Who Is the Real Author? A Computer Finds Out

In January 1996 Primary Colors, the anonymous novel by a political insider, was released by Random House. Thanks in part to curiosity about the author, it was on the New York Times best-seller list 20 weeks, nine weeks as number one. At mid-year 1.2 million hardbacks had sold, and it was being translated into foreign languages. Warner paid $1.5 million for paperback rights, and Mike Nichols paid over $1.5 million for movie rights. (Elaine May is writing the script, which will star Tom Hanks and Emma Thompson.)

As early as 19 February New York Magazine published an article claiming that the author was Newsweek columnist Joe Klein, as demonstrated by Vassar professor Donald Foster’s computer analysis of Klein’s language pattern. Klein vehemently denied the charge, but in July an incriminating letter was revealed in the Washington Post, and he was pressured into confessing. As he ruefully acknowledged in his 29 July column, “New York Magazine hired a Vassar professor with a computer to analyze the style of various suspects. It was a pretty good program.”

That is the kind of admission that some wish Walter Hooper would make about the central forty percent of “Christian Reunion,” in which Lewis allegedly denied disagreement with any Roman Catholic doctines. A.Q. Morton’s computer analysis identified that language pattern as Hooper’s, not C. S. Lewis’s.

Ascertaining authorship with computer analysis gets little attention in the press because (like DNA analysis) it is complicated. A superb new guide to the subject should be acquired by all major libraries: Analysing for Authorship by Jill M. Farringdon (Cardiff: University of Wales Press, 1996). See p. 2.

Farringdon’s mention of C. S. Lewis is brief, but “The Dark Tower” appears in a montage of headline fragments on the handsome blue dustjacket.