What does a sacred text look like? Are religious books materially different from other books? Does materiality matter? This article deals with three different aspects of material variance attested amongst the Dead Sea Scrolls, Ancient Jewish religious text fragments, of which were found in the Judean Desert. I suggest that the substitution of the ancient Hebrew script by the everyday Aramaic script, also for Torah and other religious texts, was intentional and programmatic: it enabled the broader diffusion of scriptures in Hellenistic and Roman Judea. The preponderant use of parchment for religious texts rather than papyrus may be a marker of identity. The many small scrolls which contained only small parts of specific religious books (Genesis, Psalms) may have been produced as religious artefacts which express identity in the period when Judaism developed into a religion of the book.
What does a sacred text look like? Are religious books materially different from other books? Does materiality matter? Rabbinic traditions provide various rules for the materiality of sacred books, or, as some rabbinic texts state, of books that ‘impart uncleanness to the hands’. In rabbinic literature this phrase is sometimes used in the context of discussions about the sacredness or canonicity of religious writings. For example, we find in m. Yadaim 3:5 one of the references to discussions about the status of some writings, here the Song of Songs and Ecclesiastes:
All sacred scriptures impart uncleanness to hands. The Song of Songs and Qohelet impart uncleanness to hands.
R. Judah says, ‘The Song of Songs imparts uncleanness to hands, but as to Qohelet there is dispute’.
R. Yose says, ‘Qohelet does not impart uncleanness to hands, but as to Song of Songs there is dispute’.
Rabbi Simeon says, ‘Qohelet is among the lenient rulings of the House of Shammai and strict rulings of the House of Hillel’.
Said R. Simeon b. Azzai, ‘I have a tradition from the testimony of the seventy-two elders: on the day on which they seated R. Eleazar b. Azariah in the session, that the Song of Songs and Qohelet do impart uncleanness to hands’.
Said R. Aqiba, ‘Heaven forbid! No Israelite man ever disputed concerning Song of Songs that it imparts uncleanness to hands. For the entire age is not so worthy as the day on which the Song of Songs was given to Israel. For all the scriptures are holy, but the Song of Songs is holiest of all. And if they disputed, they disputed only concerning Qohelet’.
Said R. Yohanan b. Joshua the son of R. Aqiba’s father-in-law, according to the words of Ben Azzai, ‘Indeed did they dispute, and indeed did they come to a decision’. (trans. Neusner
However, exactly the same expression is also used for scrolls or even part of scrolls that contain such sacred writings. For example, the same tractate of the Mishnah discusses matters of language, script, writing material, and ink:
The Aramaic [passages] which are in Ezra and Daniel impart uncleanness to hands.
The Aramaic [passages contained in Scriptures] written in Hebrew, or a Hebrew [version] written in Aramaic or [passages written in archaic] Hebrew script do not impart uncleanness to hands.
[Holy Scriptures] impart uncleanness to hands only if written in Assyrian characters, on parchment, and with ink. (m. Yadaim 4:5; trans. Neusner
The Mishnah tractate Megillah does not use the expression ‘imparting uncleanness to the hands’, but discusses the laws about the reading of the scroll of Esther on the festival of Purim, as well as other related issues, such as in chapter 4 the reading of Torah on Sabbath and festivals. Here, too, we find stipulations about the materiality of scrolls that are religiously used. For example, one mishnah first discusses the mode of reading the scroll of Esther and then its material:
[
[
And if not, he has not fulfilled his obligation.
[
Later rabbinic literature gives many additional and much more specific writing instructions, but in this article I will not discuss those details.
The term ‘Dead Sea Scrolls’ generally refers to a few well-preserved scrolls and 15 000 (Tov
From a material point of view, these manuscripts are an invaluable source of information on text production in Judea in Hellenistic and Roman times. However, they are also an utterly frustrating source, both because of their fragmentary nature, and because of the uncertainty of their original material and cultural context. Are those scrolls the remains of the library of a sectarian community which was focused on communal study and worship? Or did they have multiple origins, only to be hidden or discarded in the caves where they were eventually found. Material study of the scrolls has rarely been a goal in itself, or even as part of cultural history. Rather, it has been first of all a tool for philological study. Material aspects are of help in reconstructing or constructing manuscripts, the texts of which we eventually want to edit and study (cf. Tigchelaar
Material experts have occasionally studied some of the scrolls materials, but a systematic analysis of all the materials still has to start. This means that even now many of the most basic data which one would want to have are absent, or provided haphazardly and in the most general terms by non-expert editors. For example, only occasionally do editions describe the quality of papyrus, and if they do, then only in terms like coarse, average, or fine. Or, different kinds of preparation of leather scrolls (untanned; tanned; split) has been observed, but those have not yet been studied systematically. Hitherto, therefore, discussions of materiality are still largely focused on some visibly determinable features like kinds of material, size, script, scribal and paratextual features, and so on. This kind of material variance of manuscripts may assist the reconstruction of scrolls, but it can also help us to reflect on the function of texts, on textuality, and ultimately on broader cultural and cross-cultural questions. With the latter goal in mind, I will describe in three case studies a few basic material features, assess scholarly hypotheses, and try to relate some of these to a broader context.
