New Maps of Early Jewish Literature Across the Boundaries of the Real: A Response to Molly M. Zahn

I will begin by pointing out a real shift: Molly Zahn’s work is about second temple literature, and her title includes the phrase Revelation Before the Bible.^[1] The significance of the title might not have registered because it may seem obvious to many of us, but I want to flag it: in previous generations of scholarship, with notable exceptions, the corpus Zahn has been working on would have been categorized as postbiblical – after the Bible, not before. Something has changed to make Zahn’s title legible to us, or even invisible as a claim: we are ready to accept that the Bible is a category that was not available yet in the Second Temple period. But now what?

Zahn asks us to imagine “a new cartography, a new process for charting the relationships between points on a map – the paths that connect them and the borders that separate them” (Zahn 2021, 1). I will make two related points that both aim to make our maps even more capacious: first, about integrating not only extant fragments or surviving texts into our maps, but referenced, imagined, or remembered texts as well; and second, about a fifth variable, mapping texts’ locations and availability according the theories of ancient scholars themselves about where they were and if they could be accessed.

First, I want to set the stage for thinking about spatial metaphors, landscape, and new shapes of writing by sharing some art. These are sculptures by Guy Laramée, a Montreal artist who builds landscapes out of discarded books. He carves up encyclopedias and Bibles to show the process of knowledge transformation and loss over time, and to critique a culture that “insists on grabbing, on casting the flow of experience into concepts.” These books become new landscapes that aggressively reshape their previously obvious forms. I chose one of these sculptures as the cover image of my own 2016 book because it challenged me to remember that we do not need to be stuck in old containers anymore. We no longer have to think of “the Bible” as, in James Bowley’s spatial metaphor, the hub of a wheel, “with other texts revolving around it, some closer and some further away” (Bowley 2014). This is the model, for example, that shapes the Outside the Bible volumes, whose editors are up front about their own textual map:

we have arranged the selections in a way that highlights their closeness to the Hebrew Bible, starting with actual translations, then moving to various types of biblical commentary and rewritten biblical narratives, and proceeding from these to laws, liturgies, and rules for living drawn from the language and themes of the Hebrew Bible. (Feldman, Kugel, and Schiffman, n.d., 1:xvii)

The organizational principle is explicit: the sources are gathered in diminishing order of similarity and overlap with the Bible. But today, we are invited to carve all that up into new landscapes that reveal new clusters and relationships.

Breaking down entrenched categories is the first step. Zahn writes that we have a

tendency to emphasize one particular variable or feature of the texts, and use that variable to create the basic categories by which we organize the data. This is particularly unfortunate when the variable chosen is one that appears to have been irrelevant in the context in which the texts were actually produced and used, as if the case with biblical and non-biblical. (Zahn 2021, 2)

But there is another feature of texts that seems to be even more entrenched as a rigid category than biblical and non-biblical: whether or not they exist.

I think both texts that were actually available, and texts that were only referenced, remembered, or imagined, were indeed part of the literary landscape: they played important roles in how textual relationships and hierarchies were understood and constructed. In some individual cases, scholars have already acknowledged the significance of imagined texts to our understanding of extant ones. Perhaps most obviously, we know how a relationship with the imagined heavenly tablets shapes our understanding of the book of Jubilees. That text presents itself as the result of angelic dictation to Moses—dictation from a set of pre-existing heavenly tablets, though, as James VanderKam writes, Jubilees “may not exhaust all that is present in the written testimony on the heavenly tablets” (VanderKam 2010). It is these complete and authoritative tablets—not, as David Lambert reminds us, the now-canonical, extant books of the Pentateuch! (Lambert 2016)—that serve as the “unique fountainhead of tradition and authority” for Jubilees (Mroczek 2016, 47). Loren Stuckenbruck has argued that the concept of heavenly tablets served an important purpose for the continued production of Enochic texts as well: “Within the context of rapid growth of additional revelations in the patriarch’s name, the appeal to heavenly tablets reflected writers’ claims that their words were not in fact, ‘additional’ but rather constitute a provision of divine revelation that had existed all along” (Stuckenbruck 2010, 387-417, 398).

I have considered the link between real and imagined texts in another context: the many diverse psalms collections from Qumran and their relationship with the mythical 4,050 psalms of David, referenced in the most prominent scroll, 11QPsalms^a (Mroczek 2016, 38-44):

² And David, son of Jesse, was wise, and luminous like the light of the sun, and a scribe,

³ and discerning, and perfect in all his paths before God and men. And

⁴ YHWH gave him a discerning and enlightened spirit. And he wrote psalms:

⁵ three thousand six hundred; and songs to be sung before the altar over the perpetual

⁶ offering of every day, for all the days of the year: three hundred

⁷ and sixty-four; and for the Sabbath offerings: fifty-two songs; and for the offerings of the first days of

⁸ the months, and for all the days of the festivals, and for the Day of Atonement: thirty songs.

