Friday, July 06, 2018

The CBGM Applied to the Old Testament

I often get asked if the CBGM could be used on the text of the Old Testament. The short answer is yes, it could be. There is nothing in principle that excludes any tradition from being used with the CBGM. The better question, of course, is should it be and I have had mixed thoughts about that. Well, my mixed thoughts aside, now it has been.

I have just come across an MDiv thesis completed at Wisconsin Lutheran Seminary earlier this year by a fellow named Dean G. Ellis. I’m traveling and have only had a chance to skim it, but I thought I would alert our readers to it.

The title is “Applying the Coherence-Based Genealogical Method (CBGM) to the Text of the Old Testament: An Evaluation.”

Here is the abstract:
The goal of this research is to apply the Coherence Based Genealogical Method (CBGM) to the Old Testament. Deuteronomy 5 was chosen to evaluate this method. This method is being used to evaluate General Epistles in the New Testament and will result in changes to the Editio Critica Maior of the New Testament. To date, this method has not been applied to the Old Testament. This study relied on the development of new software algorithms to align the Hebrew text and perform the CBGM analysis. Initial results indicate that this method is applicable to Old Testament texts and is able to propose a model for the transmission of the text. Textual relationships were identified, and a proposed route of textual transmission was determined. This method has many promising applications within Old Testament textual studies. It also has several strengths and weaknesses that are addressed.


  1. Michael Dormandy7/06/2018 4:41 pm

    I'm just on the way back from the (excellent) OT and LXX Textual Criticism Summer School, with Kristin de Troyer, in Salzburg, so I'm thinking about these issues. My first thought is that, as I understand it, CBGM is difficult to do accurately with witnesses in different languages. As I understand it, the versions are much more important for the OT than for the NT when it comes to reconstructing the earliest possible form of the text, because many versions are much older than our oldest Hebrew MS. As for doing CBGM to compare our Hebrew MS, from what I have read and heard, the Hebrew consonantal text does not vary much between our MS, so CBGM is not necessary. I'm aware I know a lot less than a lot of people here on these issues though, and would welcome correction!

  2. Paolo Trovato7/08/2018 9:47 pm

    Dear PG, dear all,
    one of the attendees of my 4th Summer School in Textual Criticism which ended last saturday, a young graduate from Brasil, was deeply interested in CBGM which is the topic of his PhD thesis. For this reason I ws forced to read more than once and discuss with the attendees of the School your 2017 Introduction to CBGM (A New Approach to Textual Criticism…) and I must admit that notwithstanding your precious book and the fact that I have been addressing since 1978 problems of medieval textual criticism in Italian, French, Spanish and Latin texts, I found the method rather opaque and even misterious in some respects, while almost all the methodological notions which are presented as CBGM discoveries were already “discovered” in previous Lachmannian and Neo-lachmannian research (I ’ll spare you the details for the sake of brevity). The most surprising (and dangerous) innovation is, on my opinion, the refusal of the notion of family or group of witnesses simply because the labels (Alexandrian etc) are unfit. This means that you can not ever reach a clear idea about the genalogical relationships among text types. In the rest of the world we use letters (a, b, c or alpha, beta, gamma…) and this very simple choice allows us to apply the majority principle, a logical tool which is precious to address enormous sections of text which do not present peculiar problems.

    As the text at stake (NT) is a paramount one for every human being and the only way to see if a method works, is to compare its outcome with the outcome of other methods, I wonder if somebody of your scholarly community and institutions is interested in the following not so expensive test. You will choose and fund 2 phD graduate students. The 2 Ph.D students will work respectively on Catholic Letters and Act according to commor error method (= Neo-Lachmannian method). We make an agreement between universities and they register in my University (Ferrara) and in Munster (each of them will cost about 30.000 US$ for any of the 3 years). The 2 “doctorandi” arrive in Ferrara, we briefly form them in Neo-Lachmannian TC (about two months), then they go in Munster 4 months and learn how to use at their best the data banks with the variants of the Catholic Letters and Acts (they will focus only on types 2 and 3 of the grid elaborated by Alberto Cantera Glera: see Gurry and Wasserman 2017, 116) and discuss with supervisors in Ferrara the outcome. I ’ll put together a little committee of supervisors (textual scholars, specialists in NT Greek, scholars of early Christian texts from Ferrara, Padua and Bologna) who will volunteer, i.e. will work for free, in order to assist the 2 Ph.D students in their research. At the end of the first year we will organize in streaming a first workshop about the research. PhD students and revisors will compare samples of texts and provisional genealogical diagrams (stemmata) and start to discuss with everybody convergent and divergent issues of the 2 methods. Best.

  3. Paolo Trovato7/09/2018 5:19 am

    P.S. About 30.000 US$ is simply the annual grant for a PhD student in Italy. A very little part of this amount (less then 2.000 US$) is used for fees and insurance