and Jack Goldsmith. However, as the Georgia conference on The Limits of International Law made clear, a large cohort of new international law scholars is now rising that is both descriptivist and, broadly speaking, liberal internationalist. Again, a large part of the point of my seeking a spatial representation of these positions is to make clear that sovereignty questions are separate from methodology questions, however one answers them.

I have been struck - I'm sure others have been struck - by the relative paucity of rising young scholars who embrace the position of the older generation of normativists and liberal internationalists; the rising generation, even when it seeks to defend liberal internationalism, seems intent on doing so on methodologies of game theory, rational choice, instrumentalism of various kinds, and not an appeal to morality itself. That is one reason why Cornell's Robert C. Hockett's recent review of The Limits of International Law in the Minnesota Law Review is so interesting - a young, very sophisticated scholar affirmatively asserting both normative and liberal internationalist propositions.

These two axes, horizontal and vertical, are two that I have described previously. I have suggested where various paradigmatic scholars, or at least key works, might fall within that grid. Thus, for example, Goldsmith and Posner's Limits book would fall in the lower left quadrant, as a book endorsing democratic state sovereignty but also embracing rational choice descriptivism. In the upper right quadrant would be many of the classic international law theorists of the previous generation or so - Henkin, Franck, Steiner, Koh - prescriptivists of a liberal internationalist commitment. In the bottom right quadrant might be some of the new generation of thinkers - Oona Hathaway and Ariel Lavinbuk, perhaps, liberal internationalist and yet descriptivist. I myself would probably count as upper left quadrant - while respectful of the new descriptivist methods, still essentially a moralist in approach, yet a moralist committed not to liberal internationalism but instead democratic sovereignty.

My purpose is not at all to "peg" anyone anywhere that they would not self-describe - which is partly why I thought the project so suitable for an interactive wiki, in which international law scholars could be invited to locate their various articles themsleves. My purpose is genuinely descriptive, to try and find a new way to describe and represent the methodological world of international law scholarship. I should like to test whether the effort to "spatialize" the multiple vectors of international law scholarship proves useful in describing the field. I am more or less convinced that it helps, if it hedged by enough caveats - the most important being that simply putting positions on a line does not, by the mere act of putting them on a line, separate one position from another merely as a matter of degree and not kind. And quite possibly, the positions described here do not fit "along" a line at all - it may impose a certain picture that is in fact quite misleading. Nonetheless, I persist in thinking that it is a useful approach at least for these two axes.

I have now thought to add a third axis. This one is much more difficult to conceptualize as an axis, and that quite possibly because the positions I propose to stick thereon do not really fit along a line with each other. Nonetheless, let me try it out:

Spatial axis:
The exogenous versus endogenous explanation of international law methodology axis.

What do I mean by that less than transparent title? Exogenous versus endogenous? I mean by this the traditional distinction between explanations that draw from within the subject matter to be explained - explain on its own terms - and those that draw from outside the subject matter to explain it in terms of something else. One might contrast, for example, doctrinal explanations for explaining the law, explanations that draw upon legal doctrine itself to explain the law, and which thereby assert that the law has its own explanatory power - versus explanations that assert that you can only understand the law, or some aspect of it, by looking to some other explanation, such as economics, psychology, morality, etc., whatever that exogenous source of explanation might be.

Of course, this distinction is well known and applied in many fields besides law. Absolutely nothing new here - which I why I think it is useful to apply to international law scholarship and method. At the center point of the axis, we might put positive law and doctrine, as the essence of engogenous explanation. Then moving slightly further out, or perhaps embracing positive law and doctrine, we can put the Legal Process school, as one which is still fundamentally endogenous. We might also put here all the various clinical methodologies in international law, considering that they operate, behave, at least, as though the law really were the moving agent, the source of explanation for why things behave they way they do. Then, moving further out, we can begin to take on more exogenous forms of explanation, reaching outwards to legal realism, and from there to "critical" theories of law that do not necessarily explain law through law at all, but see it as the floating superstructure atop something else entirely.

What are some of the explanatory theories of international law that might fit on this line? Positive international law; advocacy and clinical methodologies that rely on legal process being taken seriously; legal process school; law as policy school; various IR theories that essentially explain law exogenously through game theory of power relations; sociological theories of international law, such as that being developed by Ryan Goodman and Derek Jinks, that seek to explain, at least in part, the binding force of international law through socializing forces; feminist legal theory; intellectual history of international law; intellectual history of international law as a profession (Koskenniemi); legal critical theories such as CLS or critical race theory; and critical theory more generally.

Certainly there are others. And particularly in this axis, I am concerned that the grounds for thinking that they can be ordered serially along a line are very weak and perhaps quite counter-productive. Moreover, I would also query whether, even if you think that it is possible to put these points serially in some fashion, this line has "midpoint" that corresponds in any meaningful way to the "midpoint" of the sovereignty axis or the normative-descriptivist axis. (One solution to that might be to order the line starting with postive law at one end, and getting more and more exogenous going further outwards - and then duplicate that ordering in the other direction of the line, so that the same positions and orderings appear for the remaining quadrants.)

It is, to be sure, dangerous to put points on lines. It has the effect of reifying the sense of differences of degree, not kind. And in some of these cases, there may well be no meaningful way to order them on a line - they are just different kinds of explanations. Still, I propose, in my upcoming paper, to suggest that the possibilities of insight through spatialization outweigh the disadvantages, provided it is taken as a limited tool.

