Regions of Interest Quantified – Solving Problems of the Analogue Methods of the Discoveries in the Judean Desert Series (Part III.4)

It has been sometime since I have posted in this series. Since I am currently on vacation, I thought I would make some time to add another installment.

In this post, let’s work our way back to an analysis of 1QSa II 11–12. As a reminder, we are interested to read and make sense of two lines of fragmentary Hebrew text. We know now how our mind reads, but we face three problems. The first problem is that these two lines are severely damaged; how we reconstruct the line implies that we make syntactical decisions that merit careful historical linguistic research. What is more, the syntactical decisions we make must factor in the space of the lacuna and as well as the spatial distance between fragments. The second problem is that we have to manipulate images for our palaeographical analysis; how we read the extant traces of ink is therefore contingent upon how we ask the computer to read the lines. The issues of methodology and conclusions are intertwined, as we need to (a) create a way to reverse engineer the text and (b) make palaeographical, syntactical, and material arguments for a particular reading or readings. And third, we need a way to document and make clear—without being overly technical—how we reached our conclusions.

In this post, let’s address the first problem. To get a good handle on the problem, we need to know the type of evidence we have and the quality of the fragments. Since the PAM images were taken over the course of several years, it is important to examine every PAM image. Once we collect the PAM images, let’s survey the evidence to see what images we want to start processing through some Computer Vision algorithms. Before we gather the images, let’s dig into some of the issues we are facing and spell out some of the methodical challenges.

Summary: Reading 1QSa II 11–12

For those who are not familiar with this text, 1QSa was discovered in the Judaean Desert of Jordan, as part of the 1947 discovery of the so-called Dead Sea Scrolls. In my doctoral thesis, I examine this text is much greater detail. Since this series is more focused on reframing philological study by focusing on digital humanities, I will not go into detail about the specifics of this text. Rather, I want to focus on 1QSa II 11–12.

In the manuscript of 1QSa, we encounter an important text relating to issues of Community Law and political tensions in the Second Temple era. In particular, lines 11–12 of column two pose an interesting dilemma. The text is fragmentary. In the first publication of this text,1 The editor, D. Barthélemy transcribed the lines 11–12 as:

1QSa II 11–12
1QSa II 11–12

Barthélemy translated the lines as follows:

Voici (l’ordre de) session des hommes de renom invités aux convocations pour les délibérations communes, au cas où Dieu 12 mènerait le Messie avec eux: Que le prêtre ait rang de chef sur toute la Congrégation d’Israel. …2

On the reading of יו̇לי̇ד, Barthélemy remarks, “Après une étude par transparence aussi attentive que possible la lecture de de mot apparaît pratiquement certaine. Mais אתם serait plus facile à expliquer si l’on admet avec J. T. Milik que יוליד est une faute de lecture du scribe pour יוליך primitif.”

Based on Barthélemy’s comments, it would stand to reason that Milik agreed with the palaeographical reading of יו̇לי̇ד. But Milik thought the scribe had fallen into error and wrote יוליד for יוליך. It is interesting to note Barthélemy’s appeal to a particular image to justify his reading. It could have been the same image referenced by John Allegro, who comments, “This word [יוליד] falls in a very badly preserved part of the fragment, and a close inspection in the Summer of 1955 gave evidence of some further deterioration. However, a special infra-red photograph taken then leaves no doubt as to the correctness of the editor’s reading.”3

The question we encounter by reading the above comments is what image were they referencing? Unfortunately, they do not tell us. We hear more about how certain images lead to an unequivocal reading by Frank Moore Cross. He remarks, “The ink is not faint; the leather is blackened, and photographing is made difficult. Those who have had access to the original—and no one is happy with the reading—have without exception agreed that ywlwd/k is paleographically fixed.”4 Cross goes on to say that he had “new enlarged infrared photographs of the section in question prepared” made in 1959.5 He continues to describe how the images have led him to make an unequivocal reading: “…the photographs are sufficiently clear, I believe, to bring to an end any further debate on the actual reading.”6 Again, no specification is offered as to what image precisely gives such a clear reading.

