Anyone interested in the textual criticism of the New Testament knows about the so-called Freer Logion. This is a passage found in only one known New Testament manuscript, Codex Washingtonianus (ca. 400), also called the Freer Codex after the art collector who bought it in the beginning of the twentieth century. The Freer Logion can be found in Mark 16, in between vss 14 en 15. So we are talking about words that were inserted into the unauthentic 'longer ending'. The Freer Logion starts as follows:
And those ones excused themselves, saying, “This age of lawlessness and unbelief is under Satan (ὑπὸ τὸν σατανᾶν), who does not permit ... (transl. Craig Evans, WBC)Although the Freer Logion became known as a manuscript reading only in the twentieth century, the first half of it was already known from the very beginning of the textual criticism in the age of the printed book. In the annotationes to his 1516 NT edition, Erasmus refers to Jeromes Dialogues against the Pelagians. Here (in section 15 of book 2) Jerome quotes, in Latin, an addition to the ending of Mark he says can be found especially in Greek copies.
Well then, while I was reading the Freer Logion in the text of Erasmus, I was surprised. Something was missing there, that is, Satan was dropped. The Freer Logion in the annotationes of Erasmus:
Et illi satisfaciebant dicentes: seculum istud iniquitatis et incredulitatis substantia est, quae non sinit ...These words may be translated as something like:
And they excused themselves, saying: this age 'consists of' (?) lawlessness and unbelief, which does not permit ...I'm not sure whether the way 'substantia' (for which Lewis & Short offer, among others, the meaning 'that of which a thing consists') is used here, is awkward or not. But apparently, this is what Erasmus found in his manuscript or manuscripts of the Dialogues, and also what he prints in his edition of Jerome. Moreover, until the end of the twentieth century, this is printed as text in every edition of Jerome's Dialogues.
It is not so difficult to see what went wrong in the transmission of the Latin Freer Logion. SUBSATANA ('under Satan') was misread (or misheard) as SUBSTANTIA. The following relative pronoun, the masculine 'qui', was subsequently changed into the feminine 'quae' to make it correspond with 'substantia'.
In fact, Migne, who published Jerome's works in the second half of the nineteenth century, knew about a Vatican manuscript that had the reading 'sub Satana'. Migne acknowledges that this reading makes much more sense than 'substantia'. However, since he thought this manuscript continued with the wrong pronoun 'quae', he did not dare to adopt this reading, and to print, without manuscript evidence, the right pronoun 'qui'. In other words: he did not want to resort to conjectural criticism. Migne's reluctance is typical for a theological tradition of fear to go beyond the manuscripts.
Migne's remark ended up in the Nestle-editions. E.g. in the most recent 28th edition one finds in the apparatus at Mark 16, in the quotation of Jerome, 'substantia (= sub Satana?)'.
A final twist, then. Migne was mistaken. The Vatican manuscript from which he knew about the reading 'sub Satana' simply continues with the right pronoun: qui. In fact, the oldest manuscript of the Dialogues known today, a 6th/7th century codex in possession of the library of Lyon, also has 'sub Satana..., qui'. This is also the reading of the modern critical edition, the 1990 CCSL volume of Moreschini. Therefore, it no longer makes sense to refer to 'sub Satana' between brackets, like Nestle-Aland does. (I have been told that in the forthcomoning third print of NA28 'sub Satana ..., qui' will replace 'substantia ..., quae'.)
Moral from this story? I would say: strange things can happen in the transmission of ancient texts. And to this I would add, as a student of New Testament conjectural criticism, we should not reject a conjecture too quickly, simply because it is too brilliant to be true. Imagine we would not have Codex Washingtonianus, nor Latin manuscripts with 'sub Satana'. And then a clever scholar would conjecture 'sub Satana'. Probably we would smile, appreciate the creativity, but reject it as too far-fetched, and point at the pronoun that needs to be changed to show that the scenario of textual corruption is highly implausible.
Again, strange things happen.