In this response I will pose a very basic question: what is an “Old Testament Pseudepigraphon”? In other words, I will address the building blocks of the volume Old Testament Pseudepigrapha: More Noncanonical Scriptures (From now on MOTP). My main interest in the context of this paper is to look for the ways in which textual units are represented in the volume and how these representations, first, relate to the manuscripts that serve as the empirical basis of the study of the Pseudepigrapha, and second, how they affect the way we – the readers – get acquainted with, imagine and conceptualize the textual units referred to as such.
Hence, the focus of my response is editorial models. To be more precise: I want to discuss the importance of scholarly models of texts, works and manuscripts in editorial practices and I will address the effect of these models to the editorial choices that eventually formed the MOTP-volume.
But first of all, before I say anything more, let me express my gratitude to the volume editors and to all the contributors for revealing all these hidden pearls. This volume contains English translations of and introductions to several texts that have hitherto not been generally known, or at least not easily available. A sign of the possible importance and relevance of such a contribution is that as soon as I started to read the volume in order to write this response, I also started using it in other parts of my research. So thank you.
“The Old Testament Pseudepigrapha, as the term is now commonly used, are ancient books”
In the first sentence of the General Introduction to MOTP, the volume editors Richard Bauckham and James R. Davila write:
In this sentence it is stated that “pseudepigrapha” are books, and as Bauckham and Davila correctly say: this is the way the term is commonly used. In ongoing research and in previous editions, for instance those of Charles, Sparks and Charlesworth, the Pseudepigrapha are talked about as books, alternatively as scriptures, writings and works. I could end my talk right here: the “Old Testament Pseudepigraphon” is a book.
Now, one of the main aims of MOTP, which is also one of the main reasons why this volume is important, is that the volume promises us more noncanonical scriptures. The last sentence of Bauckham and Davila’s introductory passage reads “This volume is a new collection of Old Testament Pseudepigrapha” (p. xvii). The volume will not give us the books that have already been published as such, but new ones.
A look at the TOC and the layout of the individual entries suggest that this is indeed so. The TOC shows that the individual units are provided with names, such as for instance, “The Book of Noah” and “The Book of the Covenant”, which identifies what will come. The standard layout of the individual entries further shows that these entries are divided into a general introduction of the textual unit, contents, manuscripts and versions, genre and structure, date and provenance, literary context, cultural and/or theological importance, a bibliography, and a translation. In other words, I expect more textual units that are books, more or less like the Pseudepigrapha I am already acquainted with through former publications.
However, when I start exploring the 39 individual entries in the volume, I find that the situation is much more complex. This complexity is due, firstly, to a chaotic empirical situation: for the majority of the entries the manuscripts that serve as empirical sources do not provide any direct access to these assumed ancient books. Secondly, the complexity is also produced by the editorial paradigm and model of work that decides the representations of those manuscript sources.
And before I go on, let me now explain how I use the term “work” and how this term relates to two other terms: “text” and “manuscript”.
I define “text” in the context of this paper as “a series of words in a particular order”, and more precisely as “the words on the page”. “Manuscript” I understand as “the text-bearing object”. This textual artefact is seen as a culturally produced material artefact, as the object that contains the text-on-page.
A “work” is a conceived compositional unit. It is a textual entity that is conceived of as a more or less consistent and coherent entity, having a more or less established identity, and which is circulating as an identifiable unit. Note, then, that the concept of a “work” is a representation and an abstraction – it is not to be confused with “the text” – it is a way we represent the text. Note also that this is but one way of identifying a textual unit – all textual units are not “consistent, coherent and identifiably one unit”. And please note that the conception of a work can both be existing in history and thus be part of what we study, and it can be a contemporary scholarly projection: something we bring to bear on the empirical material.
Further, scholars tend to understand the relationship between text and work in at least two different ways. In some studies a work appears to be the conceived sum of all the existing and assumed texts in circulation which are assumed to make up the larger, unifying whole. In yet other studies a work is conceived as the composition that the one or those who originally produced it intended it to be, or at least the assumed first and original version of the composition.
Although I do not rule out that both notions of “work” are present in the volume, as far as I can tell, the MOTP-volume is first and foremost based on that latter notion. Looking at the layout of the entries again, it is in this sense that the Pseudepigraphon in question is relevantly described in terms of date and provenance, genre and structure, literary context, manuscripts and versions, etc. The Pseudepigraphon is a work, and as pointed out in the very first sentence of MOTP, it tends to be ancient (or at least from late antiquity).
