Skip to content


A Reply to Josh Busby

Josh Busby engages my recent article “Unrest Assured: Why Unipolarity Is Not Peaceful” in a thoughtful post over at Duck of Minerva. He raises a number of interesting questions, which I hope will help clarify some of the points I made in the article. Specifically, I’d like to address three criticisms he offers:

First, Busby takes issue with my “premise” that unipolarity is not peaceful.

It’s not so much a premise of the argument as an empirical observation about the past two decades, which, as I point out in the piece, represent a disproportionate amount of the time the United States has been at war: “the first two decades of unipolarity, which make up less than 10 percent of U.S. history, account for more than 25 percent of the nation’s total time at war” (UA, 11).

In fact, Busby sees in my argument a claim “that unipolarity is not at all peaceful and much less peaceful than other periods” (my emphasis). But this is explicitly not what I argue. As I wrote: “Rather than assess the relative peacefulness of unipolarity vis-à-vis bipolar or multipolar systems, I identify causal pathways to war that are characteristic of a unipolar system and that have not been developed in the extant literature” (UA, 12). It should be clear that my piece makes no comparisons between unipolarity and other types of systems in terms of the level of conflict in each of them. Instead, it shows how the conventional view (largely based on Wohlforth’s work) that unipolarity is peaceful misses some important conflict-producing mechanisms particular to a unipolar world.

Still, Busby asks:

“How do we square this account by Monteiro of a unipolar world that is not peaceful (with the U.S. at war during this period in Iraq twice, Afghanistan, Kosovo) and Pinker’s account which suggests declining violence in the contemporary period?”

This question, according to Busby, leads to another: is my measure of peacefulness — the number of years great powers spend at war — the most adequate to conduct an “adequate test of the peacefulness or not of unipolarity”?

For Busby, a more adequate measure would be “whether the system as a whole is becoming more peaceful under unipolarity compared to previous eras, including wars between major and minor powers or wars between minor powers and whether the wars that do happen are as violent as the ones that came before.” This would indeed by a more complete test of the view that unipolarity is not peaceful. It is, alas, beyond the scope of my article, in which I wanted to achieve a much more modest goal: to show that unipolarity is far less peaceful than the conventional wisdom has it.

(Let me bracket the question of whether a conclusive test such as the one Busby proposes is at all feasible. It would certainly pose some major problems, given the need to disentangle the effects of polarity from those of other trends, such as democratization, the capitalization of the U.S. military, etc., as well as the need to separate those wars that occur within one type of polarity — e.g., Vietnam during bipolarity — from those systemic wars that themselves change the polarity of the system — e.g., World War II, from multipolarity to bipolarity. In any case, the important point is that such a comparison is well beyond the goals of my article.)

My argument that unipolarity makes room for particular conflict-producing mechanisms — involving the unipole in case it does not disengage from the world — is indeed compatible with different assessments of the overall level of conflict in the system, as well as of the lethality of those wars.

As Busby points out, it may well be that the deadliness of wars is declining. But this does not necessarily stem from polarity. I think it doesn’t. As I wrote in the article: “Some scholars might argue that wars in the post–Cold War world have been less lethal than those of the past. … It is difficult, however, to parse the role of polarity in a decrease in lethality. … Furthermore, part of the explanation for this decrease in lethality may lie in the U.S. decision to develop a highly capitalized military aimed at minimizing casualties” (UA, 19 fn. 48). In fact, other developments in the world make the recent incidence of war involving the United States particularly puzzling: “This sharp increase in both the percentage of great power years spent at war and the incidence of conflict is particularly puzzling given that the current unipole — the United States — is a democracy in a world populated by more democracies than ever before. In light of arguments about how democracies are (1) better able to solve disputes peacefully, (2) select only into those wars they can win, and (3) tend to fight shorter wars, this should mean that the United States would spend fewer years at war than previous nondemocratic great powers” (UA, 19 fn. 47).

In sum, in “Unrest Assured,” I am not making comparisons among different types of polarity, merely highlighting that unipolarity is not an unmitigated boon in terms of producing peaceful outcomes, rather it encourages particular types of conflict, namely those present in the mechanisms I lay out in the paper. I also make no points about the lethality of recent wars, which may well have been decreasing for factors unrelated to polarity.

One last clarification point before moving on, Busby notes that “In Monteiro’s world, disengagement would inexorably lead to instability and draw in the U.S. again (though I’m not sure this necessarily follows).” Nowhere in the paper do I argue that the instability created by disengagement would draw the unipole in again. Rather, I argue that dominance is not the only strategy of the unipole, which may also opt for disengagement, and lay out the conditions why, after a likely initial period of dominance, the unipole may decide to disengage from the world (UA, 21-22). After it does, it may subsequently decide to return to a strategy of dominance, but I do not make any arguments on the likelihood, much less the inexorability, of a return to dominance.

