Reply to Parker's Review of The Early Text(s) of the New Testament

I have recently read David Parker's review of The Early Text of the New Testament (eds. Hill and Kruger; OUP, 2012) with great interest, since I have contributed to that volume with an essay. The review has just published in the current Journal of Theological Studies, Vol. 64/2 (October) 2013, pp. 642-45.

However, I became suprised with the generally negative critique, although Parker concludes with the reservation that, "These shortcomings in the conception of the volume should not deter anyone from reading and benefiting from individual contributions to which it has given rise."

Parker brings up three main critical points in his review, and I will attempt to give a response to each of them in this blogpost, in particular to the second and third points, as they relate to my own contribution (I assume the editors Hill and Kruger will respond somewhere else, and speak for the whole volume and for their introductory chapter):

First of all, Parker complains that the focus of the volume is difficult to discern.  Whereas the editors say it "intends to provide an inventory and some analysis of the evidence available for understanding
the pre-fourth-century period of the transmission of the NT materials, some MSS in Section II are dated after 300 and Section III lack reference to third-century writers, in particular Origen."

I think Parker is correct about a lack of focus, which probably reflects the freedom given to each author. As the editors point out in the introduction, "the editors have not asked the individual authors . . . to endorse one approach or method over another" (p. 18). Further, Parker is entirely correct in his latter observation – there is an omission in terms of patristic authors. However, I do not understand the problem with including manuscripts (papyri) dated after 300, first because the dates themselves are seldom exact. Let me take an example from my own chapter where I write in a footnote concerning the dating of P110: "The first editor W.E.H. Cockle dates P110 to the 4th cent. but at the same time notes the similarity of script to P45 (3rd cent.) and P. Flor. II 108 (ante c. 260). The evidence points to the first half of the 4th century" (p. 87, table 5.1, note j).

Further, I am certain that Parker is aware of the importance of tracing textual trajectories between the early papyri and later MSS. For example, it would be a serious omission to speak about the early text of Luke and John (chs. 7-8), while not mentioning Codex Vaticanus (B 03) because of its date after 300.

Parker's second point is as follows:

The book does not address methodology for comparing and analysing the witnesses, beyond adopting the terminology of the Aland handbook as updated in its second edition and by Barbara Aland in an article of 2002 (manuscripts are listed under the categories of ‘Strict’, ‘At Least Normal’, ‘Normal’, ‘Free’, and ‘like D’ on p. 11).

Parker then proposes that Hill and Kruger have confused this method with the Coherence Based Genealogical Method (CBGM) – "a serious misunderstanding" – when they refer to it as the "Münster approach" to classify early manuscripts "in three major groups" (although the editors are well aware of the CBGM to which they refer on p. 8 n. 38).

To this I would reply that one thing does not have to exclude the other.  The Alands' approach to analyzing early, in particular small, papyri and categorize them as "strict," "normal," or "free" should not be seen as a relic from the past now made redundant by the CBGM. In fact, Barbara Aland's last doctoral student, K. S. Min used the method to analyze over a dozen papyri in his excellent dissertation on the earliest transmission of the Gospel of Matthew, published in 2005 in the ANTF series (a series now edited by Parker). By that time the CBGM had been applied for many years. Arguably, this method of analysis, as described by Barbara Aland in the Festschrift for Delobel, New Testament Textual Criticism and Exegesis (Leuven, 2002), pp. 1-13, and elsewhere, can justifiably also be labeled a "Münster approach" (although I would personally avoid it).

To take this one step further, one could actually question whether the other Münster approach, now in vogue, the CBGM, is a suitable tool for analyzing the text of fragmentary papyri. As Gerd Mink, the inventor of the method himself states in a very helpful essay, "Problems of a Highly Contaminated Tradition: The New Testament. Stemmata of Variants As a Source of a Genealogy for Witnesses,” in Studies in Stemmatology II (ed. Pieter van Reenen, August den Hollander and Margot van Mulken; Amsterdam: John Benjamins 2004), 71:

Badly fragmented witnesses certainly present a very special risk. The genealogical relationship between a fragmentary witness and another witness is determined on the basis of its limited text supply. As the distribution of prior and posterior variants in a text can be very different it is possible that an entirely different picture would have resulted from the complete witness, if it had been preserved. 