Aramaic was used as the official language, and the Aramaic script as official script in the First Persian Empire. This Aramaic script retained its important function also after the fall of the Achaemenid Empire. The ancient Hebrew script, amply attested in inscriptions up to the early sixth century BCE, gradually gave way to the Aramaic script in Second Temple Judea, and was not used any more by Jews after the Bar Kokhba revolt (cf. Yardeni
The rise of the Aramaic script (or ‘square’ script) and the decline of the ancient Hebrew script (often referred to as paleo-Hebrew script), is also attested in the Dead Sea Scrolls, where only sixteen manuscripts from Qumran and one from Masada are written in this ancient Hebrew script, against more than one thousand manuscripts written in the Aramaic script. How can one explain the variance of script?
Above we saw that rabbinic regulation considers only scriptures written in the Aramaic (square) script (‘Assyrian’) as appropriate for sacred scriptures. The Talmud regards the difference of script as one of the markers of identity which distinguishes Jews from Samaritans. A famous passage (b. Sanhedrin 21b) states that originally Torah (the five books of Moses) was given in the Hebrew language and in the (ancient) Hebrew script, which it then contrasts to the Jewish custom (namely Hebrew language in Aramaic script) and the Samaritan custom (Torah in Aramaic language and in the ancient Hebrew script). This difference between Jews and Samaritans apparently applied to the time of the Talmud, but can one use such later rabbinic statements anachronistically to explain the variance among the Dead Sea Scrolls?
The Sanhedrin passage speaks about Torah, and in fact, eleven out of the sixteen Qumran manuscripts that are written in the ancient Hebrew script, contain part of Torah.
The rabbinic attribution of the ancient Hebrew script to the Samaritans can hardly explain the provenance of the Dead Sea Scrolls manuscripts written in that script. Apart from 4Q22, these scrolls manuscripts are text-critically not closer to the Samaritan textual tradition than to the Masoretic or Greek. And even though 4Q22 repeatedly shares longer text forms with the Samaritan Pentateuch, and hence has been called pre-Samaritan, it seems to have lacked the ideological Samaritan additions (Lange
So if these few manuscripts in ancient Hebrew script are not Samaritan, then why were they written in the ancient Hebrew script? The very typical layout of the writing (as e.g. in 4Q22 and 11Q1) suggests a special scribal tradition, rather than an individual scribe’s variance of script. If one were to associate that tradition with a known specific group, then one could consider the Sadducees would nicely fit (cf. Delamarter
The explanation of conservatism, however, also raises the question why the majority of the manuscripts from the Dead Sea Scrolls, as well as in later Judaism, came to be written in the square Hebrew or Aramaic script. The presence of the ancient Hebrew manuscripts indicates we are not simply dealing with a wholesale shift from an ancient script to a more prevalent one, but rather a gradual shift. One may also note that the ancient Hebrew script remained in use in other media, e.g. on the Judean coins. I would propose that the substitution of the ancient Hebrew script by the everyday Aramaic script, also for Torah and other religious texts, should then be interpreted as intentional, or even as programmatic: it enabled and perhaps also reflected the broader diffusion of scriptures in Hellenistic and Roman Judea.
Excluding writing on stone, brick, and clay, the preserved manuscripts from western Antiquity are predominantly papyrus, with a gradual increase of leather and parchment starting in the first century CE, which eventually outnumbers papyrus from ca. the fourth century CE. The Dead Sea Scrolls are an exception. The scrolls from Qumran show a variance between the use of parchment (skin material)
Above we saw that some rabbinic texts explicitly state that sacred books (literally: books ‘that impart uncleanness to the hands’) should be written in Aramaic characters
What are the reasons for choosing either parchment or papyrus if one copied a literary text? Very common is the assumption that papyrus is cheaper than parchment. This is based on the fact that most documentary texts are written on papyrus. One therefore concludes that literary texts on papyrus would have been personal copies of texts, by individuals who desired to have a low-cost copy. Indeed, undoubtedly coarse papyrus would have been cheaper than processed skin, but we do not know how high-quality papyrus would relate price-wise to skin or parchment. In a delivered paper, ‘Choosing between papyrus and skin’, at a Groningen conference, George Brooke pointed at the presence of traded quality papyrus and suggested that the ‘choice between papyrus and skin reflects an intersection between high culture and popular culture, between the regional and the local’. Unfortunately however, no-one has hitherto systematically analysed the quality of either parchment or papyrus Dead Sea Scrolls manuscripts, nor their provenance, nor, for that matter, other correlations such as text density in comparison to quality of material.
