A Statistical Problem Concerning the Mar Saba Letter

The Mar Saba letter is a controversial document purportedly written by the second century theologian Clement of Alexandria and containing excerpts from the so-called "Secret Gospel" of Mark. There is evidence that the Mar Saba letter is a forgery. This evidence includes a statistical analysis comparing the vocabulary of the Mar Saba letter with that of the acknowledged works of Clement. This statistical analysis, which was based on Herbert A. Simon's model of text generation, was casually reported. Here a statistical framework for this analysis is presented. As a by-product, a more powerful test of authenticity is described.