At long last, Amnesty has weighed in on the debate between me and Ryan about its methodology for determining whether a state exercises universal jurisdiction over at least one international crime. As I expected, and contrary to Ryan’s claim, Amnesty does not consider it sufficient for a state to have incorporated the Rome Statute into its domestic legislation. On the contrary, it requires the existence of domestic legislation that extends universal jurisdiction over an international crime, whether specifically (“this legislation provides universal jurisdiction over international crime X”) or generically (“this legislation provides universal jurisdiciton over all international crimes defined in ratified treaties”). Here is the key statement from Amnesty’s response:
[T]he above mentioned conclusions are not based on counting “[s]tates as having enacted universal jurisdiction if the state is a party to the Rome Statute for the International Criminal Court or, more precisely, if the state has adopted a form of implementing legislation along with ratification of the treaty”. That would be a mistake. For example: Chad, Gabon, Maldivas, Nauru, and Zambia – which are states party to the Rome Statute are enlisted in the report as not providing for universal jurisdiction for any of the crimes defined in the Rome Statute. And Ireland and Liechtenstein – which have ratified the Rome Statute and enacted legislation implementing it into national law — are also both considered as not providing for universal jurisdiction with regard to crimes against humanity and genocide. In sum, Amnesty International considers that the domestic law in these countries has the effect of conferring universal jurisdiction over crimes defined in, for example, the Rome Statute. Therefore Amnesty International are not basing the claim that such countries have universal jurisdiction on the fact of their ratification of the Rome Statute alone but rather on domestic legislation that enacts universal jurisdiction for all crimes in treaties (including for example the Rome Statute) that they have ratified.
Unfortunately, Ryan still insists that Amnesty is overcounting the number of universal jurisdiction states. Here is his response, in relevant part:
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.
There are a number of problems with this response. To begin with, there is no “Step 1″ in Amnesty’s analysis…