Tracking down 1QSa Images

If we consult the Reed Catalogue7—a must have resource if you work on the scrolls—we find several PAM images listed for 1QSa. Reed lists the following PAM images:

  • 40.059 (bottom half of col. I – taken 1950)
  • 40.060 (bottom half of col. II – taken 1950)
  • 40.062 (top half of col. I – taken 1950)
  • 40.064 (2 frags. of col. I – taken 1950)
  • 40.513 (col. I – taken 1953)
  • 40.549 (upper left-hand corner of col II – taken 1953)
  • 40.551 (upper left-hand corner of col. II – taken 1953)
  • 40.552 (upper left-hand corner of col. II – taken 1953)
  • 41.717 (upper left-hand corner of col. II – taken 1955)
  • 42.141 (col. I – taken ?)
  • 42.459 (Cols. I & II – taken ?)8
  • 42.926 (col. II – taken ?)

In addition to the PAM images, there are several images of 1QSa in the Allegro Collection9 and the West Semitic Research collections.10 The following table shows the region of interest on the PAM images:

PAM 40.062
PAM 40.549
PAM 50.551
PAM 40.552
PAM 41.717
PAM 42.926

The Analogue Methods of DJD

The above images reveal a fairly significant problem that happens when a transcription is presented apart from images. First, it should be recalled that Barthélemy claimed to have reached his reading with practical certainty. He used two dots on his reading, יו̇לי̇ד. Cross similarly appealed to language about how certain he was of his reading, also referencing a particular image but without telling his reader what image. In their recent edition of 1QSa, Bloch et al. at least inform us of the image from which they make their reading. They read:11

Bloch, Ben-Dov, Stokel

They translate the lines as follows:12

Bloch et al. make their reading from PAM 42.926. They agree with Barthélemy’s reading and they cite Cross’ discussion as further evidence for their decision. They furthermore accept Milik’s emendation.

The second problem I would like to note is how the evidence of the images problematises the overly simple methods of using a dot/circellus of the Discoveries of the Judaean Desert series—or even the 2019 re-edition of Bloch et al. The language of certainty and appealing to a digitally altered image without spelling out the methodology is rather problematic to my mind. Perhaps, one could argue that the first editors should be given some slack, but I am not convinced it would have been so hard to say what image they used and/or say a word or two about how the image was edited to achieve such certainty. Since the above images was very likely one of the images Barthélemy or Cross consulted, it is difficult to justify their confidence. This is where digital humanities can provide not only a way to analyse these images to get some understanding about the reading, but we can use something like Jupyter Notebooks where we can spell out the methodology we use to achieve our palaeographical conclusions.13

It’s clear we have before us an intriguing issue that relates to various issues of palaeography, historical syntax, and (perhaps!) political discourse of messianic ideals in the Second Temple era. In the next post, we will advance in our analysis of this text by using Computer Vision Algorithms. We have some significant obstacles to overcome before any language of certainty can be used.

  1. D. Barthélemy, “28a Règle de la Congrégation (1QSa),” in Qumran Cave I, Discoveries in the Judaean Desert 1 (Oxford: Clarendon Press, 1955), 108–177.
  2. Ibid., 117
  3. J. M. Allegro, “Further Messianic References in Qumran Literature,” Journal of Biblical Literature 75.3 (1956): 177.
  4. Frank Moore Cross, The Ancient Library at Qumran, Rev. (Garden City, NY: Doubleday, 1961), 86.
  5. Ibid.
  6. Ibid.
  7. S. A. Reed, The Dead Sea Scrolls Catalogue: Documents, Photographs and Museum Inventory Numbers, Society of Biblical Literature, Resources for Biblical Study 32 (Atlanta: Scholars Press, 1994).
  8. This is a good example why one must work with the Reed Catalogue, in conjunction with Emanuel Tov, Revised Lists of the Texts from the Judaean Desert (Leiden: Brill, 2010) and Emaneul Tov, ed., Companion Volume to the Dead Sea Scrolls Microfiche Edition (Leiden: Brill, 1995). Reed lists this PAM image whereas Tov does not include it.
  9. John M. Allego, The Allegro Qumran Collection: Supplement to The Dead Sea Scrolls on Microfiche, ed. George Brooke (Leiden: Brill, 1996).
  10. Online: Inscriptifact
  11. Yigael Bloch, Jonathan Ben-Dov, and Daniel Stökl Ben Ezra, “The Rule of the Congregation from Cave 1 of Qumran: A New Annotated Edition,” Revue Des Etudes Juives 178.1–2 (2019): 18.
  12. Ibid., 19.
  13. Online:

Leave a Reply

Your email address will not be published. Required fields are marked *