“Speak Victorian, think Pagan.” Kaplan’s clever phrase is also the motto of Coming Anarchy.

I wonder, though, if Curzon takes that maxim as seriously as he should.

In his June 20 post, “Thanks for Nothing”, Curzon writes:

Thanks all of you who pushed for the US to break with Uzbekistan after the Andijan massacre that left hundreds dead. You got what you wanted! Uzbekistan has now imposed flight restrictions on our “K2” airbase, all while saying it has nothing whatsoever to do with our response to the Andijan massacre aftermath. (Yeah, right.) The Turkish Zaman reports that the US is now moving equipment, planes, and personnel into bases in Afghanistan, and that the base is effectively closed.

There’s more. President Karimov’s “enemies list”—which used to only contain Taliban-esque Islamists—has now been expanded to add “so-called democrats” to the roster. Eurasianet reports—accurately, I believe—that Tashkent could be preparing to break with Washington. The article also notes that it isn’t just our finger-wagging: the US is widely believed to be the main sponsor of recent revolutions in Georgia, Ukraine and Kyrgyzstan, and Tashkent wants nothing to do with us.

Karimov, of course, claims that the restrictions were put in place prior to the diplomatic tensions created by the massacre.

I hope Curzon isn’t implying that the rift between the US and Uzbekistan is the reason Karimov is expanding his list of enemies, because that’s not what the account he links to suggests. Rather, I think he’s pointing out that US disengagement and condemnation isn’t likely to put an end to Karimov’s reign of terror.

That’s certainly true, but it does not follow that the US should make nice with Karimov in order to combat Chinese or Russian influence in Uzbekistan, or in the hope that a close relationship will enable the US to exert leverage over Karimov’s domestic policies.

Curzon’s argument is based upon a particular strain of realist thought, one that owes a great deal to Kaplan’s discovery of classical realism a few years back. I have my disagreements with Kaplan, but I think he’s done a better job than most public intellectuals of understanding some of the dynamics of informal imperial rule. Nonetheless, there are a lot of problems with the kind of hyper-realism suggested by Cruzon’s post – and some of Kaplan’s work – in which we assess almost every aspect of American foreign policy in terms of a zero-sum competition for influence with potential competitors.

To put the debate in perspective, we should consider the nature of contemporary realist theory. Most current forms of realism hold that world politics are, at heart, a struggle for power. As long as the international system remains anarchical – lacks a supranational authority to make or enforce rules – it is impossible to transcend power politics. Institutions, such as the UN, won’t do the trick. Nor will trying to get everyone to agree to a set of principles, values, or international legal obligations.

For realists, states are the most important actors in international politics. This isn’t because most realists are committed to some sort of philosophical statism, but because they rightly argue that states are simply the most significant locus for the collective mobilization of economic and military power in world politics. The international struggle for power, then, is a struggle between states – particularly great powers – for security and relative advantage.

One important corollary of these propositions is that realists frown upon policy makers who place ideological and normative concerns, such as human rights, over the hard-nosed pursuit of security and power. The pursuit of such policies are likely to be counterproductive, because if a state loses relative power and influence it will be in less of a position to effectuate its values.

When I write of “hyper-realism,” I don’t have a particular theoretical construct in mind (those who expect me to bash “offensive realism,” “post-classical realism,” or whatever are going to be disappointed). What I mean, instead, is the adoption of broadly realist principles without sufficient nuance or qualification. What are the dangers of doing so?

1) an inability to clearly prioritize security interests. Not every interaction, or even every struggle for influence in a country or region, is important to the overall security of a state. Some struggles for power are not worth the costs. In fact, if we view every contest through a hyper-realist frame of zero-sum competition, it can become very difficult to craft an effective foreign policy.

In some respects, this was the problem for US policy makers in Vietnam. A Communist victory in South Vietnam was certainly tragic for many Vietnamese, but it really didn’t have very profound implications for US security or the overall shape of the Cold War, the flawed “domino theory” notwithstanding. Vietnam was a relatively unimportant periphery. Many realists rightly saw the Vietnam war not as an example of well-crafted realpolitik, but as an “ideological crusade” that soaked up US resources and undermined US influence.

At the other extreme (as John Mearsheimer has argued) since there were no real security interests either way at stake in Rwanda in 1994, there was no reason, from a realpolitik perspective, for the US not to have pushed for – or participated in – an intervention to stop the genocide.

2) a neglect of the multifaceted dimensions of power. This what I take Kaplan to mean when he writes “Speak Victorian, think pagan.” States find their lives much easier when others view their policies as legitimate: when they believe that they themselves will benefit from a state’s pursuit of its own interests, either in a material or normative sense.

Legitimacy is particular important for a state like the US. With no immediate peer competitors in international politics, one of the biggest threats to US hegemony comes from its eroding legitimacy. Anti-US sentiment is a kind of “wild card” that contributes to incipient policies of counterbalancing, diplomatic sabotage of US policy objectives, and, ultimately international terrorism.

