Colwell, Direct Copies, and Singular Readings in Dante

Paolo Trovato has alerted me to an interesting article just published in the latest Storie e linguaggi by one of his doctoral students, Federico Marchetti. The article grows out of his doctoral work and considers manuscripts of Dante that have known exemplars (Abschriften in NTTC terms). In the course of studying these, he addresses the use of singular readings as a way of determining scribal habits. This method has had its critics over the years (e.g., Strutwolf, Jongkind, myself, and now Hixson). Marchetti also criticizes it given his own results because it includes false data. Here is his conclusion:

If Po [the exemplar] had been lost, and if we had decided to investigate the scribal behaviour of Est [the copy] according to the Colwell method (that is, by calculating the percentage of textual agreement with the rest of the surviving manuscript tradition), we would necessarily have considered the scribe of Est responsible both for his own errors (which affect 0,06% of the verses) and for those that are ascribable to earlier stages of the tradition (which affect 17,6% of the verses). It is evident that such a diversity in approaches leads to a dramatic discrepancy in outcomes and their degree of reliability. Applying the Colwell method, we would certainly talk about a “scribal version” or “scribal revision” in this case, but – as we have just seen – the copyist of Est is in fact a very diligent scribe, careful about the formal features of his model as well as about its substance. In conclusion, the data collected bear out the necessity of making use of the genealogical-reconstructive method; more in general, they advocate the validity of the Neo-Lachmannian approach.

What I didn’t see in Marchetti’s article is a recognition that Royse, for example, fully recognizes this limitation or that Royse incorporates these false positives under his rubric of “the complex scribe.” But, as I have argued in my own work, the problem of deducing scribal habits from singular readings can’t simply be solved by appeal to the complex scribe since most scribally-created readings are not singular at all. I suspect part of the issue here for Marchetti is that he is after a different goal than determining principles of transcriptional probability (Colwell’s main goal). The NT, of course, is highly contaminated which is why Lachmannian approaches, whether neo- or not, have generally not been used.

In any case, my thanks to Paulo for alerting me to this essay. The entire issue will be of interest to our readers and can be found online here with abstracts.

New Contributor: Elijah Hixson

When I first started learning about textual criticism at seminary, textual criticism was not much more than one lecture on Romans 5:1 in intermediate Greek, and if you were lucky, an introductory course that was only taught once every few years. I had to wait a few years for the introductory course. My intermediate Greek professor pointed me to Metzger and Ehrman’s Text of the New Testament. After that, I found this blog and never stopped reading it. It was like drinking from a fire hose at first, but I am deeply thankful for all the posts and comments here over the years that have helped make me who I am. It is an honour to be a contributor to this blog.

I recently finished my PhD at the University of Edinburgh under Paul Foster. James Snapp recently interviewed me about this research over at his blog, The Text of the Gospels, but I will give a summary of my research and results here. Like Alan Taylor Farnes, I too decided to test the “singular readings method”. Whereas Farnes did that with Abschriften, I took a closer look at the sixth-century purple codices 022, 023 and 042. Because these three manuscripts were copied from the same exemplar, I reconstructed the text of their exemplar in Matthew’s Gospel where that was possible.

Now, I know that scholars like Royse, Hernández and just about everyone else who uses the method are careful to qualify it—singular readings really tell us about the “complex scribe” not the actual scribe. Still, after a lengthy discussion of the matter, Royse concludes that the singulars are probably due to the actual scribe in most of the cases. His words are:

In what follows, consequently, I will speak of a manuscript’s “scribe” in the ordinary way, that is, meaning the person who actually wrote the manuscript. Discussions of the scribe’s handwriting or corrections, for instance, will obviously refer to this one person. And most of the singulars should, without doubt, be attributed to this person. (p. 55)

In nuce, singular readings hypothetically could be from anyone, but they’re probably the work of the actual person who made the manuscript. If anything, singular readings tell us about the activity of the person who made the manuscript with some contamination from previous scribes in the line of transmission back to the archetype.

But is that really true?

By focusing on singular readings, one could include inherited readings from the exemplar and exclude non-singular readings created by the scribe. Those possibilities allow errors on both sides of the data. I set out to test the method in a three-tiered approach.

First, I went to all the places in Matthew where 022, 023 and 042 are all extant and compared orthography (ει/ι and αι/ε interchanges), unit delimitation, kephalaia and titloi, the Eusebian apparatus and textual changes. This comparison allowed me to build a preliminary profile of each scribe to help resolve issues reconstructing the exemplar later one where only two of the three manuscripts were extant. For example, the scribe of 023 is incredible and makes very few changes, but the scribe of 042 has a noticeable tendency to harmonise Matthew to Markan parallels. Therefore, if only 023 and 042 are extant—and they differ—, and one of the possibilities is that 023 preserves the text of the exemplar and 042 harmonises to Mark, then that is probably what happened (as opposed to 042 preserving the text of the exemplar and 023 making the change).

Second, I analysed the singular and family readings of each manuscript (readings unique to these three manuscripts). I did this to include “inherited singulars”. Of course, the inherited readings aren’t singular in my cases, but that is only because we have more than one copy of the exemplar. Stated alternatively for 022, I studied the singular readings of 022 as we would count them if 023 and 042 never existed. This modification best replicates the situation for any other early manuscript, like the early papyri studied by Royse.

Third, I reconstructed the 022-023-042 exemplar, analysed the changes each scribe made to the text and compared these numbers with what I got from the modified singular readings method.