I propose that there would have been different motives for using papyrus. One motive would indeed have been a choice for cheap material. The often semi-cursive and cursive writing on papyrus also suggests a non-formal use. For example, 4Q217, a so-called Jubilees manuscript, was written with large letters, in a semi-cursive script, on coarse papyrus. The size of the letters precludes that the manuscript would have represented an entire Jubilees manuscript or even the first part of it. Note that Jubilees is a longer composition than any of the biblical books. I suggested, speculatively, that this papyrus manuscript contained a draft for the revision of the beginning of the work (Tigchelaar
However, apart from asking why only so few literary manuscripts were written on papyrus, we could also ask an entirely different question from a history of culture perspective: why are so many Dead Sea Scrolls (including virtually all sacred texts) written on parchment? After all, papyrus was – given the archaeological remains – throughout Antiquity much more common in the entire Mediterranean than parchment, even though there is a strong increase of parchment in Late Antiquity.
The variance between parchment and papyrus also relates to the interpretation of the status of works or texts among the Dead Sea Scrolls. With the exception of the anomalous Cave 6 manuscripts, no ‘biblical’ texts have been found written on papyrus. In contrast, many other compositions which scholars regard as authoritative for those who collected the manuscripts, are represented in both papyrus and parchment manuscripts (e.g., Jubilees, Tobit, the Rule of the Community, the Hodayot, Damascus Document, MMT).
Virtually all Dead Sea Scrolls have been found in a very fragmentary state, which makes it often impossible to assess whether we have the last remaining fragments of, for example, an entire book of Genesis, or of only a part of the book. By examining physical and layout features of fragments, such as the height of a column, the size and numbers of letters in a line or column, we can sometimes imagine or even calculate the size or content of the original scroll. For example, in general there is a correspondence between a short column, say of 10 lines or even less, and a relatively short scroll. This feature enables us to conclude that in many cases we are not dealing with scrolls that contained a complete literary composition, but only part of it. This goes for many Psalms scrolls, many of which only contained a selection of psalms. But it also holds for several Deuteronomy manuscripts which contain selections from the books of Exodus and Deuteronomy that are also attested in the phylacteries, such as the Decalogue of Deuteronomy 5, and parts of Deuteronomy 6, 8, 11 and 32, and parts of Exodus 12. The manuscripts might have had a liturgical function, for example excerpted texts commonly used at rituals (Tov
However, also in other cases size or other physical features suggest that scrolls only contained parts of a work. Longer compositions would often be copied in multiple scrolls. In the case of Jubilees, one would have had two parts, chs. 1–23 and chs. 24–50; the first could have been read as a discrete literary unit, and the manuscript remains suggest it was copied more often than the second part. A different case is formed by cases where only the beginning of a work was copied. This holds for at least two Cave 4 Genesis manuscripts that probably only contained the first chapters of Genesis (4Q4 and 4Q7) on creation and paradise (or perhaps also the flood account) (cf. Brooke
Scrolls scholarship and biblical scholarship have become accustomed to the idea of textual variance. The Dead Sea Scrolls (and other textual witnesses such as the Septuagint) are witnesses to different degrees of textual variance, ranging from individual textual variants to variant literary versions of a work. The cases mentioned above, however, present a different form of variance, namely with respect to the selection and presentation of parts of a work. This raises the question of the relation of a text on an artefact to that of the text of a work (as represented in other artefacts). We should ask whether the reader or user of an excerpted or extracted text, or of a first part of a composition, is required to be aware of the compositional cotext that is not written in the artefact. Or does the complete text of an excerpted or extracted manuscript have a separate meaning on its own?
With the Dead Sea Scrolls we have the largest collection of admittedly very fragmentary Jewish manuscripts from before the Genizah. The amount of the material is stunning, and raises many questions, many of which have not even been posed before. For example, what these manuscripts tell us about the development of the production of parchment, a field of research which I expect will boom in the near future.
