Why 5 occupiers per 1000 population
    was enough in Japan but too few in Iraq

Why 5 American soldiers per 1000 occupied Japanese was enough, but 6 occupiers per 1000 Iraqis wasn’t.

A reader expert in these matters, responding to an earlier post in this space citing Sy Hersh’s report that the Iraq occupation was grossly undermanned compared to historical examples, sent links to the two RAND documents on which Hersh relied: an executive summary and the full text of James Dobbins, John G. McGinn, Keith Crane, Seth G. Jones, Rollie Lal, Andrew Rathmell, Rachel Swanger, and Anga Timilsina, America’s Role in Nation-Building: from Germany to Iraq. (2003)

My reader noted that the actual range of occupation force-to-population ratios had been wide: 100 per thousand Germany, 20 per 1000 for Bosnia and Kosovo, but only 5 per thousand for Japan. (Recall that the ratio in Iraq is about 6 per 1000.) The Japanese figure seemed, at first blush, to contradict Hersh’s thesis; if 5 occupiers per thousand population was enough for Japan, how could 6 per 1000 be grossly two low for Iraq?

In response to my query on that point, my reader sent the following response, and in response to a subsequent email the reader agreed to have it published here, without attribution [“I don’t particularly want to tone down the anti-Bush stuff as I would want to do for publication under my name (they do pay my salary, indirectly, and they are vindictive…)”].

I don’t know that my reader is right, but I do know that he’s (1) scary-smart and (2) a technocrat rather than a partisan.


The answer to these questions amounts to an indictment of the Bush administration’s astounding lack of preparation for the aftermath of a voluntary war that they initiated and conducted on their preferred timetable.

Why 5/1000 was enough for Japan has a lot to do the fact that its society and governing institutions were intact and cooperating with the occupation. In fact, Japan’s government began demobilization and other post-war transition functions before the US occupying forces arrived. So we were able to rely on indigenous functioning institutions. The US Occupation relied on the existing Japanese government and saw their function as “oversight.” The existing Japanese domestic bureaucracy emerged from the occupation stronger than ever. Extensive land reform helped gain the support of the countryside, and the economic boom that attended the Korean war sealed political support for the post war government.

Why this was true has a lot to do with the nature of the conflict and the country knowing it was defeated, and there may be some social and cultural factors that are specific to Japan.

Japan was fighting WWII to protect its empire – it was a classic state-level conflict rather than an ideological or religious or even primarily a racial one-so once we had defeated the Japanese military and eliminated the empire the cause of conflict between us and the population was largely gone. Although we did muck around some with internal Japanese structure, this was pretty much skin deep-for example, the big industrial alliances are somewhat tamer versions of the old Zaibatsu, we decentralized the police but it was immediately recentralized at the end of the occupation, and we didn’t offend the local religion in any way.

The fact that the Japanese military was in fact wholly defeated outside of Japan, the symbolic surrender on the Missouri, and MacArthur letting the emperor stay did a lot to legitimize the occupation. We relied on existing Japanese institutions for local social control. Japan also had experience with at least the forms of democracy.

These factors contrast greatly with the situation in Iraq, where Saddam had basically denuded the country of all legitimate political and social institutions, where there was no real surrender and we disbanded whatever institutions of social control existed (Baath party and Army), where our expressed aim in the conflict was a desire on our part to provide a different internal regime, where the background of a politicized religious difference provided ready tools for mobilization against the occupation (and of course Iran is mucking around in a way that had no real analogy in 1945 Japan-though the Soviet influence on Japanese communists provided an analogy a little later).

In sum, in Japan you had a largely intact society, you had limited grounds for ideological mobilization (other than nationalism, of course), and you had the powers-that-be making an accommodation to what was perceived as a legitimate occupation.

20 per thousand is also about what the British had in Northern Ireland and in Malaya, so it’s a pretty good number for dealing with an active conflict in the context of a more or less intact government. In Iraq, we are now dealing with an active insurgency in the context of a non-existent government. This would seem to require more than 20 per thousand. The question is, could we have done something that would have kept the civil administration intact or prevented an active insurgency from happening, or, preferably, both?

In Iraq, by contrast with Japan, you had a thugocracy and Army that melted into the woodwork, you had abundant grounds for ideological mobilization against an occupation that had very little legitimacy (as measured by international or regional approbation), and you had non-functioning domestic institutions that we in any case set out to disband. Plus Saddam had already let all the criminals out of jail and organized abundantly armed paramilitaries. The war was not over a year ago.

Was this chaos predictable? We knew the jails had been emptied, we knew how thoroughly Baathified the police and Army were, and we knew that we intended to de-Baathify the country, so I think it was. Once the war started, we knew from the first days that there were no organized surrenders and there were major actions by irregulars, so it was certainly predictable at that point. The administration simply had ideological blinders about how we would be welcomed as liberators.