The difficulty is that there are real tradeoffs between speaking “Victorian” – in terms of high-minded ideals and values – and taking actions that follow from thinking “pagan.” Acting “pagan” can, simply put, make it extremely difficult to credibly speak “Victorian.” Given various changes in global communications, the US does not have the luxury that some earlier predominate powers enjoyed: it cannot effectively hide its actions in one region or country from people in another.

Kaplan himself is correct in arguing, as Machiavelli does, that there are a variety of ways the powerful can maintain “plausible deniability” for actions that they support. But they won’t always work. States will, at various points, have to choose between pursuing realpolitik policies that undermine their legitimacy elsewhere, and abandoning those policies in order to enhance their room for maneuver in other countries and regions. Hyper-realists assume that, when push comes to shove, it is always better to engage in nakedly Machiavellian policies. They read the part of The Prince where Machiavelli says it is better to be feared than loved, but forget the title of Chapter XIX: “How one should avoid hatred and contempt.”

3) insufficient attention to variations in international structure . Not all aspects of international politics are governed by “pure” anarchy. Relations between greater and lesser powers always involve elements of hierarchy, even if those often fall short of the quasi-hierarchical arrangements seen in informal empires (e.g., the Soviet empire in Eastern Europe). Put differently, power-political competition between states neither exhausts the relevant dynamics of power politics nor is always the best lens to understand specific policy choices.

Does Curzon embrace the kind of hyper-realism I’ve described? Not entirely, but I do think the label has some applicability to the implicit and explicit stance he takes on Uzbekistan. A more nuanced realism, of the type I’m advocating, might lead to rather different conclusions.

To begin with, I am not at all convinced that maintaining good relations with Karimov constitutes a “high priority” security concern that should trump its potential costs – whether in terms of values, reputation, legitimacy, or strategic tradeoffs.

The Uzbek bases were clearly quite useful during the invasion of Afghanistan, but how important is it that the US retain Uzbekistan as a staging ground? The US has advanced aerial refueling capability; it also is expanding its basing capability in the region. At present, Uzbekistan is far more stable than Afghanistan, but, in the final analysis, the marginal benefits conferred by the presence of a base at Qarshi Hanabad are unlikely to be that significant.

Curzon is very concerned about opening the door to greater PRC influence in Uzbekistan. Despite my own dissection of Kissinger’s overly sanguine view on China, I also see very little evidence of an imminent, high-stakes strategic rivalry with China. Indeed, US and Chinese interests in Central Asia are not very far apart: preventing the emergence of militantly Islamicist regimes, ensuring the flow of oil through the region, and so forth. To be blunt: the US is not nineteenth-century Britain, Central Asia is not ripe for another “great game.” Why? Britain had a large stake in Central Asia because it needed to protect its most important colony, India. It saw the expansion of Russian influence, rightfully or not, as a threat to its long-term control over India. The stakes for the US are not as high, and the underlying sources of conflict with China in the region not as threatening.

Indeed, if China is pursuing direct control over oil reserves and pipelines, it is probably making a strategic blunder. As Jonathan Kirshner has argued, the costs of such guaranteed access make little sense given the nature of the current well-developed, international petroleum market. Moreover, incremental changes in the balance of influence in Central Asia are unlikely to matter if the US does develop a rivalry with China, since the epicenter of that rivalry will almost surely be in East Asia.

What about the Curzon’s argument that withdrawing from Uzbekistan would hurt the ability of the US to influence the regime? Certainly, the US has been able to transition authoritarian clients to democracy. Consider US policy towards South Korea and the Philippines in the 1980s. Yet here we face a fundamental tension in Curzon’s argument. Curzon argues that if the US leaves China (and we can add Russia) will simply fill the void. The implicit analogy is with Ethiopia under the Carter administration. But both scenarios cannot be true.

If Karimov has other viable strategic partners, then he is not really dependent upon the US. If he is not dependent upon the US, then there is little the US can do to influence his domestic policies. If, on the other hand, he needs the US badly enough that the US can exert leverage through engagement, then we have nothing to worry about in terms of China filling the vacuum left by US disengagement. If Curzon envisions a long-term strategy of turning Uzbekistan into a full-blown client state, and subsequently shaping the policies of the regime in a way that makes them less oppressive, then we’ve opened a whole new can of worms.

In order to explain why, I’m going to return to the third critique I made of “hyper-realism”: the need to recognize that not all international political interactions are governed by the logic of anarchy. One implication of this is that relations between great powers and minor powers will inevitably involve elements of authority. More generally, the dynamics of power politics change with the degree of hierarchy in interstate relations. In relations between great powers and the client states, for example, local leaders aren’t simply the representatives of sovereign states; they are also brokers between a great power and the domestic interests of the minor power.

In many empires, for example, imperial authorities rule through local intermediaries. Ideally, those intermediaries are dependent enough upon imperial authorities not to go too far astray in the pursuit of their own interests, but not so dependent that they look like nothing more than the tools of the empire when they bargain with local interests (again, Kaplan gets this right in his Atlantic article, “Supremacy by Stealth”).