The results? An analysis of singular (and family) readings of 022, 023 and 042 does not give accurate conclusions about the scribes who made them. In fact, if you add up the total number of singular and family readings from the three manuscripts, about two-thirds of them are inherited readings, not scribal creations. If you add up the total number of scribal changes in all three manuscripts, about two-thirds of them are non-singular. Instead of getting the habits of the actual scribe with a bit of contamination from the complex scribe, with the purple codices, the unique readings tell us more about the “habits” of the complex scribe with a bit of contamination from the person who actually wrote the words. I wonder if even that is accurate because of how many non-singular scribal changes went unnoticed.

Even though the method doesn’t “work” with 022, 023 and 042, it still might tell us something about scribal habits in the early manuscripts. Several of the changes I saw were instances in which scribes aligned the text to what would become the majority reading, and I don’t think that would be the case for the early papyri—certainly not as much. I’m not saying to throw the method out. It still tells us how manuscripts are unique, even if in these three instances it fails to tell us about scribes.

In the end, the project was a lot of fun. I learned a lot about three sixth-century witnesses to Matthew’s Gospel and their scribes. The manuscripts themselves are gorgeous—they were a good choice of manuscripts to spend 3.5 years looking at. I also highly recommend Byzantine manuscripts to people looking for thesis topics. Even though I am not a Byzantine prioritist myself (though I have the highest respect for our Silver-Haired Assassin), it is exciting to notice new things by working with manuscripts that have been largely neglected since their discovery and initial publication.

Colwell’s Reversal of Westcott and Hort on Singular Readings

E. C. Colwell (source)

Westcott and Hort and E. C. Colwell are often connected in discussions of singular readings as they should be. But it’s not often appreciated how Colwell inverted WH’s primary purpose for studying singular readings. 

For WH, their main interest in singular readings was that their two prized manuscripts (01 and 03) often stood alone from the rest of the textual tradition as they knew it. Singular readings did give them access to the proclivities of 01 and 03 (a la Colwell), but this was not then aimed at elucidating transcriptional probability like it was for Colwell. Instead, this exercise was aimed squarely at determining which of 01 and 03’s singular readings had “a better title to consideration,” i.e., of being original (Intro, p. 233).

For Colwell, the study of singular readings is not only to “increase skill in the evaluation of that manuscript” (a la WH), but also “to gain knowledge of the habits of a scribe in general ... and thus to increase skill in the evaluation of readings.”* For Colwell, such a study was not aimed at identifying which singulars may claim originality. Rather, it was to give us access to those readings that we can be sure are the scribe’s rather than the author’s. 

Thus, unlike WH, Colwell’s interest in singulars hinges entirely “on the assumption that these readings are the creation of the scribe.” As such, they give us access to the work of the scribe as opposed to the author. In this, Colwell has reversed WH for whom the study of singular readings was aimed precisely at identifying those singulars that are the work of the author rather than the scribe. 

In other words, both recognize that singular readings can attune us to the scribe’s habits, but each uses this fact to focus on opposite sides of the scribe/author divide. WH are ultimately interested in authorial readings whereas Colwell is really interested in scribal readings. Obviously, this has to do with whether each thinks that singulars may be original. But the difference is still important especially as it comes to “skill in the evaluation of readings.” WH have their own view of how to best attain that and it has nothing to do with singular readings. But more on that another time.

---

*E. C. Colwell, “Method in Evaluating Scribal Habits: A Study of P⁴⁵, P⁶⁶, P⁷⁵,” in Studies in Methodology in Textual Criticism of the New Testament, NTTS 9 (Leiden: Brill, 1969), 108.

The Types of Singular Errors in the Book of James

I recently finished working through all the singular readings in James according to the Editio Critica Maior (ECM). There are 1,245 of them in my dataset. I picked James because it has the most manuscripts of all the books in the ECM of the Catholic Epistles.

One result from my data is the various types of changes that occur among the singular errors. Here “error” (Fehler) is any reading that didn’t make either logical or grammatical sense to the editors (p. 27*) and “singular” is defined relative to the Greek manuscripts in the ECM. There are 493 such singular errors (39.6% of all singulars).

Using the basic categories from Royse, the singular errors in James divide as follows:

As you can see, the vast majority are substitutions (84.5%). Unexpectedly, additions outnumber omissions almost 2:1 (9.6% vs. 5.0%). This is not true of singular readings as a whole in James where, as in Royse’s study, omissions outnumber additions. You could make an argument that singular errors provide us with greater “moral certainty” (to use Hort’s phrase) about their being scribally created than non-error singulars since nonsense is less likely to be copied than sense. But I find that there is a certain trade off in studying scribal change between the guarantee of purity in the data (“moral certainty”) and the volume of the data. At some point, more data that’s less pure is actually better (i.e., more representative) than less data that’s more pure. But more on that another time.

To give a flavor of some of these errors, here are the first two verses of James.

Error] correct reading	Error #	Address
δου] δουλος	f	1.1/14
τη διαασπορα] τη διασπορα	f1	1.1/26–28
φερη] χαιρειν	f	1.1/30
χαριαν] χαραν	f	1.2/4
πειρασμος] πειρασμοις	f	1.2/14
περιπεση] περιπεσητε	f1	1.2/16
περιπεσητω] περιπεσητε	f2	1.2/16
ποικιλλοι] ποικιλοις	f2	1.2/18

I should also note that I am counting cases of these changes such that if two words are replaced, this counts as one substitution. Likewise, a five word addition would count as one case of a “2+ addition.”

An interesting study would be to consider the non-singular errors alongside the singular errors to see if there are any patterns that might help explain why some were made more than once (assuming they are non-genetic). If anyone is interested, you can download the spreadsheet with all 493 singular errors.