Works, “short units”, and “quotation fragments”
Why is this relevant? This is relevant because MOTP identifies the Pseudepigraphon as a textual entity on the work-level, whereas this work-identification may often and in various respects contradict the identifications and formats of textual units found in the manuscripts. This means that the contributors to the volume have had to navigate between a model of work on the one hand and a source material on the other. How have the individual contributors met this challenge? How have they represented the units they are dealing with?
When I go through the individual entries a complex picture emerges. I will divide the entries into three large categories based on how they represent texts-on-page in manuscripts. I admit that some of the entries are hard to place and others could belong to more than one category.
The first category of entries presents textual units which are relevantly and fruitfully represented as “works”. This category of entries deal with textual units that appear as more or less consistent, coherent and identifiable units in the manuscripts that have come down to us, and the MOTP entries reflect this. This concerns, for instance, the entry on The Testament of Job and The Seventh Vision of Daniel. The text-on-page in the manuscripts – and I certainly allow for some variance between manuscripts – seem to present texts appearing as one composition, which at least has circulated as a unit, sometimes identified in the manuscripts with a name, having a more or less recognizable structure and contents. However, this category of entries is not the most commonly found in the volume. If I push it, it makes up about 25% of the entries.
Second, another group of entries choose a different strategy which is equally relevant and fruitful, while not representing the units they are dealing with as works. These entries rather represent their material as “short units” or “self-contained units”. These units can be for instance textual clusters, sayings or excerpts collected in anthologies of various sorts – and they are represented as such in MOTP. In other words, the contributors in this second group of entries choose to follow the manuscripts, their formats and identifications more closely, but ending up discarding the work-model. Examples of entries explicitly choosing this strategy are Adam Octipartite/Septipartite and Sefer Zerubbabel (cf. p. 448). Again, if I push it, this category also accounts for about a fourth of the entries in MOTP.
The third category, which is the largest, is made up of entries that represent as works texts-on-page that are conceptualized in a rather different way in the manuscripts themselves. This category of entries include, first and foremost, entries that represent as works smaller units of text that the manuscripts locate and/or identify as integral parts of another work. In other words, this category of entries isolates units of text and re-contextualizes and represents them in terms of another work. This category of entries includes, for instance, The Book of Noah, The Book of the Covenant, The Apocryphon of Eber, and several others.
This attempt on a structured overview of the strategies chosen by the contributors in order to solve the contradictions between the work-model and text-on-page format shows that to the two last categories of entries it has been challenging to apply this model to the source material that we actually have. And to me as a critical reader, the third category is the problematic one.
An example: The Book of Noah
In order to illustrate some of the methodological challenges that results from this third approach, I will present and discuss one entry, The Book of Noah, by Martha Himmelfarb. And please note that I have chosen precisely this entry because it is both highly relevant to a discussion of models and paradigms in MOTP and a high quality entry with a robust argumentation. My interest here is to shed light on the effects of a paradigmatic editorial model.
The issue I will discuss in relation to The Book of Noah is: what is the outcome of representing an assumed reference or quotation to an assumed ancient book, found in context of a medieval work, surviving as text-on-page in even younger manuscripts, as an ancient work? Or to put it bluntly, the questions I am dealing with are: is the textual entity described here a work at all, and if it is; what work are we talking about, and to whom is this a work?
The first sentence of Himmelfarb’s entry reads: “The text translated here is the introduction to a Hebrew medical work from ninth- or tenth-century Byzantine Italy, the Book of Asaph” (p. 40). In other words, the unit that is referred to by the heading “The Book of Noah” in the MOTP-volume is a passage – the introductory passage – found in another unit, the Book of Asaph, which assumedly date from the ninth or tenth century. The translation of the passage from the Book of Asaph in the volume is based on Adolph Jellinek’s transcription of the version of the Book of Asaph found in the manuscript Codex Munich 231. This codex is assumedly the oldest of a handful of medieval manuscripts containing the Book of Asaph.
It is getting complicated. Our actual sources are medieval manuscripts. The textual unit in these sources is identified as a work, the Book of Asaph. The text we are looking at is the introductory passage of that unit in the medieval manuscript (transcribed by Jellinek), but the entry in MOTP is on either of them: it is on the work “The Book of Noah”.