It is perhaps not a coincident that it is possible to exclude fragmentary MSS when running the CBGM software. In an extensive presentation (downloadable here) of his method as applied to the Catholic Letters, Mink explains further how the method does not take fragmentary MSS into account, and it seems like many papyri actually fit the criterion he indicates:

The pre-genealogical and genealogical data used rests upon the entire corpus of the Catholic Letters (not upon single writings). When agreements with a given witness are checked or potential ancestors are searched for, smaller fragments are excluded. Smaller fragments are those which share less than 50% of the variant places in the given witness.

Hence, when Eldon J. Epp in his important essay "Textual Clusters: Their Past and Future," in The Text of the New Testament in Contemporary Research (eds. B. Ehrman and M.W. Holmes; NTTSD 42; Leiden: Brill, 2013), 564, asks what is the appropriate method for identifying the D-cluster ("Western" text), he questions the usefulness of the CBGM, since it does not take into account fragments (the D-cluster contains the fragmentary P38 and P48).

On the other hand, Epp gives a rather positive evaluation of Alands' system of classification for the papyri in another essay in the same volume, "The Papyrus Manuscripts of the New Testament," in The Text of the New Testament in Contemporary Research, 27, 

Yet, this classification system, with its considerable relativity, is not without value, for it reflects the general scribal process. For example, the close textual relationship between P75 and Codex B . . . does show a 'strict' connection with their respective exemplars–and with others along that line of transmission.

The final and probably most severe point of criticism in Parker's review concerns the second section of the volume, including my own essay:

I find it really remarkable, and a serious flaw, that none of the contributors to Section II seems to pay attention to, or in some cases even to be aware of, the Text und Textwert series (I except the chapter on the catholic epistles). These volumes make it possible to compare a witness with any other witness in a systematic manner, and to take an important step towards an overview of the textual history of a work. They are far more valuable, since they provide hard facts, than the Aland categories. The first volumes were published in 1987. How can what purport to be articles dealing with contemporary scholarship ignore them?

I have several objections to this harsh criticism, especially if it is applied to my own essay, "The Early Text of Matthew" (pp. 83-107). First of all, in my introduction, I explicitly point out that "numbers and percentages are more important than corresponding labels like "free," "normal," and "strict" and that "the validity of the results ultimately depends on the size of the sample and the specific nature and pattern of textual variation–variants should be weighed as well as counted" (p. 84). In my opinion, I use a proper method providing facts as hard as they can be in this business.

More importantly, I do not see how I could have used Text und Textwert in order to analyze most of these small papyri. When I wrote my chapter, I of course had to relate it to Min's recent and extensive examination of the Matthean papyri, and I felt that I could improve on some aspects, and in a few cases I came to slightly different conclusions regarding the classifications. But then I basically had to use the same method on the material.

In Min's case, Barbara Aland, director of the very institute that produced the Text und Textwert series, apparently did not see the need to use that tool either, because Min never refers to it in his monograph, as far as I can tell. Instead, as I have pointed out, Min uses the method of analysis which Aland continued to develop and describe (long after the development of the CBGM and the Textwerttool).

If we turn to one of the most significant examinations of NT papyri, that of James R. Royse in his magnus opus, Scribal Habits in Early Greek New Testament Papyri (NTTSD 36; Leiden: Brill, 2008), he does not use the Text und Textwert tool either, as he analyses six of our most significant and extensive papyri. On the contrary, he is entirely occupied by examining how well the scribe copied the exemplar. In his introductory chapter on methodology, we may further note that he gives a rather positive evaluation of Barbara Aland's method of analysis (pp.60ff), although he chooses to analyze the singular readings only (which is particularly useful when dealing with larger papyri).

In Parker's recent essay, “Understanding How Manuscripts are Related” in Textual Scholarship and the Making of the New Testament. The Lyell Lectures, Oxford, Trinity Term 2011 (Oxford: Oxford University Press, 2012), 65-100, in which he describes ways of understanding relationships between MSS, he does not explicitly mention the Text und Textwert tool (although the index claims so), albeit there is a brief reference to "comparison of textual data in test passages in the Gospels." Instead, he rightly focuses on the CBGM, referred to as the "Münster Method."

Apparently, Parker concentrates on the method he thinks is most useful to apply to the material in order to solve the research question. In my own essay in the volume under review, I have tried to do the same thing. I did not mention (nor used) the Text und Textwert tool because I preferred to use a different "Münster method" which I find more suitable for analyzing early (and mostly fragmentary) papyri.

I sum, I think Parker is correct when he identifies a general lack of focus in the volume under review; this is because the individual authors were given great freedom in terms of method and approach. Further, Parker has rightly identified the omission of references to third-century writers (e.g., Origen), but I object to the other points of criticism.