Second, Busby disagrees with my core argument that unipolarity is behind whatever level of conflict we have experienced during the post-Cold War era.

Busby writes:

“[I]n Monteiro’s view, the U.S. power position alone, even where the U.S. seeks to defend the status quo, is enough to generate conflict with “recalcitrant” minor powers. Here, “recalcitrance” seems to be cover for some domestic-level variables, either quixotic or idiosyncratic leadership characteristics by the likes of Saddam and Milosevic or attributes of authoritarian regimes. I’m not sure that U.S. power is doing the work for Monteiro. … Rather, I suspect that aspects of U.S. domestic politics … intersecting with domestic attributes of “recalcitrant” regimes are doing much of the heavy lifting. If we were or [sic] become different (practice restraint, focus on the home economic front for a bit) and if the regimes we face become less recalcitrant (post Arab-spring if we’re lucky, post-Kim Jong Il if we’re really lucky and something different in Iran if we’re really, really lucky), then unipolarity is not structurally determined to be violent.”

The main issue at work in this criticism is the role of domestic variables in my theory. If Busby is right, conflict is not being driven by unipolarity, but by domestic variables. Change their values and you’d have a peaceful unipolar world a la Wohlforth. Here’s the relevant section in my piece:

“The structure of the international system does not entirely determine whether or not a minor power accommodates the unipole. Still, structure conditions the likelihood of accommodation in two ways. To begin, a necessary part of a strategy of dominance is the creation of alliances or informal security commitments with regional powers. Such regional powers, however, are likely to have experienced conflict with, or a grievance toward, at least some of its neighboring minor powers. The latter are more likely to adopt a recalcitrant posture. Additionally, by narrowing their opportunities for regional integration and security maximization, the unipole’s interference with the regional balance of power is likely to lower the value of the status quo for these minor powers. As the literature on the ‘value of peace’ shows, countries that attribute a low value to the status quo are more risk acceptant. This argument helps explain, for example, Japan’s decision to attack the United States in 1941 and Syria’s and Egypt’s decision to attack Israel in 1973. In both cases, aggressor states knew that their capabilities were significantly weaker than those of their targets. They were nonetheless willing to run the risk of launching attacks because they found the prewar status quo unacceptable. Thus, for these states, the costs of balancing were lower relative to those of bandwagoning.”

As is clear from this excerpt, both systemic and domestic values play a role in determining which minor powers become recalcitrant. Systemic variables matter because the presence of a unipole — directly or indirectly, as a security sponsor — in any given region will have an impact on the prospects of states with which it is not aligned. The United States, by siding with some regional powers — e.g., Saudi Arabia and South Korea — hinders the prospects of others — e.g., Iran and North Korea — making peace less valuable for them. At the same time, to the extent these states have a history of grievance towards, or conflict with, the clients of the United states, they will have a hard time integrating and therefore will feel the predicament of extreme self-help in particularly acute ways.

Does this mean that domestic variables play no role in these relationships? Of course not. Domestic variables, such as Busby’s “quixotic or idiosyncratic leadership characteristics by the likes of Saddam and Milosevic or attributes of authoritarian regimes,” or whatever, may play a role. Since they are not determined by the distribution of power in the system, I left them out, however. The important part from where I look at it is that the configuration of unipolar alliances in the regions in which it engages reinforces these tensions that end up leading to the emergence of recalcitrant minor powers.

Perhaps the story would be more complete if I would also have addressed domestic variables, but I don’t think we need those to get the mechanism up and running. Frankly, I think this (i.e., determining which states will become recalcitrant minor powers) is an area where structural- and domestic-level theories are likely to reach similar conclusions. The more variables we throw in, the more complete our explanations will be. Of course Milosevic’s temperament, Saddam Hussein’s leadership style, or Kim Jong-il’s well-documented foodie habits may have played a role in driving conflict. But the fact is that the United States refrained from attacking any of the states they led while they had a great-power sponsor. As Tariq Aziz, Saddam’s foreign minister, lamented immediately after the war, “We don’t have a patron anymore. … If we still had the Soviets as our patron, none of this would have happened” (quoted in UA, 28) I think he was right. Polarity was the operative transformation in placing them within the reach of U.S. military power.

Third, Busby sees more of a connection between my theory and Walt’s balance of threat theory than I acknowledge in the piece.

