Introduction to the Legacy System of DJD
In the previous post of this series, I introduced how the early editors of the Discoveries in the Judaean Desert (DJD) series adopted a sigla system to annotate characters. The sigla system served to affix additional criteria to a transcription as a DJD editor read a fragment. Thus, the DJD system was designed to communicate the psychological certainty of the editor. In the successive editions of DJD, this sigla system did not change all that much, and one can find the system explained in DJD 12 (as well as a few other DJD volumes). The sigla that directly relate to letters can be summarised as follows:
|◦||ink traces visible but letter uncertain|
|[ ]||lacuna in fragment or surface of fragment is damaged|
|vacat||interval in the writing of the scribe, space was intentionally made|
These sigla are given further definition in DJD 12. It is worth quoting Eugene Ulrich in full on the understood function each siglum has when used in a transcription.
Letters that are ‘certain’ or virtually certain are simply transcribed. Letters are considered ‘probable’ if the ink traces very probably [sic] form one particular letter though they could also form a second or third letter with features identical to the preserved strokes; these are indicated by a dot above the letter (e.g., ב̇). Letters are considered ‘possible’ if some ink is preserved and it conforms to the suggested letter but could also form any of several other letters sharing that feature; these are marked with a circlet above the letter (e.g., ב֯). Claims made for letters so marked can be only as solid as the basis supporting them. ‘Unidentifiable’ letters, that is, those for which ink remains on the manuscript but which cannot be indentified [sic] with confidence, are indicated by circlets in the middle of the line (e.g., ◦). As regards [sic] the unidentifiable letters, an attempt has been made to reflect the number of letters by the number of circlets, but this remains ambiguous, because at times it is difficult to determine whether, for example, two ink traces formed parts of one letter or of two.1
From the above quote, we see that letters marked with a siglum of probable (א̇) and possible (ב֯) are traces of ink whose palaeographical interpretation is ambiguous because of undefined features required for our brain’s letter box to make sense of the geometric contours. What is this idea of a brain’s letter box? The ‘letter box’ is the colloquial expression used for the anatomical region of the brain known as the left occipito-temporal area or visual word form area. Reading, in other words, begins with a visual recognition which triggers two other parts of the brain, which have to deal with phonetics and semantics of language. One part deciphers semantics and the other part deciphers speech sounds. The parts of the brain that relate to semantics and speech sounds are formed in us as kids, but when we learn how to read, our brain creates an interface between the semantics and speech sounds vis-à-vis the letter box.2
With the DJD system, the difference between using a possible and probable siglum can directly depend on the Hebrew/Aramaic proficiency of the editor, in terms of exposure to spoken and written Hebrew. Based on what we know about how the brain reads, the system does not purport to say anything about the material features of the character. Some editors, however, did use the sigla as a way to comment on the material quality of the reading. In DJD 40 (1QHodayota), for example, the definition of the sigla was completely changed. The editors used the circlet (e.g., ג֯) to indicate that “there is damage, whether the reading is certain or uncertain” (DJD 40:xv).
The DJD system is thus designed to communicate the psychological certainty of an editor—even in the case of DJD 40. But is it a good system to keep in a digital edition? Yes and no. It’s a good system to keep depending on the relationship between a transcription and an image of the transcribed artefact, but the system has many flaws for it assumes too many things about the reading process and how editors would come to their palaeographical conclusions. Let’s spell this out with the example of 1QSa II 11–12.
Complicating the Legacy System of DJD
In the previous post, I raised the question of what you would expect to find based on the DJD transcription of 1QSa II 11–12? Here is the transcription per the DJD editors:3
Obviously, the issue relates to the word,יו̇לי֯ד. The reading has significant implications on how we read and make sense of these two lines and perhaps even the entire document. What is more, the reading—as we should now expect—entails decisions about palaeography, morphology, syntax, and semantics, and therefore has a direct implications for larger social, political, and legal reconstructions of Second Temple Judaism. Before we take a look at an image of the artefact, let’s examine what other readings are possible:
Thankfully, Yaakov Licht4 provides a summary of the likely readings, and wisely opts to leave the word entirely ambiguous in his transcription:
Licht surveys each of the viable transcriptions:
Now, this is a significant amount of variation, in terms of palaeography, in terms of morphology, in terms of syntax, and in terms of semantics. Here is an image of these two lines. The image gives an entirely different impression of the material status of the lines, and the DJD 1 reading of יו̇לי֯ד.
So what would you read? Perhaps the question of what you would read first needs to be filtered through the question: How would you read this line? The answer of course is not as easy as the letter box, but it certainly relies on the letter box. We’ve reached the end of the legacy system of DJD, and we must innovate an entirely new method that will require dipping our feet into Digital Humanities.5
In the next post, I will examine how the legacy system of DJD can be significantly improved upon in a digital environment. This will involve introducing an idea that I presented in 2014, at the Material Philology in the Dead Sea Scrolls in Copenhagen and at the West Coast Qumran Study Group: Difficult Texts and Digital Tools in Portland, Oregon (lecture is available here). This idea involves tagging images themselves with Regions of Interest (ROI) and storing these regions along with their definitions in a database. Thus, the next post will take us to Part III – Regions of Interest Quantified – Solving Problems of the Analogue Methods of the Discoveries in the Judean Desert Series.
- Eugene Ulrich et al., “Introduction,” in Qumran Cave 4 VII Genesis to Numbers, DJD 12 (Oxford: Clarendon Press, 1994), 1–7.
- To my mind, this should have significant ramifications on how we teach languages such as Hebrew, Aramaic, Greek, and/or Latin, but unfortunately it does not. To learn more about the cognitive science at work here, see Stanislas Dehaene, Reading in the Brain: The New Science of How We Read (Penguin, 2009); see also “How the Brain Learns to Read – Prof. Stanislas Dehaene – YouTube,” Oct. 25, 2013, https://www.youtube.com/watch?v=25GI3-kiLdo.
- Barthélemy translates the lines as follows: “Voici (l’ordre de) session des hommes de renom invités aux convocations pour les dêlibérations communes, au cas où Dieu 12 mènerait le Messie avec eux: Que le prêtre ait rang de chef sur toute la Congrégation d’Israel …” DJD 1:117.
- Licht, Jacob. מגילות הסרכים ממגילות מדבר יהודה: סרך היחד, סרך העדה, סרך הברכות. Jerusalem: The Bialik Institute, 1964.
- This is a loaded term in many respects. I will engage this idea in the near future, but for now, I would like for you to consider this very important essay: How Not to Teach Digital Humanities