1995 Andrew Criddle Paper: On the Mar Saba Letter

Criddle’s 1995 note treats the Mar Saba letter as a statistical outlier within Clement’s corpus and then asks how far such an outlier, by itself, should push us toward inauthenticity. Using standard observations about how an author’s total vocabulary expands, he argues that in a genuinely new Clementine piece one should usually see more brand-new words than reappearances of earlier hapax legomena; in the Mar Saba letter, the opposite happens, with a small text reusing an unusually high number of Clementine one-off words and introducing surprisingly few new ones, which he reads as moderate evidence for a forger artfully sampling Clement’s rarities.

Criddle - On the Mar Saba Lette… The rejoinders emphasize how fragile that inference is: the sample is tiny; the letter’s content deliberately overlaps Clement’s pet themes; the “expected” ratio may not hold across genres; and there is no independent control showing that such clustering of scattered hapax cannot occur naturally. Behind the technical back-and-forth sits a broader caution: stylometry can sharpen suspicions, but in a short, provenance-less piece it cannot, by itself, deliver a courtroom-grade verdict, and prior probabilities (how likely one thinks an authentic new Clement letter is) heavily color how much weight the anomaly seems to carry.

A useful older reminder about this kind of evidence comes from A. E. Housman, writing as a textual critic in 1921: “A textual critic engaged upon his business is not at all like Newton investigating the motion of the planets: he is much more like a dog hunting for fleas.” The Mar Saba debate, on this view, shows fleas being found—odd lexical ratios and stylistic quirks—but also how easy it is to mistake the irritation of an anomaly for proof that the whole animal is spurious.