Theory Talk #12: Robert Jervis

Thursday, July 24, 2008

Robert Jervis on Nuclear Weapons, Explaining the non-Realist Politics of the Bush Administration and US Military Presence in Europe


Theory Talks proudly presents a Talk with Robert Jervis. While Jervis is one of the principal representatives of the Neorealist current within IR Theory, he has gained a lot of attention for his cross-breeding with psychology and game theory.


In this Talk, Jervis – amongst others – clarifies some of the more complex theoretical issues at stake in IR, criticizes the Bush policy from a Realist perspective, and helps to explain the failure of the Treaty on Nuclear Non-Proliferation.


Print version of this Talk (Pdf)


What is, according to you, the biggest challenge / principal debate in current IR?


In my opinion, the two principal debates are quite old.


The first debate is the one that Kenneth Waltz (Theory Talk #40) developed in his book Man, the State, and War (1959) and treats the causal value of the three different levels of analysis: on which type of factor should you focus in order to explain outcomes of the most important questions in IR?


I think a lot of current debates still struggle with this underlying question of on which to focus, having the choice between (1) the individual level; (2) the domestic level; or (3) the international level. The first level, called the ‘first image’ in IR, focuses typically on human nature or on the beliefs of individuals; the second image involves the nature of the state, its social system and domestic politics; the third focuses on the general characteristics of international system as a whole and on the more specific aspects of the state’s external environment.


Any explanation can, of course, combine elements of different levels, and there are questions we cannot answer within any of these levels, but still a lot of discussion can usefully be framed by which level we should concentrate on. Realists tend to work either with the first or third image, while liberals of course are typical second image thinkers, because they say that a certain type of domestic arrangement will lead to better international behavior.


Also, it is very important to look which level decision makers work with: I think one of the keys to understanding the Bush foreign policy is that he and his staff are clearly inclined toward second image beliefs. They’ve been quite explicit about the fact that what they care about is the nature of others’ regimes: they divide the world into democracies and tyrannies, which they think explains their behavior and therefore strong influences how the Bush administration behaves toward them.


The second central problem is that of endogeneity. It has only recently been treated explicitly but, again, is one with which other generations of scholars have grappled. Two aspects are relevant.


First of all, we as IR scholars are theorizing about people who hold their own theories – take Bush again, who doesn’t explicitly state ‘I hold second image beliefs’ or put this in a footnote in his National Security Strategy document, but it’s quite easy to dig it out of his statements and behavior. Now this brings up a big problem: can Realists explain the behavior of decision makers who are not Realists and subsequently don’t believe in or act on but reject Realist beliefs? Reagan and Gorbachev in seeking to abolish nuclear weapons clearly were not Realists. So we’re not only theorizing about people who act, but the actors whose behavior we’re trying to study have their own theories, which are of the same type as our theories.


The second part of this problem is that of untangling causal effects. Do alliances cause war? Or does the fear of getting into a war cause alliances? Do arms races lead to war or does the threat of war lead to arms races? Are these kinds of causal relations reciprocal? The events we see are based on the behavior of states that are anticipating what other states are going to do. When going through our arguments, whether they are about specific questions or generalities, we have great problems with convincingly arguing for a pattern of unidirectional causation, leading to continuous challenges that no one has ever formulated a definitive answer to. But being aware of these problems is quite helpful in moving us along.


What is your position or answer to this challenge / in this debate?


As critics have already pointed out, I’m not particularly consistent, which doesn’t actually bother me. I don’t think there’s any answer in these debates, but you have to be aware of what you are doing: you don’t want to mix too many different types of explanations. States, even when they’re in the same situation, behave extremely variedly, so we have to look at the values states hold. I think that it’s perfectly legitimate for the scholarly community and individual scholars to work with different levels of analysis to answer different questions. I wrote a book about system effects, and that means mostly the third image, but I don’t have any problem working with cognitive psychology in order to answer a different range of questions, for example why Bush adopts this foreign policy I completely disagree with.


How did you arrive at where you currently are in IR?


