Robert drove a great deal of traffic to the Duck with his provocative posts on retrenchment and US alliances. His efforts to “grade” allies by strategic importance has led to some interesting results and fascinating discussions. But I think he’s working with an overly narrow view of what alliances are good for. Here I agree with Steve Walt in general, although I have a somewhat different spin.
Robert’s analysis treats alliances as if they are basically cooperative security arrangements, produce little in the way of externalities, and amount to a net negative investment only offset by the relative “strategic importance” of an ally. Thus, Robert advances three major criteria: (1) direct security interest, (2) need, and (3) values/symbolism. This can be summed up as “where is it located? Can it defend itself? Does a US commitment help Washington in the war of ideas?”
One problem with these criteria should be obvious. As Robert is quite aware, “security interests” can get pretty fuzzy once we move beyond “contiguous territory.” It isn’t even clear that common borders make necessary allies; after all, who exactly will be invading the US via Mexico? Regardless, one of the problems with “selective engagement” strategies hinges on precisely this ambiguity: it is too easy to slip into either *nothing* being a vital national interest or *everything* being crucial to national security. The “values” criteria is, I submit, even more subject to abuse.
But even if we wave away problems with the three criteria, my suspicion is that Robert’s assumptions about alliances are wrong. Alliances — whether formal defense pacts, strategic partnerships, or whatever — are also control mechanisms by which states reduce the autonomy of their partners: sometimes states forge alliances precisely because they judge other states to be unreliable (for a classic treatment, see Snyder’s Alliance Politics). In some strategic partnerships, the capabilities of the ally are irrelevant: the relationship enables power projection by providing extra-territorial bases and access. Alliances often involve significant externalities, in terms of the signals they send to potential friends and enemies alike. These signals aren’t just about “values,” but the scope and nature of US commitments. In short, an assessment of US alliance commitments requires a more nuanced outlook than Robert provides, and his consequent analysis may be significantly flawed.
Mexico is a good example. It isn’t so much that the US forges a close relationship with Mexico to pool military resources, but because it is more costly for the US to impose formal hierarchy on the country. When it comes to US basing and access agreements, it is true that these can develop relatively quickly — as in Central Asia after 9/11 — or that lost opportunities can return as a result of shifts in power and threat. But it is also the case that long-term basing arrangements tend to enjoy a degree of stability that often proves elusive in newer US-host relationships. Which amounts to a rather elliptical way of saying that it isn’t always efficacious to abandon strategic partnerships to save some short-term cash.
Adopting a more nuanced view of the “importance” of various bilateral and multilateral partnerships doesn’t take away from the importance of the exercise. I agree that everyone needs to spend more time thinking, and writing, about US strategic interests and which international relationships are most important in light of those interests.
But, at the end of the day, we need to be clearer about what we mean when we talk about “strategic overextension,” “relative decline,” and other key terms in the debate over “retrenchment.” The connection between bloated military budgets and strategic partnerships is not necessarily a direct one. US relative decline may force Washington to adopt clearer strategic priorities — such as pivoting toward East Asia — particularly when it comes to providing robust security guarantees. However, a lot of the recent “damage” done to the blood and treasure of the United States stems from wars of choice — and one in particular — that had very little to do with the central connective tissue of its alliances and partnerships.
The irony of focusing on the number of these relationships — rather than how much the US commits to them or broader aspects of the US defense budget — is that Washington’s alliance architecture is probably its best hedge against the rise of potential competitors. It both allows for the pooling of resources and it also reduces the likelihood of rising powers being able to wedge apart possible balancing coalitions. So, yes, the US needs to be smarter about how it allocates resources and less willing to resort to expensive military ventures of dubious geo-strategic value, but nothing about “retrenchment” requires slashing and burning its strategic relationships with other states. At least not in the foreseeable future.