So, what about balance of threat theory? How does my theory relate to Walt’s? For Walt, a state is a threat when it fares highly on an composite index of: strength/power, geographical proximity, offensive capabilities, and intentions. My point is that the strength/power and offensive capabilities of the unipole are such that, when combined with the limited capability other states have to devise its intentions, they make the United States a grave threat for any state in a region in which U.S. forces (or security guarantees) are present.

So the variation you would expect from Walt’s theory in terms of who is a threat and who is not is wiped out when it comes to relations between non-aligned minor powers and a unipole present in their region. Does this mean that Walt is wrong? No. It means that unipolarity is a situation in which Walt’s theory is not particularly useful because the unipole’s power, as it were, trumps the other dimensions that would produce variation in Walt’s world. (Colin Elman has posited this before, as I note in UA, 24, fn. 63.)

I hope to have addressed Busby’s excellent criticisms.

Be Sociable, Share!

categories: IR theory, national security, publications, research, war. | tags: , .

Posted at 8:15 pm


3 Responses

Stay in touch with the conversation, subscribe to the RSS feed for comments on this post.

  1. Konstantinos Travlos says

    Good response and clarifies a lot of things

    Some points

    1) You can conduct the type of study you referred to early on ((Let me bracket the question… In any case, the important point is that such a comparison is well beyond the goals of my article) using large-n ,COW, and Polity IV. There is a wealth of studies conducting such tests in JPR,JCR, CMPS. And after resolving what uni-polarity means you can create a measure.

    2) Why did you use years a state is at war as your measure. Both you and Busby talk too much about war duration. That is a bad measure of the war-propensity of a variable. War duration is probably more driven by specific in-bellum mechanisms and dynamics that have less to do with international factors, and even less with something as far off as polarity.

    Furthermore your theory does not tell us why the unipole should have long wars. Only that it will have more wars than otherwise because it is unable to credibly guarantee not going to war with a recalcitrant state. Not just more wars but more militarized conflicts with such states.

    War initiation is what you both are looking at, not war duration, let alone war lethality (that has little to do with international political factors) . To put it simply your unit of analysis is not the appropriate.

    If you do war initiation then per COW the US was part of 3 war initiations from 1950-1990 (Korean, Vietnam, 2nd Laotian ) and 4 post 1990s (Irag II,Kosovo,Afghanistan, Irag III)

    Which is a difference of only 1 war between the bipolar and uni-polar period. Which is a large problem for the face validity of your story. If polarity has a strong effect that difference is too small. Even a 50% difference 2 vs. 4 is probably too small.

    It doesn’t help that if we just look at Militarized Interstate Disputes (MIDs, conflicts that kill less then 1000 people) than the US participated in 167 MID initiations between 1950 and 1989 and 46 from 1990 to 2001 which means that in bipolarity the US on average was participating in a MID initiation 0.22 times a year, while in unipolarity 0.21 times a year.

    Again not a big difference. That is the main issue for me (as well as the point I made about the possibility of finding way to be credible thus resolving the issue of credibility)

    Konstantinos Travlos
    UIUC

  2. Nuno says

    Konstantinos,

    Thanks for responding. One quick point. You say:

    “If you do war initiation then per COW the US was part of 3 war initiations from 1950-1990 (Korean, Vietnam, 2nd Laotian ) and 4 post 1990s (Irag II,Kosovo,Afghanistan, Irag III)

    “Which is a difference of only 1 war between the bipolar and uni-polar period. Which is a large problem for the face validity of your story. If polarity has a strong effect that difference is too small. Even a 50% difference 2 vs. 4 is probably too small.”

    Again, I am NOT arguing that unipolarity is more prone to war than bipolarity. Only that it is more prone to war than you would think by reading the existing literature on unipolarity, which portrays the system as being peaceful.

    I hope I’ve made that clear. Thanks again.

  3. Konstantinos Travlos says

    Yes and no. The issue is that you have provided a possible causal mechanism that few thought of about before. That is a contribution, and indeed a very characteristic Chicago contribution. However if your mechanism actually only exerts a very minor influence then in a way you are driving another nail down the concept of polarity as an important variable. Such a clear causal process and yet possibly a minor effect.

    That said I think you have made the concept important enough to warrant large-n evaluation. Where I really think that the polarity concept can greatly help is at the level of how easy or harder different polarity systems facilitate or make harder the creation of norms and regimes that can provide an alternative to war. Your clear theory might indicate that it may e harder in uni-polarity then most people thought as you point out. That said the sad thing is that If my memory does not betray the polarity people are still too focused on capability distribution as opposed on regime creation.

    And I still think your theory is very good.



Some HTML is OK

or, reply to this post via trackback.