Well, I’ve been at it for over forty years, so that’s a long story, but I’ll try to make it short. My interest in perceptions grew out of the Cold War, which I was born into; and in graduate school, I got interested in the debate between what I later labeled as ‘spiral theorists’ (who take it that people are better off resolving conflicts) and ‘deterrence theorists’ (who take it that a defender of the status quo has to use threats to enforce deterrence) over the big issues of nuclear strategy and Cold War policy. These theories not only dealt with analytical questions (‘how to understand the Soviet Union’) but also with prescriptive questions (‘how to deal with the Soviet Union’), and I came to see that they turned on different images of the USSR. When I realized that, I decided to get into political psychology, where I focused on what states perceive and how they signal their own intentions – they want others to have a certain view of them. Signaling is basically the topic of my first book, perceptions are the focus of the second, and two subsequent ones argue that nuclear superiority doesn’t matter –according to deterrence and the nuclear revolution, having more guns won’t help you. In these latter books, I was influenced not only by my theoretical dispositions but also by the consulting work I did for the Central Intelligence Agency (CIA) that gave me access to the arguments for why we needed more weapons and nuclear options, arguments I found no more informed or sophisticated than those used in public.


So the Cold War has been a red line through my work; and when it ended, I kept on applying what I’ve learned about international politics to the changes in the US foreign policy. I think, in the end, it’s a dual struggle to be an IR scholar: you struggle not only with theoretical debates amongst peers, but also with what the newspaper tells you every day. I still go through the experience daily of asking myself ‘now why did they do that?!’


Now in terms of individuals who inspired me, I would name some obvious people like Kenneth Waltz, Tom Schelling, Glenn Snyder; and then there is one scholar who’s not as well known as he should be: Arnold Wolfers, who was I think the most sophisticated, subtle, and well-grounded of the early generation of Realists.


What would a student need to become a specialist in IR?


I don’t think there’s just one answer to that question, but I do think you need a good grounding in general social science, a lot of political science, and international history. And above all, you have to be willing to change your mind when you find new arguments or events that contradict your deeply embedded ideas. So you basically need to develop a combination of stubbornness (you need to follow your own ideas out and stick with them for quite a while) with the ability to listen carefully to what others say and to take seriously new evidence. So you ask me: how to get there? That’s the big question, which unfortunately I cannot answer. Things like choosing a dissertation topic are extremely hard. Fritz Stern, a colleague of mine, once replied a student who came to ask him for help with his topic: ‘young man, I would rather choose your wife for you than your dissertation topic!’


You’ve written a lot about perceptions and misperceptions, for example your 1976 book Perception and Misperception in International Politics, in which you focus on foreign policy, according to realists like Mearsheimer an essential complement to structural theory. But Mearsheimer insists that structural realism doesn’t deal with perceptions. Does that make you a social constructivist avant la lettre?


To some extent, absolutely: some twenty years ago I gave a talk, and a student came up to me afterwards, saying: ‘so you’ve become a constructivist now?’ I responded: ‘Actually, it’s the other way around.’ My first book, The Logic of Images in International Relations, is a combination of what we would now call social constructivism and rational choice theory. They have much more in common than either one wants to acknowledge.


If social constructivism is to develop seriously, it has to be grounded in psychology. The importance social constructivists attach to peoples’ ideas, beliefs and values should translate into attention to the field that deals with these subjects par excellence: psychology, especially the subfield of social psychology, since like constructivism it is social as well as psychological.


Unfortunately for all of us, the field of social psychology, which thrived between the thirties and the sixties, has become much less robust as sociology and psychology diverged. If a social constructivist wants to make serious claims, I am convinced he or she has to draw upon psychology. Up to this point, most do it either without knowing it, implicitly or badly – or they don’t want to acknowledge their indebtedness to another field, which is of course understandable for a theory struggling to establish itself. Every school wants to exaggerate the degree to which it’s new.


Due to the economic problems of the US, there are stories being heard about the decline of US’s relative power. Under what conditions will American hegemony end and do you see this happening any time soon?


As a Realist, the only condition under which I see American hegemony end is when another state reaches the power of the US – without that, there just isn’t a challenger. I don’t like this, but despite all the internal problems the US faces right now, I just don’t see any challenger rising soon. The idea of China as a global challenger to the US is laughable; it’s simply too weak for the coming fifty years and there’s even the chance that it’ll collapse into civil strife. And the economic crisis is affecting other countries just as much as – if not more then – the US: the world economy is strongly interdependent, so if the US crashes, everyone else does.