According to Himmelfarb, the conception of a “The Book of Noah” goes back to Jellinek. He published the passage, together with two other passages, under that title in 1855. Himmelfarb notes that no such work has in fact survived, and that many scholars doubt that it ever existed, but also that other scholars attempt to reconstruct the content of such a book. Himmelfarb holds that it is possible that there once existed such a work. However, she also points out that the work from which the passage in the Book of Asaph was drawn was not necessarily restricted to Noah material – this assumed book does not have to be a “Book of Noah” as such.
The main discussion of Himmelfarb’s introduction is the relationship between Jubilees and the assumed Book of Noah, the possible use of a common source, as well as the possible influence of one, both, or either of the two on the Book of Asaph and other medieval texts. Based on specific features of the language as well as the similarities and differences in content between Jubilees and The Book of Noah, Himmelfarb suggests that both works might have drawn on a common Hebrew second century bce source. In other words, although Himmelfarb does not date The Book of Noah, this entry primarily puts the discussion of the work into a discourse on Second Temple Judaism and the later circulation of its texts.
The hypothesis that “The Book of Noah” is a Second Temple Jewish work is one possible hypothesis. It is certainly likely that the narrative clusters relating to Noah have long roots, and it is also possible that there was an ancient conception of a work ascribed to Noah. This last assumption is suggested in the text itself (v.11), and other sources may point in that direction (Jub 21:10; maybe 10:3).
However, as Himmelfarb’s introduction also makes clear, we still know too little about how assumedly older textual units and storylines were transmitted and transformed in medieval milieus. Himmelfarb’s contribution is transparent and informative on this issue, but it also displays a certain struggle with the format of the volume, and the procedure of moving from medieval text snippets to ancient works.
My main interest here is methodological: when this passage from The Book of Asaph is represented in an entry about the work The Book of Noah, that is, when the basic unit is “a work”, this is methodologically challenging and we should ask ourselves about the gain and the relevance of doing it. There might have been a work known roughly as “The Book of Noah”, but can we really get to it, and why is it an aim to get to it?
In the manuscripts that we have, the text labeled as such in the MOTP-volume is part of another work. That location of the text-on-page is probably decided by the author/redactor, or the scribe/copyist etc. of The Book of Asaph in the Middle Ages. If they refer to an assumed work ascribed to Noah, it may well be that this is primarily a reference to their – medieval – assumption of the existence of such a work. That assumption is not enough to conclude that there was such a thing as a second temple conception of a work.
And even if there was a second temple conception of a work, as Jubilees may indicate, we don’t really know what the relationship between a conception of work and textual contents was. There are several examples from late antiquity and the middle ages of mistaken identifications (e.g., Books of Baruch), of texts circulating under several names or no fixed names (e.g., Enoch-literature, the Palaea Historica, etc.), and sometimes there is a continuum between references to corpus – work – and stories about the acts of a figure when a textual unit is mentioned and named in historical sources themselves, which may suggest that to some extent the precise identification of one work vis-à-vis another was not the most important issue after all.
And importantly, even though the passage in question here might once have been excerpted from a book of Noah, it is also likely that is circulated, for instance, as part of a collection of excerpts or as parts of other works, just like we have it in context of The Book of Asaph, maybe even alongside an assumed “Book of Noah”. Hence, the work-representation is only one of several historical possibilities.
We need to ask ourselves: to what extent is it still “An excerpt of the Book of Noah” when we find it in The Book of Asaph, and to what extent is it something other? Is it relevant to talk about a given passage as “a passage excerpted from” another work, or are we better off seeing that passage as integral to the work the manuscripts seem to suggest that it belongs to?
And how about historical work-identifications: are they changing as well, or do we assume that ancient works-identifications appear as fire- and waterproof entities – as conceptions that move through history without friction?
And finally, it is possible that the use of the term “book” and “work” in the entry is only a conventional label. But, conventional labels may in fact paint a picture of textual units that were not necessarily there.
MOTP and the work-model
The result of this brief discussion of the editorial choices in MOTP suggests that the basic textual unit of the volume is a work. At the same time, two of the groups of entries that I established above, show that this template can be problematic.
