Your interesting observation raises many questions about the practices of the editors of this text in the past. That they should have missed this reading in so many manuscripts staggers belief.
However, I doubt that “committentes” is the correct reading. The author – incidentally, I don’t believe that it was Jerome – claims that some scribes or translators write down (“ponentes”) the water, spirit and blood, but miss out the Father, Word and Spirit. The only possible sense I can wrest from “committentes” is that some scribes or translators “commit to writing” the testimony of the Father, Word and Spirit. “Committentes” would thus mean the same thing as “ponentes” – but then of course the author’s pointed contrast disappears. He is unhappy that some authors leave out the Father, Word and Spirit, not that they also commit these words to the page. Arguing simply from the sense of the passage, I think “omittentes” must be the correct reading.
There are several possible ways in which the “o-” could have been read as “com-”. Once this had happened in one manuscript copy, it was likely transmitted to further copies unless later scribes intervened. It would be interesting to trace the variants in the entire text of the prologue in the various manuscripts to see if this variant could be isolated to a particular textual family. Of course, such an error could have happened independently more than once.
Firstly, the “c-” might have crept in through visual similarity with the “o-”. But this still leaves some details unexplained.
Alternatively, it is possible that Fuldensis (or its archetype) was copied from a defective exemplar in which this word was illegible or damaged (a real possibility if the archetype was written on papyrus). The scribe of Fuldensis might then simply have guessed at the missing letter or letters.
A third possibility is perhaps the most plausible: it is possible that an early scribe mistook the “o” for an abbreviation. The Tironian sign for “con-” or “com-”, preserved as an abbreviation in many different kinds of Latin hands through the middle ages, looks like a reversed “c”, which is easily confused with an “o”. (See Ulrich Friedrich Kopp, Tachygraphia veterum 2 [= Palaeographia critica vol. 2.2], p. 52). This would explain the misreading quite economically.
Then, as you have shown in some of the manuscripts, some later readers corrected “committentes” to “omittentes”, because they clearly realised that “committentes” just doesn’t make a whole lot of sense.
I’m so glad that you brought this to our collective attention. Even if we seem to be dealing with a rogue reading, you are quite right to ask what the absence of this variant from any critical edition says about editorial standards in the past. It is possible that the judgement of the editors of yesteryear was compromised by short library opening hours, poor lighting or failing eyesight. (Those filthy splotches in Fuldensis are caused by a reagent used in the nineteenth and early twentieth century to bring out faded letters, which shows that an early editor was really struggling to read this page.) However this oversight might have happened, I hope your discoveries encourage us to do better in the future.