There’s a large American military force in Europe (116.000 personnel, according to Global Research). Your Realist colleague John Mearsheimer argues in his 1990 article Back to the future: instability in Europe after the Cold War, that this American military presence in Europe guarantees European peace, and predicts that Europe would fall victim to internal strife between Germany and other states. Is that really how Realists think about the viability of European integration? Put differently, is it, from a realist perspective, really in US interests to prevent Europe from becoming a ‘billiard ball’ and thus a unitary actor that could balance against the US?


This article of Mearsheimer is a great example of taking the Realist logic until the end, and reaching a crazy conclusion. But it’s useful to think about why it’s a crazy conclusion. I do believe that the military presence of the two big powers in Europe during the Cold War was necessary to the European unification we’ve seen. But I think we could now take away our troops without European countries fighting each other – for a variety of reasons, nuclear weapons being one of them. Although I don’t think Europe will unite in my lifetime, it does form a robust version of what Karl Deutch called a ‘security community’ – war is unthinkable between them – but I’m cautious: we could withdraw, but I see no reason to run the experiment, and I think European diplomats feel the same way. Even if it’s just a five percent chance that Europe goes bad, I think American military presence is a small price to pay to avoid that. And keeping some troops in Europe doesn’t cost much more than keeping them in the US, especially because we get some payments of the Europeans.


Also, there are different types of Realism, or rather, theories that take different parts from Realism. There’s the Realism of neoconservatives like Paul Wolfowitz and Dick Cheney who wrote the Draft Defense Planning Guidance (1992), where they explicitly stated that the US should inhibit the rise of the peer competitor – that meaning, of course, Europe – because history tells us that that leads to instability and dangerous rivalries. While that is clearly a form of Realism, my Realism makes me wish Europe would unite and balance US power. I want to emphasize that I would welcome a unified Europe, which would function as a balance against the US and perhaps stop it from doing foolish things. But again, I don’t think it’s going to happen.


You’ve argued that every little thing that happens in every part of the world affects American interests, just because the US is everywhere. Literally: ‘Where's a part of the world we don't care about? Doesn't exist.’ Africa, full of failed states that can potentially harbor terrorists, certainly seems like such a place. So why not intervene in Sudan with more engagement, or in the whole of Sub-Sahara Africa with something equal to a Marshall Plan?


You’re right; we’re not doing a tremendous amount if you look at what we could do. But if you look at it in historical perspective, we are doing a lot. In the 19th century, Britain cared about Sudan because of Egypt, and France cared because it was a way to pull the lion’s tail. Then, no other great power cared about Sudan, while now everybody cares about Sudan. Even if we don’t send troops, I think that’s already a big difference.


And if you ask: why no big Marshall Plan? I think no-one has the confidence that that would work – except maybe for Jeffry Sachs. If people really had faith that large amounts of foreign aid would work, the public would support or even demand it. In the US, public opinion holds that we spend too much only because the estimates people have are incredibly distorted – they think it’s one of the biggest items in the Federal budget.


Nuclear weapons represented the biggest threat during the cold war, and now the big issue in US foreign policy is nuclear weapons in the hands of terrorists. The Treaty on Nuclear Non-Proliferation has been signed exactly 40 years ago, and disarmament hasn’t really taken off; India is even becoming a nuclear power under US auspices. How would you explain this from a realist perspective? Is it a prisoner’s dilemma? And, if it doesn’t matter if a country is ‘friendly’ or not, why did the US let India have the nuclear bomb? Why not let Iran have it as well?


I think non-proliferation is a prisoner’s dilemma from the standpoint of a lot of potential proliferators: in large parts of the world, countries want nuclear arms if their neighbor has them, but they’re willing to follow the Treaty if they get guarantees that their neighbors do so as well. As such, the Treaty has worked fairly well.


But the US under Bush is clearly seeking to change the rules. This goes back to his second-image convictions: for Bush, India behaved well, as it is a democracy. Iran and North Korea, according to the Bush Doctrine, have to be treated more strictly than the Treaty says, because they behaved badly – and again, the reason for this is their domestic regimes. What he’s in effect doing, is saying: ‘no, the idea of the same rules applying to everyone is wrong, because not all countries are the same’. He wants certain standards of justice to be applied to liberal democracies and different ones to ‘bad’ regimes.


