In 2016, I gave paper that revisited the old question of the rise of the codex. There has been a lot of work done over the years on why the codex may have come to replace the roll, but there has been much less concern about the question of how: What technological developments were involved in bringing about the multi-quire codex with its characteristic stitching that became the medium for the transmission of literature for a millennium? I concluded that when we look at the varieties of single- and multi-quire codices, tablets, and notebooks that have survived from antiquity, the real difference in terms of the techniques of connecting the elements (wooden boards on the one hand,  papyrus or parchment sheets on the other) was the jump from the single-quire codex and its usual binding with tackets to the multi-quire codex with its much more complex link-stitch binding. In response to the paper, AnneMarie Luijendijk posed a characteristically perceptive question that I was unable to answer: If the development of the link-stitch and the multi-quire codex was such a technological jump, where did the technology come from? Until a few weeks ago, I still was unaware of a satisfying solution.

That changed with my visit to Georgios Boudalis’s exhibition at the Bard Graduate Center, The Codex and Crafts in Late Antiquity. In the exquisite exhibit and the accompanying book (The Codex and Crafts in Late Antiquity, New York: Bard Graduate Center, 2018), Boudalis clearly and compellingly demonstrates the close connection between the sewing of textiles and the binding of multi-quire codices. As Georgios put it in the title of his paper at the symposium: “Can a Book be Compared to a Sock?” The answer is a resounding “Yes,” and this observation marks a giant leap forward in our understanding of the development of the codex.

Bronze statuette of an artisan, with object under belt; image source: The Metropolitan Museum of Art, Rogers Fund, 1972, 1972.11.1

I recognize that I can’t do justice to all the sharp observations in the book, but I’ll try to summarize its main contributions. After an introduction to the standard history of the codex, Boudalis turns to the “precursors” of the multi-quire codex, namely wooden tablets and single-quire codices. His treatment of tablets brims with fresh insights about how the physical remains of these wooden slabs might give clues to the ways they may have been bound together. These are accompanied by excellent drawings. His section on single-quire codices is similarly well done (although some statements about the earliest Christian books require modification; see below). In addition to making good use of surviving physical artifacts and literary references, Boudalis also offers novel and illuminating interpretations of pieces of iconographic evidence as well. I’ll just highlight one of them: Especially intriguing is his suggestion that a bronze statuette at the Met assigned to the first century BCE might be carrying in its waistband not a set of wooden tablets as is usually assumed, but rather a small single-quire codex of papyrus or parchment with a spine strip not unlike those that sit beneath the covers the Nag Hammadi codices. Boudalis notes that the item in the belt appears to bend under the pressure of the belt and thus must depict an object made of a material less rigid than wooden slabs.

Bronze statuette of an artisan, with object under belt, detail of object from front and side; image source: The Metropolitan Museum of Art, Rogers Fund, 1972, 1972.11.1

Judging from the description on the Met website, the statue appears to be without archaeological context, so its dating is open to some question, but Boudalis’s observation is nevertheless tantalizing as a potentially very early depiction of a codex. These chapters on tablets and single-quire codices build to the conclusion that the multi-quire codex in a way combines the technologies of these two formats:

We have found that the multigathering codex does combine features borrowed from both the wooden tablet codex and the single-gathering codex. Sewing through the fold is probably inherited from the single-gathering codex, while sewing multiple gatherings into a single structure appears to have been adapted from the wooden tablet codex (p. 153).

The means for this combination is the technology of the loop-stitch (Boudalis’s preferred term for what is more often called the link-stitch). This feature of the multi-quire codex is set within the field of contemporary technologies by comparisons between the sewing of bindings with that of textiles. Again, we are treated to excellent and innovative illustrations to explain the connection, such as the figure below:

A clever illustration from Boudalis’s book that helps to conceive of the binding of a book as a fabric

The following chapters vividly show how other parts of the codex—spine linings, endbands, fastenings, etc.—are equally closely connected to various different ancient crafts. Finally, Boudalis poses the question of what type of setting could serve to incubate such technological developments? He notes that one clear answer is monasteries, where both the production of textiles and the production of codices is well attested. The book concludes with a catalogue of the items in the exhibition, both the ancient items and the many helpful models. Having seen these items in person, I can say that it is a pity the images in this section are not larger, but in most cases they still illustrate the points that Boudalis wants to convey.

This is a brilliant book. Some of Boudalis’s ingenious proposals are more imaginative than others. The lack of well-preserved examples of early stab bound codices or the cords that held together wooden tablets prevents us from knowing for certain how these items were bound. Boudalis’s reconstructions are rightly and openly signalled as hypothetical. And they are entirely plausible, but they are not (yet!) capable of being confirmed from existing archaeological remains.

I spotted only a couple relatively minor factual errors in the book. Discussions of the Nag Hammadi codices and the Bodmer Papyri are occasionally a bit misleading. It is claimed that Nag Hammadi Codex I and Codex XIII were composed of two quires each (p. 37 and p. 42, note 12). It is indeed true that some early analyses concluded that these books had two quires (e.g. Doresse 1961, 28-29 ; Krause and Labib 1971, 14), but more recent and thorough investigations have come to different conclusions. In the case of Codex I, according to the best (indeed, the only possible) reconstruction, we are dealing with three quires of 22 sheets, 8 sheets, and 6 sheets (Robinson 1977, xvii-xviii). Codex XIII is more ambiguous. With only the remains of eight leaves surviving, it is hard to tell the original make-up of the complete codex, but among the surviving leaves, there is evidence for only a single quire (Turner 1990, 359-360). Elsewhere, the texts making up the Bodmer Composite Codex (LDAB 2565) are referred to as if they were each individual codices (p. 48, note 3). But such small slips are not central to the arguments Boudalis is making and take nothing away from the achievement of this book.

We students of the early codex are very much in debt to Boudalis for this truly groundbreaking and engaging work. It really is a game-changer in the study of the history of the book.

References:

Doresse, Jean. “Les reliures des manuscrits gnostiques coptes découverts a Khénoboskion.” Revue d’Égyptologie 13 (1961), 27-49.

Krause, Martin and Pahor Labib. Gnostische und hermetische Schriften aus Codex II und Codex VI. Glückstadt: J.J. Augustin, 1971.

Robinson, James M. The Facsimile Edition of the Nag Hammadi Codices: Codex I. Leiden: Brill, 1977.

Turner, John D. “Introduction to Codex XIII.” Pages 359-367 in Charles W. Hedrick (ed.), Nag Hammadi Codices XI, XII, XIII. Leiden: Brill, 1990.