Sunday, September 2, 2012

Big countries, small wars, different mindsets

Political scientists and other scholars have spilled a lot of ink on the question of why big states lose small wars, i.e., wars against weaker adversaries. Phil Arena recently pointed me to
Patricia Sullivan's 2007 JCR article, which I have looked at (meaning looked at, not read every single word of). Her main argument, put in simplified form, is that big states are more likely to lose small wars when their objective is coercive, i.e., when it requires the adversary to change its behavior, as opposed to when the objective can be accomplished simply with brute force (i.e., overthrowing a regime or conquering territory). The main reason, she argues, is that big states are more likely to underestimate the costs of achieving coercive objectives. Sullivan has a typology of objectives on a continuum with brute-force objectives at one end and coercive objectives at the other. [
Cf. Schelling, Arms and Influence (1966).]

Interestingly, the objective "maintain regime authority" falls in the middle of Sullivan's continuum. This is interesting because if you had to choose a three-word label for the U.S. objective in Afghanistan, it would be "maintain regime authority" (against those who seek to overthrow it). That was also basically the U.S. objective in Vietnam, as Sullivan suggests (i.e., the stated aim was to maintain an independent, non-Communist South Vietnam).

The main point I want to make is that looking at this article highlighted (once again) for me the distinction between those who emphasize the idiographic in their methods versus the nomothetic, or to put it in simpler terms (this would probably drive PTJ up the wall, never mind), the difference between those who do historical case studies and those who do formal modeling or quantitative work (yes, some people do both in the same book or article, but we'll put that aside for now).

Sullivan's approach would suggest that the wars in Afghanistan and Vietnam are basically similar because the objective (maintain regime authority) was the same. Of course she would acknowledge there are local differences, but she is not concerned with exploring them; she is interested in a model that explains, at some kind of quasi-'law-like' level, when big states lose small wars, and she gets there via a 'homogenizing' approach, so to speak. So if you took her approach, even though her concern is not explicitly with policy debates or decision-making, you might be quite receptive to analogies between Afghanistan and Vietnam.

A case-study approach might suggest something quite different. As opposed to a receptivity to the Vietnam-Afghanistan analogy, it might suggest a wariness about such an analogy and indeed about analogies in general. Yuen Foong Khong's Analogies at War (1992) showed quite convincingly that analogies to Munich and to the Korean War exercised a harmful influence on the Johnson administration's Vietnam policymaking. The takeaway lesson of that book, one could argue, is that because even smart people find it difficult to use analogies properly (i.e., in a sufficiently discriminating way), one should be wary of the mobilization and use of historical analogies, especially in a broad-brush way (e.g. "Afghanistan is like Vietnam"), in policy debates. Each case should be looked at primarily on its own, in other words.

One might say there is no contradiction here. Sullivan is interested in explaining outcomes, not prescribing a method for policymakers to use in decision-making. Khong is interested in showing why and how policymakers tend to misuse historical analogies. They are doing different things but not contradictory things. OK. Nonetheless, if you are a scholar with a more nomothetic mindset (e.g. Sullivan), if a policymaker called you and asked you what to do, you might start thinking in terms of historical analogies, because you are used to homogenizing historical cases and treating them as data points to be coded. Whereas if you have a more idiographic mindset, you might be less prone to think in terms of analogies, i.e., in terms of similarities between cases, and more prone to emphasize that each situation is unique. And I think that might be true even if, for purposes of getting your dissertation or article or book past the relevant authorities, you made a general argument that mobilized case studies in its support. Close contact with the historical specifics of cases, even when mobilized to support a general thesis or argument, is bound to sensitize one to differences and unique elements. In other words, even if (as is very often the case) you are doing case studies to develop or back up an overarching theory, you are bound, almost despite yourself as it were, to acquire some wariness about the merits of generalization.

Of course, our two hypothetical scholars might end up, policy-wise, in the same place: for example, both might have decided to oppose the Obama administration's "surge" in Afghanistan (or to favor it, as the case may be). But they would have reached their conclusion, whatever it was, by rather different routes.

P.s. Just to be clear (and repetitive), that the U.S. lost in Vietnam doesn't necessarily mean it's going to 'lose' in Afghanistan. This partly depends on how one defines 'victory'. (See this post and the attached comments.)