When great powers deal with lesser powers, elements of this kind of relationship will always be in play; yet they can be sure that they will rarely strike an effective balance. Local regimes, for instance, will often have access to alternative sources of money and support. They can turn to other states or great powers. This is the position Karimov is in right now, even though the implications of the competition for influence are not as great as many observers suggest.

Another difference between imperial relations and relations between greater and lesser powers is also one of degree. In many empires, local rulers have – or develop – independent bases of support. Empires usually try to limit this, with greater or lesser degrees of success. Relations between great powers and lesser powers fall towards the far end of the spectrum: local regimes almost always have entrenched local power networks. This shouldn’t be surprising, in that they are leaders of sovereign nations rather than formal proconsuls or viceroys.

But here’s where things become really complicated. Great powers are best off when they maintain some ability to pivot between the leaders of a state and the state’s own population. Doing so allows them to insulate themselves against the risk of regime change, as well as to minimize the leverage a local leader has a “broker” between the great power and the small power’s various domestic constituencies. This means that, optimally, a great power should try to avoid putting “all of its eggs” into a local regime, particularly one that has a lot of divergent policy goals from those of the great power.

Such risk management is important for a number of reasons.

First, failing to do so can drag the great power into local disputes and conflicts in which it has little at stake.

Second, when that regime is corrupt, autocratic, and unpopular, the great power may also face a dangerous feedback loop: it is forced to adopt policies that favor the interests of a regime rather than those of its people, in doing so becomes more dependent upon the regime, and thus is forced to accommodate the regime even further if it wants to maintain its strategic interests. In the worst-case scenario, the great power will cease to have viable alternative partners within the other state; any change in regime means a complete loss of influence for the great power. At the same time, its support for that regime may increasingly undermine its reputation and legitimacy with other partners.

This is, in fact, one reason it can be more prudent to encourage the democratization of minor allies, even if those who win the election are less pliable than the autocrats they replace. It enables a great power to spread the benefits of the partnership more evenly across its ally’s population, thus insulating itself from future regime change. If, over the long haul, a great power faces a risk of the cycle I discussed, it may be better to cope with a less pliant alternative today than a hostile regime in the future. This is the dilemma, I would argue, that the US may have gotten itself into with the House of Saud, and one reason why invading Iraq looked so attractive to the Adminsitration: as Bill notes, the US could move its bases out of Saudi Arabia into another Middle Eastern country, one which, if the invasion went as planned, would (1) be a more reliable ally and (2) where the presence of bases would be less likely to motivate terrorism.

These add up to a whole host of reasons why transforming Karimov into a genuine client may be a game that isn’t worth the candle. Because many different foreign interests are jockeying for position in Uzbekistan, the US will have to make very extensive commitments to emerge as the dominant contender. These commitments will unduly enhance Karimov’s brokerage position with the US – in other words, it will give him more influence over the US then is really prudent for a great power to allow. If the kind of feedback loop I discussed above kicks in, the US will find itself very dependent upon Karimov and increasingly unable to position itself as a potential partner for opponents of the regime. Put simply, this could ultimately work against the main US strategic interest in the region: preventing the rise of new threats from militant Islamic groups.

Moreover, the more the US is in bed with Karimov the more it will harm its ability to “speak Victorian.” There are real reputational costs to being overly cosy with this regime, and the costs will only grow in magnitude if the US pushes to make Uzbekistan into a full-blown client state. It is very difficult to quantify the impact of such blows to US legitimacy, but given the fundamental way in which the “War on Terror” is a war of ideas and the specific importance of legitimacy for the US as the only global great power, I think it would be a mistake to overlook them. In the case of Uzbekistan, the benefits of the current relationship, and the risks of a more robust one, simply aren’t enough to justify the potential costs to America’s other strategic interests.

UPDATE: I started working on this post some time ago. I’ve now learned (via) Nathan Hamm at Registan.net, that David Hoffman and Cory Welt have already made a similar argument at the American Prospect. Reading their article makes clear why I write for academic, and not policy, journals. They put the argument in rather stark terms:

Whatever other power-projection missions K2 might fill, however, we must keep in mind that there is only one raison d’être for the U.S. military presence: to support combat operations in Afghanistan. Any other goals — hedging against unspecified, uncertain future threats — do not justify the potentially astronomical costs of that presence, including a loss in our moral authority to promote democracy and the possibility that U.S. forces will end up on the wrong side of a popular uprising against a dictator.

Hamm responds that:

We can more or less write off the hope that we will be able to get much serious democratization work done in Uzbekistan. Our total withdrawal presents no chance for improvement in the country. Even if we do incur costs of prestige, we incur great security risks be eliminating our presence in an increasingly brittle state that runs risk of collapsing into a chaotic civil war. For all of these reasons, if we can maintain some kind of military relationship in Uzbekistan without having to bend over backwards to placate the Uzbek government, we probably should at least think thrice before jumping ship.

Obviously, I disagree. It may be true that the US has not alienated the domestic population of Uzbekistan yet, but we cannot expect to gain strong benefits from the partnership without risking the kind of dynamics I outlined above.

Filed as: Uzbekistan, realism and empire