The Defamation Lawsuit Against One of the "Freakonomics" Authors: Analyzing the U.S. Court of Appeals for the Seventh Circuit's Decision
By JULIE HILDEN
|Monday, March 2, 2009|
On February 11, a three-judge panel of the U.S. Court of Appeals for the Seventh Circuit dismissed a defamation lawsuit against Steven D. Levitt, one of the co-authors of the book "Freakonomics." (Levitt is an economist; his co-author, Stephen Dubner, is a journalist and was not a defendant in the suit.) The suit was brought by John Lott, a fellow economist whose work is discussed in the book.
In this column, I'll explain why the suit proved unsuccessful, and why I think the panel's decision was right in one respect, but off-base in another.
The Paragraph and the Email that Were the Basis for the Defamation Suit
"Freakonomics" discussed Lott's work in just one paragraph, in a section on explanations for why crime rates dropped sharply in the 90's. Lott has argued that allowing law-abiding citizens to carry concealed weapons will diminish crime. But Levitt wrote:
"[T]here was the troubling allegation that Lott actually invented some of the survey data that support his more-guns/less-crime theory. Regardless of whether the data were faked, Lott's admittedly intriguing hypothesis doesn't seem to be true. When other scholars have tried to replicate his results, they found that right-to-carry laws simply don't bring down crime."
Lott claimed this paragraph was false and defamatory, particularly in that it suggested that his original results must have been dishonest, for they could not be replicated.
In addition, Lott added a later, separate claim that alleged that, in an email exchange with a third economist, Levitt had again defamed him. In particular, Lott claimed, Levitt had accused Lott of buying support for Lott's "more guns, less crime" theory by paying for the publication of a journal filled only with non-peer-reviewed articles that took a positive view of Lott's work.
However, the parties reached a settlement as to that second claim. Its value was likely low because what was at issue was simply a communication to one person, and any damage that Lott incurred likely could be at least largely corrected now by his contacting that person and making his case – though, granted, that person might have passed on the content of the email to others too, thereby raising the potential damages. (By contrast, contacting every reader of "Freakonomics" would be practically impossible, and contacting even every economist who might have read "Freakonomics" would be extremely burdensome.)
The Seventh Circuit Panel's Reasoning – and Why It is Persuasive
With the email claim settled out, the Seventh Circuit grappled only with the "Freakonomics" paragraph claim. After explaining at length why Illinois law applied (for Lott had later argued for Virginia law), the court noted that under Illinois law, as interpreted by the Illinois Supreme Court, a statement is not defamatory if it is "reasonably susceptible of an innocent construction." This standard led to a win for Levitt, for the panel held that it was satisfied that the "Freakonomics" paragraph at issue was reasonably susceptible of an innocent construction.
The panel's analysis turned on Levitt's word "replicated." Lott argued that the paragraph meant only one thing by saying that others had replicated Lott's work and been unable to get the same results: It meant that Lott must be faking the studies. However, the court reasoned that "replicated" could mean instead that "scholars tried to reach the same conclusion as Lott, using different models, data, and assumptions, but could not do so." This would be an innocent interpretation, the panel explained, because it would not impute any dishonesty to Lott -- for the reason that the results could not be replicated could simply be that the experimental conditions were not exactly the same.
The panel also suggested that the debate over whether the conditions were similar, and what conclusions could be drawn from particular results, belonged in an academic forum, not a courtroom. In the end, the panel suggested, Lott's work should face not a courtroom jury but a jury of his peers: fellow economists.
I believe the panel's reasoning is persuasive as to the "replicated" statement. Granted, anyone who has taken a high school science class knows that to replicate an experiment is to try one's best to do that same experiment under the same conditions that originally applied – as painstakingly recorded by the original experimenter. But importantly, here, the wording was that others had tried to replicate Lott's "results," not his "experiment." That wording could encompass attempts to use data from various different cities and time periods, other than those Lott had directly studied, to test the same hypothesis. And various results might apply in various cities and times, with none among those results being false or fraudulent.
The Court's Holding on the Invented-Data Material: More Questionable than Its Replicated-Results Holding
Much more dubious, however, in my opinion, is the court's decision to give "Freakonomics" a pass for another part of the paragraph at issue – the part referring to "the troubling allegation that Lott actually invented some of the survey data that support his more-guns/less-crime theory."
The panel seemed to think that this statement did not defame Lott because Levitt "distanced" himself from the invented-data allegation, by putting it aside to focus on the replication of the experiments' results instead. But that may not be the right interpretation of the law.
It's black-letter law that a repetition of a libel is still a libel; the fact that someone else said it first is not a defense (though it may affect the apportionment of damages). And Levitt did not rebut the invented-data allegations after raising them – which would have negated any possible damages from his repeating the allegations. Instead, he just bracketed these allegations and moved on to another issue – the issue discussed above, of whether Lott's experiments could be replicated.
Overall, too, the paragraph seems quite damning to Lott: "Freakonomics" technically reported a dispute among economists about whether Lott invented data, but arguably insinuated that Lott well might have done so, by coupling that report with other questions about his scholarship.
In sum, it would be difficult to read the paragraph at issue in the "Freakonomics" case – reproduced above – without getting the idea that Lott's scholarship was dubious at best and dishonest at worst. Yet the Seventh Circuit panel still concluded that the paragraph was not libelous. In that respect, the panel may well have erred.