⁹ And all the songs which he spoke were four hundred and forty- six. And songs

¹⁰ to perform over the possessed: four. The total was four thousand and fifty.

¹¹ All these he spoke through prophecy which had been given to him from before the Most High.

The fantastical cache of 4,050 compositions is a symbolic number that not only evokes vast proliferation, but also specifically hearkens back to the 4,005 Solomonic compositions of 1 Kings 5, and is divided into calendrically significant sections. Inscribing itself into earlier literary traditions and reflecting cosmic order, this imagined and not entirely accessible corpus may be a conceptual matrix for the creation of actual psalms collections. Specific scribal products may give partial access to the totality of this sacred corpus, but are never their full embodiment.

Thus, the texts that are known or theorized about in antiquity, but not extant or not available (at least for now), do form part of the scriptural imagination of ancient people. They have conceptual relationships with the texts that do remain – and should be mapped as part of our new cartography.

Taking imagined texts seriously as salient elements of ancient literary awareness can help us see how some of the references to texts we think we can identify in ancient Jewish narratives might be no less imagined than the wholly fictional ones we have just discussed.^[2] References to Torah in second temple literature are a particularly good example. It is easy to identify, for example, whatever Ezra reads in public in Nehemiah 8 with what we know as the Torah. And yet, even if this narrative episode is based on a historical moment of public reading, the text involved would certainly have been different from one we recognize; even in the narrative context of Nehemiah, the “torah of Moses” is imagined to contain material not found in our Pentateuch. But whatever real text might lie behind the Nehemiah scene, Ezra’s Torah is accessible only as “a work realized in storytelling”^[3] about the return from exile. As such, this Torah has similar generative and evocative power to a wholly imagined text, a capacious placeholder that might be reimagined anew in every generation of readers. Another striking example from the Letter of Aristeas involves descriptions of physical manuscripts of the books of the law, manuscripts that range from poorly copied to impressive scrolls in Hebrew (§30, §176-77), to quality Greek translations ((§302, 308-13)—while the material documents in this fictional narrative are wholly imagined, it is tempting to think we at least know what their contents were supposed to be, and map them conveniently onto our Torah. But Aristeas provides few clues to their contents (notably, it never describes them in terms of five books; for an explicit reference to Torah as composed of five books we must wait for Philo, Eternity 19). The manuscripts play a crucial and vivid part in the drama of Aristeas, but their relationship to “real” texts that we can identify is as sketchy as the 4,050 psalms of David or 4,005 Solomonic proverbs in relation to extant biblical texts—a pliable, shapeshifting imaginary corpus that can accommodate new texts and changing conceptions of scripture. Texts in narratives, whether they are known to be fully fictional or whether it is possible to roughly identify them with a text we actually know, often function in similar imaginative ways.

Thus, the texts described and referenced in second temple literature are not always quite our texts, even if we might be tempted to identify them as such. A related point has to do with material accessibility: our texts might not have been readily available to all of our ancient writers, even if we can more or less securely date their composition. Historically, ancient libraries themselves were never comprehensive. Texts we actually have might have been just as inaccessible and distant to some as the 365^th book of Enoch.^[4] At Qumran, Esther might have seemed just as “imaginary” as the Hymnbooks of Job’s Daughters indexed in the Testament of Job.

While we know that our record of the ancient Jewish library is only partial, ancient people also had an awareness that their library was not complete. They also knew there had been more out there than they could collect and access now. Both “real” and “imagined” texts were part of that broader library that was not fully accessible—and telling real and imagined texts apart is not as straightforward as we might sometimes think. So, first, indexed and remembered texts should be part of our map.

Second, I want to add a fifth variable that can help us sort texts in new ways. In addition to creating flexible and temporary groups based on Zahn’s four variables – literary form, voicing, relationship to other known texts, and social location – what if we also mapped texts according to the way ancient writers imagined their location and availability? This would mean taking the cartography metaphor almost literally, and creating categories and relationships based on native cosmic maps: where the ancient writers thought these texts were.

Inspired by Robert Kraft’s language of how sacred knowledge becomes “earthbound” (Kraft 1996, 199-216, at 207), can we map exactly where, for early Jewish scholars and audiences, texts were located, where they’d come from and where they’d ended up, and if, by whom, and when they could be accessed? Are they in heaven or in a library, partially available in a concrete manuscript, or buried in a jar? Inscribed on a rock to be found by accident, sealed by students or angels for a future time, transmitted through patriarchal ancestors or ecstatic women, recollected from scattered fragments by Ezra, or accessible because they are embodied in a sage? So, I take the question of “mapping,” “landscape,” “cartography” quite literally, but as an emic category, and ask what a cosmic or theological map of writing might have looked like for some of the producers and consumers of our texts.