Kindred Winecoff said...

Two issues.

1. "Each case should be looked at primarily on its own, in other words." And how, exactly, can that be done? We always approach new cases with reference to old ones. I don't think there's any way around it, and if there were I'm not sure the "on its own" approach would outperform the "look for similarities with past events and act accordingly". I don't think it's unreasonable to believe that we use mental models (i.e. heuristics) all the time anyway, so trying to create better models is better than relying on worse ones. (I imagine Phil would say something along these lines.)

This is where typologies are important. Hopefully these typologies will be created (or revised/improved) through deep process tracing or other historical approaches -- including nomothetic approaches, for me anyway!

More important is to be honest about how two situations are different as well as similar. This brings me to:

2. I don't think it's very helpful to think of Afghanistan and Vietnam as having the same goal of maintaining regime authority. Vietnam was not about Vietnam... it was about slowing or stopping the spread of global communism in Asia. Afghanistan is about eliminating safe havens for transnational terrorist groups. I don't think the US cared about the Thieu or Karzai regimes per se at all.

Why is this important? Well, thinking in terms of local aims (e.g. "regime authority") would lead us to believe that both Vietnam and Afghanistan were catastrophic failures. Thinking in terms of systemic aims (e.g. "anti-communism/terrorism") would lead us to conclude that both Vietnam and Afghanistan were at least partial successes at the systemic level, albeit with plenty of failures at various lower levels.

I owe you a response to your comment at my place. I'll get to that next.

LFC said...

Interesting points, I will have to think about them some, or at least about #1.

I can say one thing w/o too much reflection, though, and that is that "slowing or stopping the spread of global Communism in Asia" required maintaining the Thieu (or another non-Communist) regime. So I think it's hard to separate those two things.
Plus I'm inclined to think the Vietnam war was not much of a success on any level. A point I've made before -- buried in some obscure post or other -- is that once the Indonesian Communists were slaughtered by the Indonesian regime en masse in '65, the domino theory lost a lot of its credibility. (Mueller makes this pt, so it's not original w me.) After that, the spread of Communism in Asia shd probably have stopped ringing such urgent alarm bells in Washington. Of course not everyone agrees -- Walt Rostow argued in retrospect that the Vietnam war 'saved' other Asian countries from Communism. But then, as the architect of the Johnson bombing campaigns and general super-hawk, he had a personal interest in saying that...

(I'm sure I haven't done full justice to the Sullivan article, btw, but that's the way the cookie crumbles...)

Kindred Winecoff said...

I'm not sure if slowing/stopping the spread of communism did require maintaining the Thieu regime. It may have simply required demonstrating that the costs of expansion would be high, not low. Note that communism essentially stopped spreading post-1960s. Perhaps a lot of reasons for that, but it's possible that US security commitments became more credible once they demonstrated a willingness to incur costs. So China chooses rapprochement in 1972 rather than conflict; maybe that doesn't happen w/o US military intervention in E Asia.

I'm not actually sure I believe this... just saying that it's a reasonable argument and the timeline fits.

So I don't buy domino theories either, as programmatic/mechanistic outcomes. But I'm also not willing to say that massive US military interventions over a period of decades had no effect whatsoever on the decision-making of other leaders. It seems like there's a reasonable middle ground where US intervention in Vietnam/Afghanistan has some deterrent (or other positive) effect relative to the counterfactual. More than likely this is unobservable, but that doesn't make it any less real.

LFC said...

[warning: rambling]

This is partly a generational thing, perhaps: I lived through the Vietnam war albeit as a kid, was in 6th grade in '68, I remember, albeit quite vaguely, being on the Mall for the first big moratorium march wh/ I think was in '69 (I know I was there even tho I have few or no actual tactile memories of it, if that makes sense). I was very involved in the McGovern campaign in 72, even though too young to vote.

No way am I going to agree that the costs incurred, on Americans and Vietnamese and others, were worth it to make US security commitments credible, even if that was the effect, wh i'm very very skeptical of.

