Thanks for the comment, Chris. You raise interesting points.

1. Maybe the ship has sailed on case-specific exceptions. I’ve suggested that if there is a systemic exception we don’t need case-specific exceptions. Does that work in reverse? If there is a case-specific exception can we dispense with case-specific exceptions? At first blush I’d say no. Can’t we imagine a case where there is no fraud and the case goes entirely according to the prescribed procedure, but the procedure itself fails to satisfy due process? For example, suppose a foreign state still uses wager of law. Or suppose a case where the executive is so powerful and terrifying that judges always do its bidding, even if there’s no procedural fraud in the particular case. What do you think?

2. I think “may” is probably the right choice for Article 4(4). I’m just thinking about this from the American perspective; in federal practice (though not, e.g., Massachusetts practice) a judgment is not stayed pending appeal unless the appellant files a supersedas. So “must” would, I think, be overprotective. The bigger “may” versus “must” problem seems to me to be Article 5. The UFCMJRA has some mandatory grounds and some permissive grounds for non-recognition. Not so here! Again, what do you think?