We could begin with two axes (Figure 1).

Figure 1

Then, we can plot the texts we have along with the texts our ancient writers imagined according to their location and availability, mapping conceptual connections between them.

The heavenly tablets, for example, are in heaven, themselves inaccessible, but partially reflected in Jubilees. The Torah is, of course, a complicated case, and some concept of Torah can be found in every quadrant (a little later, the rabbis will have explicit debates about this very issue). It can be cosmically pre-existent; available in the person of a teacher; temporarily lost or damaged; or re-revealed and re-constituted from scattered fragments.^[5] Psalms are also a larger corpus than any particular collection, but partially available in scribal products, as are Solomonic texts, many of which are not readily accessible (or, somewhat later, thought to be deliberately suppressed). Other texts might once have been available, but have been buried or hidden for a future time, like the books preserved in jars in the Testament of Moses. Some exists in another plane, like the sealed books to be opened at the end of days, or Enoch’s ongoing writing in Eden. One stage of this mapping experiment might look something like Figure 2.

Figure 2

We could add many more examples. But this first attempt should illustrate the general point: the “corpus” that I suggest we “map” includes lost or imaginary texts as well as extant fragments that we actually have and can study. Our maps should also identify clusters and plot paths between them, cutting across the boundary not only between “biblical” and “non-biblical,” but also between “real” and “imagined” texts.

This impressionistic first attempt allows us to visualize how some texts sprawl across various modalities. Perhaps those are the texts with the greatest generative vitality: Torah, for example, exists in the celestial world, but also in human representatives and material manuscripts that are sometimes accessible, and sometimes not; Enoch’s writings can also be found in at least three modalities: as real texts we can read (e.g. the 1 Enoch and 2 Enoch collections); as imagined corpora which are mostly unknown (e.g. the 365 books mentioned 2 Enoch 23:3); as otherworldly writing (Enoch’s continuous writing in Eden until Judgment Day, e.g. Jubilees 4:23). Solomonic and Davidic lore also appear across categories, creating room for the continued expansion or “filling in” of these traditions over time. Imagining inspired writing as spanning across various modes (extant and imagined, heavenly and earthly, accessible or hidden) helps us visualize the extant writings we study in a fuller interpretative context, and better track the conditions that made certain categories of writing most generative of new literary creativity.

Location and accessibility, then, are two related variables by which we can map the literary landscape and textual relationships. And these categories and distinctions, I think, can help us reorient the questions we ask about the history and nature of scripture. We have been in a bit of a rut when talking about the history of early Jewish literature as a largely process of canon formation, trying to pin down just how and when specific texts became scripture. We no longer want to think in terms of “biblical” and “non-biblical” for the Second Temple period, but we have replaced these with “authoritative” or “inspired” texts that we cannot help but assume mostly tracks with what is later “canonical.” But this relationship is not always linear. When we look at the way ancient writers themselves talk about their own texts, we can see that the emergence of a “canon” did not always preclude the existence of other inspired, authentic, revealed texts outside that list. Josephus insists that Jews do not have “myriads of discordant and competing volumes, but only twenty-two volumes containing the record of all time” (Ag. Ap. 1.38); and yet, extrapolating from 1 Kings 5, he also claims that Solomon wrote three thousand books of “parables and similitudes” and one thousand and five books of “odes and songs” (Ant. 8.44), and that there are multiple books of Daniel (Ant. 10.267). Categories of inspired, authentic, or authoritative literatures do not always help us understand all the ways ancient communities sorted their texts, or imagined how they had been sorted already in their ancient past. Later writers continued to imagine a larger body of texts—sacred writings lost, hidden, sealed up, or locked away in heaven—beside, beyond, and behind the extant and available biblical texts they possessed, as a persistent and durable context for “the Bible” even after its apparent fixing. As David Lambert shows in the next part of this series, and in his forthcoming What is Scripture?, readers do not encounter the Bible as a bounded and finished textual product even today. We could create a similar map across the boundaries of existence not only for the “pre-canonical” world of early Judaism, but for every era of biblical reception.

Can mapping the location and accessibility of real and imagined texts help us tell a new story about what a collection of scripture is? Are there more complex categories into which we can sort texts besides just the binary of authoritative and not authoritative, authentic or not, inspired or not? If so, how might this change the shape of our collection as we continue to study, teach, and publish on it? The metaphors of mapping, and Zahn’s invitation to consider multiple possible sorting criteria, might do no less than present us with an entirely new field.