Lott has filed a response to Levitt's motion to dismiss.
He doesn't have a good argument on the question of the meaning of "replicate", basically just asserting that it means "analyse the same data in exactly the same way" and ignoring the other usages it has.
On the one hand, I hope Levitt's motion to dismiss is successful because Lott's case appears almost completely without merit and it's undoubtedly costing Levitt money.
On the other hand I'd LOVE to see Lott tryign to defend his reputation in open court.
Tim -
Thanks for keeping us up to date on Lott's suit. Your links at elsblog pretty clearly show that Lott himself doesn't (or at least didn't) use his overly restrictive definition of replicate, and it was in the Florida voting affair of 2000, where he really tortured statistics, no less.
Though I've read enough about Lott to turn blue, my memory doesn't serve - - were there any other studies of the "more guns, less crime" conclusion that specifically countered Lott, as Levitt claimed in the book? That would torpedo the suit right there.
I made a [table](http://timlambert.org/2003/07/0728/) of all the studies.
Tim - thanks for the link. Assuming that your analysis from July 2003 is complete and correct, and that there was no deluge of "more guns, less crime" research since then, I think I would recommend to Lott that he withdraw at least his first claim. (However, I admit that my own feelings are precisely the same as Ian G's, above.)
The second claim about the email, is still interesting. Did Lott "buy" the special issue of JLE (I think so), and was it peer refereed (Lott may be right here, but how is Levitt's comment defamatory?)