I appreciate the Letter to the Editor from Hugo Relva, Legal Adviser at Amnesty International responding to my earlier post, in which I questioned the coding procedures that Amnesty used in its major survey of universal jurisdiction. First, Relva writes that the overall percentage of states with universal jurisdiction provisions is 76% not 86%. I made a mistake in stating the latter figure, and have added a notation to the original post correcting this error. [As a side note, the number may be closer to 84%; the Amnesty study also states: “a total of 163 states (approximately 84.46%) can exercise universal jurisdiction over one or more crimes under international law, either as such crimes or as ordinary crimes under national law.”]

Nevertheless, Relva’s reply does not address the concern I raised about the coding procedure, which is based on Amnesty’s erroneous understanding of the legal requirements of the Rome Statute of the International Criminal Court. Relva explains that the survey did not code every state party to the Rome Statute as a universal jurisdiction state. I didn’t say otherwise. And I agree that is not their coding method. On the contrary, as I wrote in my exchange with Kevin Jon Heller (emphasis added):

What Amnesty does seem to suggest, however, is that a subset of ICC member states – due to a particular form of national implementation of treaties — are universal jurisdiction states.  Those are states where domestic law provides for courts to have jurisdiction over any crime required by treaty. And in this regard Amnesty’s coding is based on the organization’s explicit assumption that the Rome Statute contains an implied “obligation to exercise universal jurisdiction.”

In other words, the problem with the coding procedure is that it appears to involve the following two steps:

Step 1: the proposition that the Rome Statute obligates state parties to enact universal jurisdiction for ICC crimes

Step 2: the decision to code a state as having enacted universal jurisdiction if it (a) is a party to the Rome Statute and (b) its domestic law provides for jurisdiction over crimes obligated by international treaty

As I explained in my original post, Step 1 is flawed. The Rome Statute does not include universal jurisdiction, and has no obligation whatsoever for state parties to provide (extraterritorial) jurisdiction for ICC crimes.

I suspect that the reason Amnesty sets forth the two steps as a part of its coding procedure is because it is meaningful – i.e., that it makes a difference in their results. It is difficult to discern, from the Annexes of the study, which particular states might be affected, because the relevant information is not provided.

It is encouraging to learn, from Relva’s letter, that no government has asked the organization to correct the country information in the study. If this coding procedure comes to light, however, some governments might.