The category of entries that represent their textual entities as “short-“ or “self-contained units” (my second category, above) dropped the notion of work and let the manuscripts decide their representation of the units. The third category of entries (above) chose the opposite solution. They present the unit as a work, to a certain degree despite the manuscripts.
These two categories of entries have one important issue in common: they are both dealing with smaller units of texts. They are not dealing with the kind of compositions that are well known to us from Charles’, Sparks’ and Charlesworth’s editions of Pseudepigrapha, where the Pseudepigrapha to a larger extent (but not always) appear as compositions which could fruitfully be approached as works.
The entities that are left for MOTP are to a large degree precisely shorter units that may or may not attest to the existence of ancient works. If we look at the manuscripts that contain them, they may also appear to be excerpts, independently circulating passages, “words” and lessons, story clusters, free floaters, snippets, etc.; sometimes existing independently, sometimes as self-contained units alongside other such units (collections of excerpts, compendia etc.), sometimes alongside complete texts which were conceived of as works (liturgical lessons), sometimes part of more “works” than one, etc. And let me put this straight: this is common. This is how texts, works, passages, and snippets tend to circulate in late antique and medieval Christian milieus. This is what circulation looks like if you look at the manuscripts.
Whether we want to represent these bits and pieces in the ways in which the manuscripts have them, or in accordance with a model that assumes that they stem from or otherwise identify ancient works is an editorial choice. This is a choice the volume editors reflect on in the General Introduction, saying: “The independent existence of the works apparently behind some of the quotation fragments is speculative, but we judged the case for these at least to be worth presenting” (p. xxvi, n. 33). I agree, the units here referred to as “quotation fragments” are certainly worth presenting and I am very glad that they are part of the volume. But it is quite another question if they had to be represented as works!
And to me, this is an important choice.
First, the choice will contribute to our historical knowledge about antiquity and our notion of how texts and traditions circulated in the middle ages. As pointed out above, we know that texts circulated in many more formats than the “work”-format. This kind of variance is to be expected, and why should so-called Pseudepigrapha be any different? We need to ask ourselves about the extent to which it is fruitful for readers of MOTP to get acquainted with this material in the work-format, or whether this format hides too much information, or even creates a map of second temple Judaism and late antiquity that may not fit the territory.
Second, there is a methodological issue involved here: how do we value the information contained in the source material when “the ancient books” are what we aim for? When it comes down to it, our sources are materially existing manuscripts. And one of the things that I really like about the MOTP-volume is the fact that both the volume editors and the individual contributors are really knowledgeable about the manuscript situation and that they share this knowledge in a transparent way – much more than I commonly see in publications like this one. This should be applauded. At the same time I think it can be fruitful to repeat, again, that our manuscripts are cultural artefacts, produced at a particular time and place, and that the producers of manuscripts also took part in a continuous shaping of the texts and textual identification contained in them. This can readily be seen in terms of manuscript variance – a variance that is commented upon by many of the contributors.
The question is if this historical variance matter. Does the identification and context of use of the text units by those who engaged with these works in later periods than the assumed original matter to our analysis? The paradigm of historical criticism generally suggests that it does not. That paradigm, which still is dominant in our fields of research and which tends to decide editorial models, fixes our eyes on the earliest phase of an assumed work as a phase were the work is finished and starts circulating as such, and from this point of view it would make sense to assume that later changes to texts in circulation do not really change that first identification of a textual unit.
I think we need to keep in mind that these assumptions are produced by a historical-critical paradigm of texts and works. It is our task as scholars to decide when this paradigm is no longer fruitful in order to make sense out of the material that has actually come down to us. And we should keep in mind something that the volume editors point out for us: the large bulk of empirical material, the manuscripts containing Pseudepigrapha, was produced after the ninth century. Hence, they are the material cultural products of precisely those who don’t matter to our analysis when the earliest phase is what we aim for.
Third, this is a question about paradigms and their capacity of creating worlds: Does the Pseudepigraphon have to be a book, identifiable as a “work”, in order to be something in its own right at all? Maybe we need that level of representation in order to isolate these snippets as anything other than the large and continuous flow of biblical interpretation?
Finally, I want to end this response where I started by expressing my gratitude to the editors and contributors to the volume for sharing their knowledge and making all these texts available to us. And luckily, there is more to come. I look forward to it!