Wednesday, July 25, 2007

Translation Issues and the UBS Rating System

In her recent discussion of the translation issues surrounding 1 Cor. 14:30-40, Suzanne McCarthy briefly refers to the rating system employed by the editors of the United Bible Societies (UBS) Greek New Testament to indicate the probability of various readings. According to Bruce Metzger's A Textual Commentary on the Greek New Testament (p. 14*), the general outline of the system is as follows:

A: the reading chosen by the editors is "certain"
B: the reading chosen by the editors is "almost certain"
C: the editors "had difficulty in deciding which variant to place in the text"
D: the editors "had great difficulty in arriving at a decision" (relatively rare; e.g., Matt. 23:26, John 10:29)

In the case of 1 Cor. 14:30, the placement of vv. 34-35 was assigned a "B" rating; a few uncial codices (D, F, and G) and a handful of Latin witnesses place them after v. 40. McCarthy notes that while this rating leaves some room for doubt, the transposition of vv. 34-35 is rarely footnoted in English Bibles. While I haven't done any substantial research on this particular point, it served to remind me of a larger problem: the seeming arbitrariness of the rating system. Rom. 5:1 is an even more glaring example. Here is the complete verse, as found in the UBS/Nestle-Aland text:

Δικαιωθέντες οὖν ἐκ πίστεως εἰρήνην ἔχομεν πρὸς τὸν θεὸν διὰ τοῦ κυρίου ἡμῶν Ἰησοῦ Χριστοῦ

In the NRSV, the verse is rendered, "Therefore, since we are justified by faith, we have peace with God through our Lord Jesus Christ"; many other major English translations adopt similar formulations. However, the original hands of the earliest and best manuscript witnesses (א, A, B, C, D, etc.), along with some patristic witnesses (particularly the recorded teachings of the early heretic Marcion of Sinope, which date to the mid-second century), replace ἔχομεν ("we have") with εχωμεν ("let us have"). In summarizing the UBS editors' preference for the former, Metzger notes that in antiquity the two terms were phonetically identical, and suggests that the the early occurrences of εχωμεν may have been the result of a simple scribal misunderstanding. The decision was assigned an "A" rating.

Whether one concurs with the editors' conclusions or not, it seems misleading to state that a particular reading is "certain" if it is not found in the most important and most frequently cited manuscripts. Some Yale professors encourage their students to simply ignore the rating system, but such blissfully intentional ignorance isn't always an option. The revised edition of the Guidelines for Interconfessional Cooperation in Translating the Bible, a document produced by the Vatican in 1987, states that translators should follow the UBS text in the case of an "A" or "B" rating. If entire denominations are placing their trust in this system, then it seems reasonable to request that it be based upon consistent, clearly defined principles, and that it reflect the formidable difficulties which hinder the reconstruction of the "original" New Testament text.


Zephyr said...

I agree with your final suggestion that the UBS rating system be based on consistent, clearly defined principles. However, the Vatican's Guidelines does allow for following a C rating. From section

"Translators should normally follow this text for readings rated as A or B in The Greek New Testament but may choose other well attested readings when the text has a C rating."

Matthew Burgess said...


The Vatican Guidelines do allow for following a "C" rating, but what good is such an allowance if many ratings have been artificially inflated? Regardless of one's opinion regarding the "correct" reading of Rom. 5:1, the distribution of the manuscript evidence warrants a more cautious rating. But because the UBS editors did not exercise such caution, some translators are prohibited from adopting a reading supported by our earliest and best manuscripts, and most readers will never know that a major textual problem exists here. And I'm sure that other passages are affected by the same issue. That was my main point in this post... the problem of "grade inflation." ;-)