Kenneth Waltz Interview: Conversations with History; Institute of International Studies, UC Berkeley

Theory and International Politics: Conversation with Kenneth N. Waltz, Adjunct Professor of Political Science, Columbia University; February 10, 2003, by Harry Kreisler
Photo by Jane Scherr

Page 2 of 7

Being a Political Theorist

Before we talk about your first book, what exactly does a political theorist do?

Thinks. There are two kinds of political theorists, really. One is a political theorist who writes about other people's political theories. And that, of course, in the traditional political theory field, is what is done mostly. It's a reconsideration of Hobbes on this point, or Locke on that point.

The other thing, if one wants to try to develop theory or promote its development, is to think about ways of doing that. [My] first book was really a sorting out book, and the second book was an attempt to develop a theory of international politics in a sense in which it had not been developed before.

Before we talk about those two books, help us understand, first, how does a person prepare to do theory? What kind of curriculum should one undergo to do that in a satisfactory way?

If you want to deal with theory as theory and not as the history of political philosophy -- which, incidentally, is a part of the field that I love, and it's a marvelous literature, but I didn't feel inclined to rehash it -- if one wants to develop theory, the direct route is to read a good deal of philosophy of science and take courses. Good courses are available in the philosophy of science. The first requirement, if you're going to work on a theory or the improvement of somebody else's theory, is to figure out what a "theory" is, which very few people seem to do. I've spent a lot of time reading the philosophy of science, because it's a very difficult question: What is a theory? What can it do? What can it not do? How do you test its validity or seeming validity? It's a profound and difficult subject in its own right. It also is a field in which there is great literature, and it was a pleasure for me to read in the philosophy of science, and not to have to read a lot more political science.

Are you allowed to say that as a former president of the American Political Science Association?

Well, I do!

So what does a theory do?

First, in order to have a theory, you'll have to have a subject matter, because you can't have a theory about everything. There's no such thing as a theory about everything. So you'll have to say, " I'm going to try to develop a theory of, in this case, international politics." The first question is, how can you think of international politics as a domain in its own right, as something that you could possibly have a theory about?

And how do you decide that you can do that?

You will have to figure out a way of defining it as an autonomous field of study. The closest comparison is the development of economic theory, where, before the physiocrats (that is, before roughly 1760), economists wrote about all kinds of things, and mostly at the level of bookkeeping, of what we might now call accounting -- family and business accounts, that sort of thing. It was only with the physiocrats, who greatly influenced Adam Smith, that the concept of an economy as something that could be studied in its own right developed. Once that concept existed, then it became at least possible to have a theory about how national economies work: what regularities appear, what repetitions occur, how you can think of it as a self-sustaining enterprise. The breakthrough is the physiocrats, and then the great follow-through was Adam Smith.

So if you're going to do theory of international politics, then at one level you probably have to be grounded in the history of the domain, so to speak.

I don't see how you can do it without knowing a good deal of history. But the main thing is to have a conception of international politics as something that can be studied in its own right. It's something that, for example, two major figures in the field -- Hans Morgenthau and Raymond Aron, a Frenchman -- thought was impossible. How can you isolate international politics even for the purpose of study from everything else that goes on? Their answer was, you can't. In that case, it's not possible to develop a theory of international politics. International politics is something that's influenced by everything else -- a national economy, national politics, international politics -- and it's all interrelated, there's no way of separating it. So the first requirement was to develop an idea of the structure of international politics, which would make it possible to think of international politics as a subject matter that could be studied in its own right. That's what I did in The Theory of International Politics.

Now, how does one then evaluate a theory? Is "usefulness" a good way to evaluate a theory? How do you know when a theory is useful?

Whether or not a theory is useful is decided by the body of people who find it worthwhile to use the theory or to argue about the theory. As Steve Weinberg, who's a Nobel Laureate in Physics and a very reflective physicist, has said, ultimately, the test of a theory is that people (meaning the people in the field) find it worth dealing with, arguing about, criticizing, trying to apply.

And the purpose of a theory is to explain what's going on -- how the order hangs together. Clarify that for us.

What a theory does is present a mental picture of a part of the world, and in that picture are identified the major causal factors at work. The theory specifies the relations among those, and the necessary relations as they're necessary within the terms of their theory, among those major causal forces, which we often now refer to as variables, (adapting a scientific terminology that's not always useful). That's a simple way of putting it.


Then you can compare that picture and the supposed causal forces at work with the real world. That's always a problematic exercise, because theory is very simple. What theories do is leave most everything out. You're simplifying, you're looking for what is salient, what are those central propelling forces. Obviously, they're not the only forces at work, so you've got that problem that natural scientists have, too; but they can usually, or often, control for perturbations that come in from outside the system. Whereas in politics, in international politics in particular, you can't. So it makes for endless argument and a lot of fun.

The theory has an economy about it, because at one level you can say, "Well, there's so much going on in the world, let's put it all into theory." But that's not the game here. In other words, there's not a one-to-one relationship, this not an effort to replicate everything that's going on.

If it were, then the ideal theory would be identical with the real world, right? And, instead, theory is a simple instrument which you hope to be able to use in order to understand and explain the real world. The emphasis is on explanation, not on prediction. Prediction is nice. If you can predict, fine. But the key requirement: if a theory is not able to explain what's going on, then it's not theory, or it's a worthless theory. It's not a theory at all.

Next page: A Theory of International Politics

© Copyright 2003, Regents of the University of California