Most of the questions asked in earlier scholarship were historical and textual. Now that we have access to all fragments other queries become possible, such as how all these different manuscripts were used, and why they were produced in their particular way. For example, why would one produce parchment manuscripts containing apparently only the first chapters of Genesis? Were those short copies for personal reading or reference, or for communal study, or were they commercial copies? Or is it possible that in a period which testifies to the increasing scripturalisation of Judaism, such scrolls served primarily as religious artefacts, as objects that were not necessarily read but served as tokens of one’s religious identity?
In the past the Dead Sea Scrolls collections were seen as the library of an elite sectarian community, created for study and worship, and not necessarily reflecting the use of texts in Jewish society at large. In my view, the large material variance rather invites us to connect broader the scrolls with Jewish society, and to ask about possible functions of specific scrolls outside a sectarian community context. Whilst much of the scholarship in the field remains focused on questions related to authority, canonisation and canon, a culturally and historically more important question is that of the use of physical scrolls as religious texts and artefacts.
The author declares that he has no financial or personal relationships which may have inappropriately influenced him in writing this article.
Older scholarship hypothesized a council at Jamnia about 65 C.E. at which one also determined the extent of the canon. In the last half century this hypothesis has been challenged and dismissed by most scholars (cf. Lewis
For a brief overview of pertinent sources, see Tov (
There are many surveys of the Dead Sea Scrolls (see Tigchelaar [
Lists of all the ‘manuscripts’ are found in Tov (
1Q3 (1QpaleoLev-Numa; 1QpaleoLevb?), 2Q5 (2QpaleoLev), 4Q11 (4QpaleoGen-Exodl); 4Q12 (4QpaleoGenm); 4Q22 (4QpaleoExodm); 4Q45 (4QpaleoDeutr); 4Q46 (4QpaleoDeuts); 6Q1 (6QpaleoGen); 6Q2 (6QpaleoLev); 11Q1 (11QpaleoLeva). In
4Q101 (4QpaleoJobc); 4Q123 (4Qpaleo parJosh); 4Q124 (4Qpaleo Unid. Text 1); 4Q125 (4Qpaleo Unid. Text 2); and 11Q22 (11Qpaleo Unid. Text). The fragment Vat 57241, first published by G. Lacerenza (
Perhaps one should distinguish between parchment, leather, and papyrus, but since I am not aware of any statistical data available on the ratio of parchment and leather, and since the latter seems to be the exception, I will refer to all animal skin as parchment.
Cf. Falk (
No papyrus remains from Caves 2, 3, and 5. Two minute papyrus fragments (11Q28) were found in a box (#988) with some small 11Q1 fragments. 1Q70+1Q70bis are thirty-one mostly minute fragments from perhaps as many as five different documents, most of which may have been non-documentary.
This is a common sense assumption, suggested by several participants of the KNAW colloquium. However, the finds of Cave 6 suggest the opposite. Much more remains from the papyrus manuscripts than from the parchment manuscripts.
A statistical approach should also take into account the dates of the manuscripts. One cannot simply adopt the palaeographic dates given in the DJD volumes, and my own preliminary survey shows remarkably few cases of papyrus manuscripts from the so-called Herodian period.
6Q4 (6QpapKgs; the large size of the letters makes it unlikely that the manuscript contained the entire books of Kings); 6Q7 (6QpapDan); 6Q30 (6QpapProv; though one cannot know whether the manuscript contained a book of Proverbs). The tentative identification of one papyrus fragment 6Q3 (6QpapDeut?) is paleographically problematic, while another papyrus fragment 6Q5 (6QpapPs?) may be identified as preserving letters from Ps 78:36–37, which in itself does not necessitate it came from a manuscript of the Psalter.
4Q69 (4QpapIsap) is a papyrus fragment with remnants of Isa 5:28–30 in four lines, and has been published as an Isaiah manuscript. Given the profile of Cave 4, it is most likely a remnant of an Isaiah pesher. Note that the Isaiah pesharim generally have long Isaiah quotations. See Hartog (
See the Trismegistos site (
A Samaritan provenance of the text, published in Talmon (
In the 2000s two papyrus fragments of the Book of Enoch, purportedly from Qumran, surfaced, and are now in the Schøyen collection (4612/6 and 4612/12). However, the authenticity of the fragments has not been established, and several features suggest a modern provenance.
In 4Q216, the hand is fairly fluent, but slightly less regular. The original initial spelling of the word ‘angel’ (
Cf. similar questions posed in Brooke, ‘Genesis 1–11’.