Independent and Dependent Variables

Oona Hathaway and Ariel Lavinbuk, in their review of Goldsmith and Posner in the Harvard Law Review, essentially move to prise apart two conceptual issues in international law scholarship, what they call "rationalism" and "revisionism." Without wanting to put my labels on them, in some respects their "revisionism" matches up to my horizontal axis of sovereignty - the revisionism at issue being sovereignty versus liberal internationalism. Likewise, their rationalism somewhat matches up to (some of) the descriptivist positions I have put on the vertical axis of descriptivism/prescriptivism. And, as they comment, there is a perennial but unjustified conflation of the two positions, just as, we could add, there is a perennial but unjustified conflation of the normative and liberal internationalist position.

The question that arises out of the separation of the horizontal and vertical axes, then, is the relationship of the two axes once they have been prised apart. As axes, they can of course be conceived as variables, and the question that arises is whether they are (i) two independent variables, (ii) two dependent variables of a third independent variable, (iii) the sovereignty variable independent and the methdology variable dependent, or (iv) the methodology variable independent and the sovereignty variable dependent.

The big reasons why I find it helpful to spatialize the axes, in other words, are, two. First, to show that the two can be prised apart and are more accurately thought of as separate issues, separate questions. Second, to be able to frame methodically the question of independence and dependence. (Note that I have deliberately put the methodology axis as a vertical axis, in order to avoid the automatic reflex to assume that if it is horizontal, it is intended as the independent variable. I want to regard all four possible relations of independence/dependence as genuinely open.)

And here we have a first contrast of positions. The Hathaway-Lavinbuk position is, essentially - if I am wrong, I invite them to correct me, but I believe this is correct - that what they term revisionism is independent of rationalism - and vice versa. The two variables are independent, that is, (i) above. By contrast, at least as I read Goldsmith and Posner, and understand them to accept (and I invite correction on this), they see rationalism, rational choice and game theory, as fundamentally forcing the position on the sovereignty axis - in other words, they hold (iv), the methodology variable independent and the sovereignty variable dependent, at least (and this condition is important) insofar as the position on the methodological line is true.

The primary claim of Goldsmith and Posner can thus be restated thus: If (and only if) the methodological point adopted - rational choice in their case - is a true description of the world, then the range of true positions on the sovereignty axis (the revisionism variable) is forced (to the sovereignty side of the line). But if, on the other hand, one chooses a methodological position that is false, then they make no claim as to forcing a position on the sovereignty axis, and indeed make no claim at all.

(And note that this truth-value claim is crucial to the kind of analysis adopted here. Unlike in, for example, economics typically, or many other parts of science, the fact that I have arranged the positions according to some non-arbitrary, some partly meaningful criteria along a line (or so I hope), does not make the positions simply different by degree. On the contrary, there remain significant differences of kind at least possible. The crucial implication is that a move along a line, a change of position along the line - even from one adjacent position to another - might move all the way from true to false as a description of the world. Thus, when describing a change in a presumed independent variable, it does not necessarily produce a corresponding change in position of the presumed dependent variable - because the change in independent variable might have moved from truth to falsity, leaving no implication for the dependent variable. This is part of the risk of using this kind of spatialization - the assumption that movements along the line are all movements along changes of degree, not changes of kind, and not changes of kind that involve moving from a true proposition to a false one.)

Where Next?

Even with these cautions and warnings about presuming too much about lines and variables, I suggest that this kind of spatialization has some clarifying value to help frame the debate about explanation in international law scholarship. It is not rocket science, obviously - but that is part of the point. I have deliberately sought relatively uncontroversial propositions about method in order to seek to construct a model that might receive relatively broad agreement as a framework for debating the questions of what is independent and what is dependent, if anything. I am seeking to eschew theoretical commitments in it, but instead seeking to provide a frame.

However, I have not sought in this post to do the more difficult, and much more fraught, task of defending a certain, or any particular, ordering of the exogenous-endogenous axis, or how it relates to the methodological and sovereignty axes.

Nor have I sought to explain where I would locate in this discussion the traditional axis of realism and idealism in international relations - although I would say briefly that conceived as an explanatory proposition, which is why we care about it here, it essentially comes in as an explanatory position largely exogenous to international law, and hence a position on the exogenous-endogenous explanation axis. And expressed that way, located there, on that axis, it becomes another way, perhaps, of explaining why realism and idealism have never been central to methdology in international law (without expressing a view on whether that is a good thing or not).

(I'm delighted to say, too, that Jose Alvarez has kindly invited me to participate in a panel discussion on scholarship issues in international law at the joint AALS-ASIL meeting in Vancouver next June. That discussion will be focused on more "critical" questions of scholarship, the academy, its relation to the world of international law, advocacy, etc., but perhaps some of these concepts will still be relevant. I will also post something about curricular revision at Harvard Law School, adding multiple international and comparative law components to the first year curriculum. In particular, I want to address not the usual public issues that I blog about, but instead the international economic law issues that I teach - what on earth should the IBT course cover, for example?)

I very much welcome comments on this project, either in the comments, or else to my email at school: kanders@wcl.american.edu.

(Note: This post originated back in the spring, when I wrote a post on blogospheric scholarship, responding to a Harvard Law School conference discussion on blogs and scholarship, and then extended in discussions at Opinio Juris, (my original post here). It suggested a certain kind of "wiki" scholarship appropriate to blogs and online work because it would allow collaborative postings that would allow scholars to self-identify within the "framework" I have been tentatively urging above. I have been spending quite a lot of time between other, more mmediate projects thinking about the methodology questions in international law scholarship, and wanted to add something else about it beyond that original post. My plan is to get a paper out that proposes this framework - and then to somehow get the tech people at my school, or somewhere, to set up a modifiable web site with cool graphics to be able to show this graphically. I'm not so good at the tech stuff.)