Transposition Errors and Moveable Type

In yesterday’s daf (Bava Kamma 81b), we meet an Amora named Rabbi Yehuda ben Kenosa.

אֲמַר לֵיהּ רַבִּי חִיָּיא: שֶׁמָּא רַבִּי יְהוּדָה בֶּן (קְנוֹסָא) [נְקוֹסָא] תַּלְמִידִי הוּא, וְכׇל מַעֲשָׂיו לְשֵׁם שָׁמַיִם. כִּי מְטוֹ לְגַבֵּיהּ, חַזְיֵיהּ. אֲמַר לֵיהּ: אִי לָאו יְהוּדָה בֶּן (קְנוֹסָא) [נְקוֹסָא] אַתְּ, גְּזַרְתִּינְהוּ לְשָׁקָךְ בְּגִיזְרָא דְפַרְזְלָא.

Rabbi Ḥiyya said to Rabbi Yehuda HaNasi: Perhaps it is my student Rabbi Yehuda ben Kanosa. And if so, all of his actions are undertaken for the sake of Heaven; he is not acting out of haughtiness. When they reached him and Rabbi Yehuda HaNasi saw him, he said to him: If you were not Yehuda ben Kanosa I would have cut off your legs with iron shears, i.e., I would have excommunicated you for your impudence.

In the Hebrew text, they have the correction to Nekosa. But in Rav Steinsaltz’s interpolated Modern Hebrew commentary and in the derived English translation / commentary above, they keep it as ben Kenosa. I learned in a Steinsaltz gemara, he noted the variant girsa when he discussed the chaim of the person on the page.

What is happening with this variation? It is a transposition of one letter for the subsequent letter. Perhaps because the word qenas (fine) is familiar, but qesan is not?

However, I would note that this variation appears specifically in the printed texts, not in the manuscripts in Hachi Garsinan. Thus, the printings:

As opposed to the manuscripts:

Related. I also sometimes write my Jewish Link columns well in advance, and I’ve started on Bava Metzia 2a, where Ravnai / רבנאי is quoted from just a bit earlier (since all of Bava Kamma / Metzia / Batra is one tractate called Nezikin), from Bava Kamma 113, that ומצאתה implies that the lost article came into the finder’s hand. But in Bava Kamma 113, it is Ravina / רבינא who says it (in printings and manuscripts). So too in Bava Metzia 2a, in the manuscripts, it is Ravina who said it, and is Ravnai . So while I plan on including a short discourse on the rare Amora named Ravnai, this is really a scribal error.

That got me thinking, once again, about different modes of transmission and the types of errors that are common in each. So, oral transmission, you’ll have errors of sound similarity or of mental recall (for associated figures). For handwritten written, you can have orthographic similarity. Transposition of two letters seems possible, especially from a a busy sofer.

However, I think it might be the case that for printed sefarim, with the then-new technology of Gutenberg’s printing press, transposition errors will become far more common. That is, Gutenberg’s big innovation was moveable type. Instead of painstakingly making a carving or mold of the entire page, you had individual letters, when you could assemble together to make words and sentences on an entire page on a matrix, and then print that. I would guess that this could more readily lead to transposition errors. Similarly, once people began using typewriters and keyboards, typographical errors involving adjacent keys became more common.

Update: Upon further research, Ravnai is not a good example of a transposition error due to moveable type. The third parallel sugya, Bava Metzia 27a, has Ravnai:

ההוא מיבעי ליה לכדרבנאי דאמר רבנאי ומצאתה דאתאי לידיה משמע

The Gemara answers: According to the Rabbis, that term is necessary for the derivation of the halakha in accordance with the opinion of Rabbenai. As Rabbenai says in interpreting the verse: “And so shall you do with every lost item of your brother’s, which he has lost, and you have found it” (Deuteronomy 22:3), that the term “and you have found it” means that it assumes the status of a found item only when it actually enters his possession.

and we do see manuscript evidence of Ravnai rather than Ravina for this instance. So which it is still unclear which is original, and Munich 95 has Ravina in the other two parallel sugot, including Bava Kamma which is the primary sugya, whatever transposition occurred happened before Gutenberg.