Mr. Gilson would be well advised to check his own factual assertions before condemning others. (Full disclosure: John Lott and I overlapped in graduate school and have been friends for many years.) Lott's findings on gun ownership and crime rates in fact have been replicated multiple times by every scholar who has attempted to do so, and the findings have proven robust in papers published in peer-reviewed journals. The "coding errors" to which Gilson refers were made by other scholars attempting to extend Lott's analysis to additional years; Lott is not responsible for those mistakes. Lott's purportedly "sloppy research" has resulted in the publication of approximately 100 papers in peer-reviewed journals; so much for the sloppiness accusation, which is ill-defined in any event. In the lawsuit referenced above, the judge ruled against Lott on one cause of action; but the other cause of action remains. Only were Lott's work as sloppy and dishonest as Gilson's comments would Lott have anything about which to apologize.

Benjamin Zycher

Thanks for your comments, Benjamin. Regarding my statement that Lott's findings have not been replicated by other researchers, I refer readers to Chris Mooney's article on our website, which details how other scholars were unable to recreate the findings upon which Lott's "more guns-less crime" thesis was based. That alone would suggest that it is not possible that Lott's findings "have been replicated multiple times by every scholar who has attempted to do so." As for my claim that no one's ever been able to replicate Lott's findings, I confess that I deferred to the Freakonomics guys on that one, perhaps too quickly. (That broad claim is at the heart of Lott's suit against Freakonomics co-author Steven Levitt. More on that in a sec.) But Lott's work is hardly a slam dunk, since his use of data is so contentious. As for the "coding error" matter, Mooney's article shows that these errors originated with research done by Lott and his associates, not his critics. Lott tried to explain away these errors as "minor" and not significant enough to affect his conclusions. Finally, you are correct that the judge did not dismiss Lott's entire claim against Levitt. He may still pursue his claim that Levitt defamed him in an email to another economist. But the court threw out Lott's contention that Freakonomics defamed him by asserting that his work could not be replicated by his colleagues. I've amended my post accordingly.