Spatializing the representation of international law scholarship (post 3 of 4)

Spatializing the representation of international law scholarship. Let me try to put this slightly more formally, by offering a spatial model of the intersection of method and ideology in international law scholarship. [I don’t have spiffy graphics yet, so please bear with me; this also draws on earlier posts on this blog.] My goal here is not to make a normative point about international law scholarship, but instead to offer a way of representing and comparing the work of different scholars by reference to their method and ideology, in a way that plainly separates the two concepts.

Indeed, I propose to separate them so much that they become two different graphical axes.

Imagine an X axis, horizontal, representing the range of positions between what I will baldly assert is the critical ideological question of international law scholarship – sovereignty or liberal internationalism. There is a range of possible positions here, running from sovereignty as a value for its own sake, sovereignty justified by the autonomy claims of a democratic community, multilateralism among sovereigns, multilateralism that pools sovereignty, liberal internationalism, and finally genuinely global federalism. (I will not elaborate on this range here; I discuss them in a couple of earlier articles – in my review of The Limits of International Law, and in my review of Slaughter’s A New World Order.)

Now imagine a vertical Y axis, representing a range of methodological positions with prescriptivism at the top, above the X axis, and descriptivism on the bottom. Within the range of methodological positions we might put at the top of prescriptivism moral idealism – roughly the view that the content of public international law ought to be strongly informed by a moral vision of that law, whatever precisely that moral vision might be. At the bottom we might put two different descriptivist alternatives – rationalism of the kind that leads to deductive rational choice and game theory, and empiricism of the kind that leads to, for example, Oona Hathaway’s empirical work on the effects of human rights treaties on human rights outcomes.

[Clever graphical insert]

(In a full modeling, we might include a third Z axis – one corresponding to method in a different sense, not the sense of prescriptive versus descriptive, but something. That sense might be how closely, or not, methods of international law are endogenous or exogenous to the legal materials of the law itself. They might include such methods as the legal process school, positive or doctrinal legal analysis as examples of endogenous methods and, as examples of exogenous method, feminist international legal theory, various schools of IR thought that look for non-legal explanations for movements of international law, or, most strikingly, Koskenniemi’s simultaneously endogenous-exogenous study of international law as a profession, at once internal to legal thought but also exogenous in the form of sociology (some brief commentary on his work, here, plus a reference to his latest, very interesting paper on SSRN, here).

(For that matter, one might also imagine an axis that looked to whether the method in question was fundamentally legal, political or social theory in its explanatory focus, in what it took to be theoretically revelatory. I will leave aside the complications of a third, or fourth, axis in this discussion, however.)

There are important skeptical questions one might ask about these proposed axes. They have the virtue of clarifying by separating. But do they also obscure? Are these the right ways to characterize the debates over ideology and methodology? And even if they are, by setting them out as axes, they create ranges of positions – whereas these positions might not be well-represented as linear points as though in a range.

These are all legitimate questions and quite possibly fatal objections. Nonetheless, I persist with this spatial, graphical model to see if it seems to illuminate anything about the current state of international law scholarship. I feel most comfortable representing the sovereignty-liberal internationalism positions on a linear range, because that seems accurately to capture the sense that more international federalism equals less sovereignty and vice-versa, at least as sovereignty is traditionally understood. I also am reasonably comfortable putting prescriptivism and descriptivism as opposite points on a line – although it seems also true that empiricism and rationalism are really at the same point on this line on the descriptivist side. Why? Because although many people in their actual work separate the two, in fact when it comes to accepting conclusions as factual, ordinarily we want both an inductive, empirical confirmation that the world conforms and a deductive explanation that tells us why that should be so. We want them both.

If, however, even for discussion’s sake such a coordinate grid is accepted, what might we learn out of it, if anything? Suppose we think in crude terms about where we might locate one or another figure in international law scholarship. In fact we should proceed article by article or book by book, but for illustration’s sake, let’s proceed with a couple of scholars.

Upper right hand quadrant. Here we locate scholars who are both prescriptivist in their method and liberal internationalist in their ideology. For example, Louis Henkin, Henry Steiner, and Harold Koh, and a great many more besides, especially of the last two generations of international law scholars.


Lower left hand quadrant. Here we locate scholars who are descriptivist in their method and sovereigntist in their ideology. For example, Goldsmith and Posner.

[Another clever graphical insert]

Those two quadrants essentially describe the state of discussion up until the past couple of years. They also summarize much of the confusion between ideological and methodological commitments that have characterized the debate. But increasingly, the other two quadrants have much of the most interesting argumentation:

Upper left hand quadrant. Here we locate scholars who are sovereigntist by ideology but whose methodology is prescriptivist or normative. I would include myself here – someone whose argument for democratic sovereignty is essentially a moral argument, every bit as much as, say, Louis Henkin’s argument for a liberal internationalist legal order is a moral one.

Lower right hand quadrant. Here we locate scholars who are liberal internationalist by ideology but whose methodology is descriptivist. This is where much of the intellectual ferment in American international law scholarship is taking place among the younger generation – scholars who are able to engage in the discussion of IR theory, game theory, rational choice, empirical and statistical studies, but who wind up favoring liberal internationalism rather than Goldsmith and Posner’s sovereignty position. Sometimes – certainly not in all writings – one might put in this category Slaughter, Hathaway, Lavinbuk, Raustiala, Guzman, etc.

[Still another clever graphical insert]

Does this tell us anything? Well, maybe it didn’t need all this graphical machinery to do so, but at a minimum it tells us that you cannot read off ideology from method or vice-versa. And that much of the most interesting work is being done today with liberal internationalist commitments but rationalist methodology, Goldsmith and Posner notwithstanding. Which is to say, these scholars argue in effect that if Goldsmith and Posner had captured rationalism in a more sophisticated fashion that better models the real world, their results would have been quite different from what their book argues.

It also bears noting that this two axis representation likely leaves many non-US scholars wondering where, or if, they fit in this representation at all. The reason is that in order to address what, for many of them, is their scholarly method – positive law analysis or doctrinal analysis – we need to add the third axis that I have here left aside, the axis that takes into account endogenous versus exogenous explanation in international law.

Independent and dependent variables. This brings us back more formally to a problem raised earlier in connection with Goldsmith and Posner. The separation of issues onto axes, mapping of the intersections between them, does not, by itself, address the question of dependence and independence. A theory of international law, , might assert such a relationship, either one of correspondence or of causation - as Goldsmith and Posner’s does – as a consequence of the internal workings of the theory. But it might not. Hathaway and Lavinbuk, in their review of Goldsmith and Posner, call for a recognition that method and ideology are independent variables. But of course for any particular theory of international law, they might not be – they might stand, on the contrary, in a precise relationship of dependence.

The point is, Hathaway and Lavinbuk are correct to say, there is no necessary relationship of dependence or independence; it depends entirely on the internal claims of any particular theory. And again, I would suggest that some of the argumentation that takes place today within academic international law occurs because of prior assumptions about necessary relationships of dependence or independence of method and ideology, which might be true of any particular theory – or might not. The clear separation of what is at stake makes it easier (maybe) to test whether a theory of international law asserts dependence or independence, and with respect to what.