After all, the major US allies in Europe were opposed to the US role in Vietnam. The British and French (and the latter shd have known) thought the US was out of its mind. So, if anything, it cost US credibility w allies, who also shared an interest in preventing the spread of Communism. The war was conducted in an immoral fashion to some extent -- carpet bombing, napalm (the effects of which just now are starting to be cleaned up in parts of Vietnam), etc.

There were other ways to establish the credibility of US sec. commitments than 58,000 dead US soldiers, and ditto for the costs of the Korean war, I suspect (tho that's a somewhat different situation).

A commenter the other day on CT told a story of being drafted in '66 and being assigned to graves detail. He put, I think he said, 6 or 7 thousand bodies in body bags in the course of one yr. This is one person assigned to one Army graves detail.
Afghanistan is a different case, as I argued in the post, but I can at least understand, even if I don't necessarily totally agree with, the visceral opposition to US mil. intervention on the part of some of those who are older than I am and had to deal w Vietnam more directly.
Final pt: The historiography of the Vietnam war is flourishing these days -- a new bk just out by Frederik Logevall about the French period mostly, another new bk written from the N Vietnamese archives (I forget the author's name) -- and I think the weight of this scholarship, tho I'm not absolutely sure, wd support me on the 'credibility' issue.

Kindred Winecoff said...

Yeah, I really don't know all that much about Vietnam so I'm not prepared to argue strongly in favor of any side. I certainly know enough to agree that moral atrocities were committed -- I'm a big fan of Hitchens' book on the case for trying Kissinger with war crimes -- and I think it's entirely possible, and perhaps even overwhelmingly likely, that the costs outweighed the benefits.

My point is that that doesn't mean there were no benefits. Nor does it mean that the benefits were negligible nor even that the cost/benefit balance would have ended up much improved under reasonable counterfactuals. (I'm not meaning to diminish the human cost/experience, although it may seem as if I am.)

I could tell a story in which the Europeans didn't like the US involvement in Asia not because it *did not* affect the balance of power there, but because it *did*. I.e., perhaps the Europeans would have preferred the US focus on guaranteeing their security rather than the security of the Koreans or the Vietnamese or the Japanese or whomever the US thought they were protecting. Perhaps, i.e., that US involvement in E/SE Asia cost the US credibility with allies in Europe, but perhaps it gave the US credibility with allies in E/SE Asia. Or, perhaps, with the US's antagonists. (Again: China)

I don't know that this is true, but is certainly could be. US policymakers at the time seemed to be thinking in these terms. Maybe it was a miscalculation. Maybe the fears over Soviet expansionism were fully justified. Maybe the final costs of Vietnam were unforeseeable in 1964, or perhaps they were foreseeable and just unforeseen.

Counterfactuals are hard, yadda yadda.

I just return to this point: US involvement in Vietnam and elsewhere changed stuff. Some of those changes were bad. Some of them might not have been. Best to consider both at once rather than only one of them.

No need for further response if you're fatigued. If you do reply I likely will forget to check, so maybe call my attention on my blog or via e-mail.

Phil Arena said...

Interesting discussion. Sorry I didn't catch this sooner. I wasn't very active online this weekend.

"I don't think it's unreasonable to believe that we use mental models (i.e. heuristics) all the time anyway, so trying to create better models is better than relying on worse ones. (I imagine Phil would say something along these lines.)"

Yes, I would. :)

I'd also say that I think you're being a bit unfair here, LFC. I can't speak for Sullivan, but myself, I would never give policy advice based on analogies. Nor would I ever assume that my simplifying model (be it statistical or theoretical) has captured all the relevant details, thus obviating the need for any appreciation of the specifics of any given case when deciding how to handle that case. I find myself in the minority on many issues, but I'm actually not sure this is one of them.

You are probably right that people whose primary research methodology is the case study are more inclined to say that "every case is unique". I know some (certainly not all) such scholars are inclined to reject the possibility of generalization altogether, which would imply a belief in the "uniqueness" of cases that far exceeds that which most statistical modelers see. But I think you're turning statistical modelers into a caricature.

LFC said...

Both comments noted and I will reply, though not right now. In fact may not be for a few days, as i'm unavoidably busy the next couple of days.
But I will respond as soon as i can -- after all, I can't let Phil have the last word. ;)

Phil Arena said...


Take your time. Have a nice Labor Day.

LFC said...

Thks, same to you.