The court dismissed Count I because the sentence on page 134 of Freakonomics is capable of an innocent construction. The relevant construction is based on the reasonable reader, and the court determined that the reasonable reader is based on the general population, not the academic population (though I don’t think it matters). The court did NOT say that it would have come to a different conclusion had it applied a standard based on the academic population, but one might draw this implication from the opinion.
The following excerpt comes from Robert’s Rationale, a non-academic blog I just discovered through Google (and one I don’t know much of anything about beyond this post):
The judge ruled against the first allegation. The word “replicate” does have a specific meaning in science: repeating an experiment precisely. However, the book was in large part marketed to the general public, which would take the sentence to mean that other scientists, using different methods and statistics, have come to different conclusions.
In defamation law, there’s a practice of “innocent reading,” that is, if a statement can be taken two ways and one of them is not libelous, the non-libelous one should prevail. What’s interesting here is that there are two audiences; one could read it two ways, but the other could only read it in the libelous way. The judge ruled that the general public’s reading was the one that counted, even though many academics also read the book (it is, after all, an application of econometric methods).
Lott himself makes a related point on his own blog about the popularity of Freakonomics with academics: “I think that the market for the book was also aimed at academics. The book is apparently marketed to a large number of economics classes and is read by academics.”
So, will the popular view of this case be that Levitt defamed Lott to his smaller, academic audience, but that he was not liable because he wrote the book primarily for a larger, general audience?