Well, Which Is It?

Yesterday I spent some time at the library, researching a particular aspect of Julia Child's career. I had an idea for a piece — which may or may not work out — and I needed to do some initial investigation.

Reading through a number of her biographies, side by side, I was struck by the inconsistencies among them. For instance, Laura Shapiro's slim book Julia Child says that Mastering the Art of French Cooking only sold 16,000 copies in its first year, not taking off until a year after its release. Noel Riley Fitch's detailed Appetite For Life says, "By August, less than a year since publication, Mastering had sold 100,000 copies … and was in its fifth printing."

When writing of Louisette Bertholle's royalty amounts, Fitch says that they were 18 percent (versus 41 percent each for Beck and Child) for conceiving the idea. (She did very little on the book itself.) Joan Reardon, in an article about Mastering for the Summer 2005 issue of Gastronomica, says that it was 10 percent.

These are not books about days of yore, with archivists and researchers piecing together scattered, weathered scraps of data. Some of the participants in the Julia Child story are still alive. Child herself was when Fitch's book came out in 1999. And I imagine Knopf, the publisher, still has records from that time. Shouldn't they be more consistent?

My inclination is to trust Fitch's account, if only because of the extensive detail. (You could make the case that Reardon's 10 percent is a typo; the rest of the piece lines up with Fitch's account, at least for the parts I focused on.) But Shapiro says she used Fitch quite a bit. Does she have new information about initial sales? Or is this an editing issue: Did Shapiro mean that the book only sold 16,000 copies in 1961 (it came out in October of that year)? Or perhaps her note that sales didn't take off until fall of 1962, which might have been before October, actually lines up with Fitch's account, who merely lumps the entire first-year sales together without giving a breakdown.