Compared to Japan, you also have borders that need to be patrolled (both Iranian and Arab militant infiltration have been a problem) -a mission that we have not provided forces for and that increases force requirements.

So right away you needed the equivalent of a police force and a border patrol (The US has 2.3 police per 1000-but how many would we need if we emptied the jails and prisons, and if there were organized hostile forces? The numbers are not exactly comparable because police have a lower duty cycle than military forces, but offsetting this is that military forces have a lower “tooth to tail” ratio because they need to provide all their own logistics.). You also needed to establish local governments, and you needed dedicated counter-insurgency forces. That begins to give you an idea of why the needed ratio would be probably higher than what has worked elsewhere. The looting (which seems to have been a continuation of war by other means) right after the military victory showed how out of control things were-although partly our forces seem to have not been told it was part of their job to provide law and order, in addition to the numbers and the types of forces being inadequate for this mission.

Now if we had been able to set up a truly legitimate Iraqi authority, then these needs might have been relatively brief-but in fact because of the Kurd/Sunni/Shia distinctions and because of DoD’s misguided dalliance with Chalabi, there was no realistic plan to develop a legitimate Iraqi government. What we should have done is quickly established legitimate local administrations (Amazingly, some US military units did this on their own initiative but there was no overall plan or doctrine for doing so and so the results have varied greatly) and then built upward from there-the whole governing council has been mostly a distraction. And we should have done this under military occupation, rather than wasting the effort setting up the parallel CPA structure-which barely gets out of the green zone in Baghdad. This of course would have involved the military and the Pentagon thinking seriously in advance about “nation building,” which it refused to do.

In other words, the “right ratio” in Iraq would have been very large (maybe 20 per thousand for counterinsurgency plus 10 per thousand for policing and other civil functions plus whatever you need for border patrol) -or, better, ZERO, in the absence of a better political and social policy for occupied Iraq and a more legitimate occupation. My guess if you had gotten the reconstruction politics right, and if the war and the occupation had been more legitimate (UN cover, support in the Arab and Muslim world), and you had some real language capability and cultural understanding (either in US forces or by attaching Arab elements to US units), that you could have gotten away with something like 10 to 15 per thousand Military Police -type forces in addition to something approaching 5 per thousand war-fighters on hand on hand at the end of major conflict for engaging/deterring active insurgency and border patrol, and that you might have been able to get these forces out after a year or so… (I think you might have done with somewhat less than 20/1000 for the country as a whole because the Kurdish areas and large stretches of the South could have been much easier problems than the “Sunni Triangle”-Ideally you would have liked to start out at something like 25 per thousand and rapidly tapered off as situations allowed-but this would clearly have required major participation by non-US forces.)

But I don’t think even 100/1000 would have been enough with the kind of units we deployed (essentially no Arabic language capacity and no skill or doctrine in political and social reconstruction) in the context of an occupation with very limited political legitimacy-in fact deploying such units in this context might have just produced pitched urban battles earlier, and mobilized the whole Arab world against us.



Another reader says my correspondent has it just about right, and adds two more references which he says would have been well-known to Pentagon planners: another RAND report, this one by James Quinlivan, “Force Requirements in Stability Operations” dated 1996, and Barry Posen, “The Security Dilemma and Ethnic Conflict,” Survival, vol.35, no.1 (1993). My reader describes the Posen piece as follows: “A rather concise explanation as to why disarming and reintigrating multi-ethnic state is so difficult; and it drives home the point about U.S. troop numbers, and why having sufficient troops on the ground to ensure security of the population is so very important.”

Still another reader points out that under the Japanese occupation the Japanese were forbidden to possess firearms, making the whole business much easier. He thinks the failure to disarm the Iraqis was at the root of all our other troubles. I’m not qualified to judge whether disarming the Iraqis was feasible, or what the requisite force levels for that task would have been.

Author: Mark Kleiman

Professor of Public Policy at the NYU Marron Institute for Urban Management and editor of the Journal of Drug Policy Analysis. Teaches about the methods of policy analysis about drug abuse control and crime control policy, working out the implications of two principles: that swift and certain sanctions don't have to be severe to be effective, and that well-designed threats usually don't have to be carried out. Books: Drugs and Drug Policy: What Everyone Needs to Know (with Jonathan Caulkins and Angela Hawken) When Brute Force Fails: How to Have Less Crime and Less Punishment (Princeton, 2009; named one of the "books of the year" by The Economist Against Excess: Drug Policy for Results (Basic, 1993) Marijuana: Costs of Abuse, Costs of Control (Greenwood, 1989) UCLA Homepage Curriculum Vitae Contact: Markarkleiman-at-gmail.com