According to you, we should analyze the US from a realist perspective: ‘it’s behaving like a normal superpower’, you’ve stated in your Conversation with Harry Kreisler. But would a realist approve of the US policy in Iraq?


With a group of Realists, I took a stance by placing an add in the New York Times in the run-up to the Iraq war, saying there is no need for this war and that the reconstruction would be extremely difficult. Yet again, here we touch on the question of whether and how Realists can analyze the behavior of countries whose leaders do things that Realism sees as really very foolish.


Classical Realists go back to human nature, to argue that if both individuals and collective actors like states are not externally restraint by other strong actors, they will in the end be unrestrained and go off to do wild and foolish things – “absolute power corrupts absolutely”. That’s also the point of departure for the Constitution of the United States and of the balance of power in international politics. Kenneth Waltz, towards the end of the Cold War, wrote an article in which he predicted that the US would do very foolish things, just because it’s got the power to do so – which is why I think that Realism can explain the basic tendency for states to do things that Realist scholars disapprove of.


But what’s so frustrating, is that the bulk of the scholars, both those who approve of American policy, in the States and in Europe (yes, they exist in Europe too!), and those who disapprove, tend to use second image arguments, that is, ones based on the domestic configuration of the US, be it the religious factor, the presence of the neoconservatives, or economic pressures. And these people don’t like Realism because it undercuts their arguments and claims that other states similarly positioned would act (and have acted) similarly. By simply pointing to the sheer size of the United States and the lack of any state big enough to balance it, all this talk about domestic factors becomes unnecessary. But I think that in this way, Realism does shed light on behavior that scholars condemn.


Last question. When invading Iraq, we found out Saddam bluffed about having nuclear arms, most probably to keep his neighbors out. Couldn’t Iran be bluffing as well?


The narrow answer I would give is that Iran could be bluffing, although I doubt that it is. I’d note that the Shah’s regime wanted nuclear weapons; and that there are a lot of reasons to expect a country like Iran to want them under current circumstances.


The broader issue is about intelligence and knowledge. I worked US intelligence on this and think it is clear that intelligence really thought that Iraq had weapons of mass destruction, and the Bush administration just made that threat look a little more certain. Saddam’s view of the world was really crazy: up until the very moment we invaded he didn’t worry about the US, thought we wouldn’t invade, and was more worried about Iran and a Shiite rebellion and his own generals than about us. How can it be that our threats were not believed? Saddam is yet another example of unforeseeable behavior on the part of States or decision makers: he committed suicide, not only figuratively but literally – something that is nuts for Realists. A regime should not commit suicide; if something like a National Interest exists, committing suicide is definitely not part of it. On the other hand, a lot of what he did – seeking to increase his power and to protect himself – fits well within Realism. So his case is really puzzling unless you get deeply into how he saw the world.


The big challenge both for IR scholars and for governments that this raises is that states who interact closely and exchange messages on a daily base can still live in their own perceptual and conceptual worlds. That’s why I think IR often resembles not poker but the Japanese short story and movie, Rashomon.


Robert Jervis is the Adlai E. Stevenson Professor of International Politics at Columbia University. He is a fellow of the American Association for the Advancement of Science and a member of the American Academy of Arts and Sciences. He is a past president of the American Political Science Association and his publications include Perception and Misperception in International Politics, System Effects: Complexity in Political Life and Social Life, The Illogic of American Nuclear Strategy, and most recently American Foreign Policy in a New Era.


Related links


Who influenced Jervis

  • Read chapter 5 of Arnold Wolfers Discord and Collaboration (1962) here
  • Read Robert Aumann’s and Thomas Schelling’s Contributions to Game Theory: Analyses of Conflict and Cooperation (Nobel Prize Committee 2005) here (pdf)


Jervis’ work


  • Read Jervis’ 2003 article Understanding the Bush Doctrine (Political Science Quarterly) here (pdf)
  • Read Jervis’ 2005 article Why the Bush Doctrine Cannot Be Sustained (Academy of Political Science) here (pdf)

Print Version of this Talk (pdf)