Reconsidering the Cases of Humanitarian Intervention

To the Editors (Alex J. Bellamy writes)

In a recently published piece, Robert Pape makes some misleading and erroneous comments on my published work.¹ First, Pape writes, “Alex Bellamy, a staunch advocate of R2P [the responsibility to protect initiative], catalogues episodes of mass atrocities to clarify ‘R2P’s preventive agenda,’ with a total of twenty-one qualifying for intervention from 1990 to 2010” (p. 212). Pape provides no reference to support this statement. In truth, I have never produced a list of “cases” that “qualified” for intervention. The datasets that I have produced relate to cases where the lowest casualty estimates suggest that at least 5,000 noncombatants were intentionally killed. Nowhere have I suggested that this “qualifies” these cases for intervention. Actually, I have been generally critical of abstract talk about criteria and thresholds for armed intervention, as well as the pervasive and erroneous tendency to treat R2P as synonymous with humanitarian intervention, both of which I believe to be disconnected from political realities.² Since I began working on R2P a decade ago, I have repeatedly expressed caution about the use of force for protection purposes for reasons similar to those aired by Pape last year.³ In my first book on R2P, I concluded that “non-consensual force is a highly unreliable form of protection.”⁴

Second, Pape claims that my dataset includes “virtually all instances of armed conflict around the world” during the 1990–2010 period (p. 213) and that, because of this, “one can reasonably wonder whether there are any meaningful limits to R2P” (p. 213). In this respect, it bears repeating that I have never compiled a list of cases that “qualified” for “intervention” or for which R2P “obligates” such action. Moreover, the claim that a set of twenty-one cases of mass killing spread over twenty years includes “virtually all instances of armed conflict” during this period is clearly untrue. To take just one example, the Uppsala Conflict Data Project’s dataset identifies approximately 165 armed conflicts during this same period.⁵ Clearly, my own dataset of mass atrocities covers just a small portion.⁶

I appreciate this opportunity to correct the record.

Robert A. Pape Replies

Reading Alex Bellamy’s letter, it is refreshing to know that both Bellamy and I are skeptical about the value of armed intervention in many cases.¹ The real disagreement appears to be about whether Bellamy’s very good data on mass atrocities reflect responsibility to protect (R2P) principles for intervention. Even if he does not specifically qualify each case in this manner, my use of his dataset of mass atrocities as a set of cases that would qualify for intervention according to the standards of R2P is perfectly appropriate. Let me explain.

Bellamy’s dataset contains a list of mass atrocities that, in principle, would trigger the international community’s responsibility to protect. Specifically, to better understand their context, Bellamy collected a dataset of “episodes of mass killing defined as a minimum of 5,000 civilians killed intentionally.” Most of these episodes occurred during armed conflict, but some did take place in peacetime. Using this dataset, Bellamy then goes on to argue that “atrocity prevention requires tailored engagement that targets both peacetime atrocities and those committed within a context of armed conflict.”²

True, Bellamy never describes the measures that should have been taken in particular cases and whether his dataset is the complete list of events that would trigger R2P. Further, he would prefer to use nonmilitary measures to prevent atrocities before they occur—provided these nonmilitary measures actually work.

Bellamy, however, also expects that in some cases nonmilitary measures will fail and armed intervention will be necessary. In 2009 he wrote, “When non-military ways are unable to protect endangered civilians … R2P calls for the deployment of military force. Military force can be used either to protect populations from attack or to coerce or compel compliance by targeting those responsible for attacks on civilians.”³ Similarly, in 2012 he stated, “Sadly, preventive efforts will not always succeed. That is why the...