Selecting and Organizing Recipes in Late Antique and Early Byzantine Compendia of Medicine and Alchemy

This month, we’re excited to collaborate with History of Knowledge to celebrate the upcoming conference, Learning by the Book: Manuals and Handbooks in the History of Knowledge. The five-day event takes place at Princeton in June and features a “blogged conference” to complement traditional panel presentations. For the next few Thursdays, the Recipes Project will cross-post selections from the conference (with RP readers noting  the extended length, in keeping with HoK posts). These features are just a taste of more than thirty works produced for the conference, and readers are invited to read the full selection here. Enjoy!


Matteo Martelli

Ancient recipes are usually short texts; one can easily find more than one recipe written on a single papyrus sheet or on the page of a Byzantine manuscript. Despite their brevity, however, they open an invaluable window onto a wide array of techniques and practices used to manipulate the natural world. Ancient recipes could pertain to various fields of science and technology — from cosmetics to cookery, from agriculture to horse care. In this post, particular attention will be devoted to two contiguous and, to a certain extent, overlapping areas of expertise: medicine and alchemy. As we will see, the works of two important authors, Oribasius and Zosimus of Panopolis, reveal the ways that recipe collections forged new forms of knowledge transfer in the fourth century CE.

In antiquity, medical recipes were easily exchanged among experts. Physicians used to send letters containing recipes to each other, as evident in Graeco-Roman papyri. Moreover, recipes were sold to people interested in specific formulas. And they could be quite pricey! In the second century CE, for example, a friend of famous physician Galen of Pergamum (second-early third century CE) was ready to spend over a hundred gold pieces to purchase highly valued recipes, some of which were preserved in “two folded parchment volumes.”[i] About a century earlier, the Roman physician Scribonius Largus (mid-first century CE) referred to the price of valuable formulas he had included in his Compositiones for a powerful drug against abdominal pains or an antidote made of hyena skin.[ii]

We can safely infer that recipes were collected in Antiquity. They were shifting atoms of knowledge that could be disseminated in a variety of treatises of different genres or simply piled into collections of variable length. The accumulation of technical knowledge could produce recipe books, usually in the form of lists or compilations of (often anonymous) recipes. Papyri offer strong, albeit fragmentary evidence for this process. A telling example is a fourth-century medical book usually referred to as The Michigan Medical Codex, which consists of thirteen leaves containing formulas for different plasters and salves.[iii] In a codex format, the papyrus has been identified as a manual copied for a practicing physician, who in some cases corrected the text or even expanded it by adding personal notes and recipes in the margins. In the alchemical field, two well-known examples of recipe books written in codex form are the so-called Leiden and Stockholm papyri (third-fourth century CE), which have been variously linked to workshop practices (Figure 1). They were defined either as handbooks for ancient craftsmen (e.g. goldsmiths, dyers) or as copies of the workshop notes of an artisan.[iv] The two papyri include more than two hundred recipes on how to dye metals, stones, and textiles (wool in most cases).[v]

Figure 1. Leaf from the Stockholm papyrus, freely available at the Word Digital Library:
Figure 1. Leaf from the Stockholm papyrus, freely available at the Word Digital Library:

These kinds of recipe books could be quite difficult to navigate, due to their lack of structure and fluid arrangement of the collected material. Readers often find no guidelines to assist them in the difficult task of locating specific procedures and techniques in a given collection. Moreover, these compilations often provide no information about the criteria for selecting and accumulating recipes. Important questions remain difficult to answer: to what extent does collected information correspond with the state and characteristics of a given discipline? How exhaustive is the selected material? To what extent were these collections used as reference works? Or were they local, produced by a single workshop or a scholar in contact with a small circle of artisans? What kinds of authority did the authors or compilers of ancient recipe books rely upon in selecting instructions to be included in their collections?

The three “manuals” or “handbooks” mentioned so far (the Michigan Medical Codex and the Leiden and Stockholm papyri) date to between the third and the fourth century CE, a moment of transition when “traditional” bodies of knowledge were inherited, selected, and re-organized. This cultural transfer and rearrangement of texts and practices had a strong effect on the ways that recipes were transmitted and organized. This is especially evident in the works of two almost contemporary authors: the so-called medical encyclopedia by Oribasius (fourth century CE), physician of the Roman emperor Julian the Apostate, and the alchemical books by the Graeco-Egyptian alchemist Zosimus of Panopolis (third-fourth century CE).

Figure 2. Bologna, Biblioteca Universitaria, MS 3632 (f. 97v), 14th-15th century CE The physicians Oribasius (left) and Philippos (right)
Figure 2. The physicians Oribasius (left) and Philippos (right). Bologna, Biblioteca Universitaria, MS 3632 (f. 97v), 14th-15th century CE.

On the one hand, these authors had to cope with an already rich and well-established tradition. Oribasius regularly exploited Galen’s huge medical corpus as well as the works of many other (less known) physicians. He extracted passages and quotations from earlier, authoritative writings and re-arranged them to build his own compendia. Even though less systematic, Zosimus’ approach to early authorities is equally dense. He constantly refers back to those figures of the first and second centuries CE who were identified as the founders of the alchemical art: Pseudo-Democritus, Maria the Jewess, and Pebichius, to name but a few.

On the other hand, Oribasius and Zosimus tried to provide as comprehensive a picture as possible of the disciplines they were committed to. In the introduction to his major compilation the Medical Collections, Oribasius spells out his aim “to seek through the most important writings of all the best authors and collect all that is of practical use to the very purpose of medicine.”[vi] Zosimus probably had a similar goal. According to the Byzantine lexicon Suda (Ζ 168 Adler; tenth century CE), he wrote an alchemical oeuvre in twenty-eight books. Regrettably, this work is no longer available in its original form, since only excerpts or kephalaia have been included in Byzantine manuscripts. However, one can get a glimpse of its structure by considering the twelve books preserved in Syriac translation, which I am currently editing and translating into English.[vii]

Both Oribasius and Zosimus shared a similar effort to systematize their fields. They were similarly committed to developing strategies in selecting and legitimizing the technical recipes they re-organized in their own works. A fresh comparison of their writings with the almost contemporary recipe books mentioned above can help to highlight these strategies. In fact, it is possible to track the movement of some recipes from the “manuals” on papyrus to the new, more exhaustive works of Oribasius and Zosimus.

Recipes were attributed to authoritative figures and organized in sections devoted to specific areas of expertise: the treatment of a single disease, for instance, or the description of a particular craft. Explanatory sections introduced the recipes, thus providing critical information for situating the copied procedures in a broader (either technical or theoretical) context. On the one hand, the combination of theoretical parts with bodies of recipes anticipates the structure of Latin alchemical handbooks in the Middle Ages.[viii] On the other hand, the tendency to be as exhaustive as possible could lead these authors to write vast treatises that were difficult to handle for a practicing physician or alchemist. Oribasius was certainly aware of this risk. He wrote a summary (Synopsis) of his Medical Collections for his son Eusthatius: “for when they (i.e. professional physicians) read what I have stated concisely and in outline, they will remember the whole of each field of knowledge, and without having to carry with them a heavy weight it will possible for them to be sufficiently equipped with what is needed in practice.”[ix] Meanwhile, Oribasius’ summary is presented as a kind of “portable” reference book. This perhaps suggests the meaning of modern terms “manual” or “handbook,” given that the Greek word encheiridion (usually translated as “manual, handbook”) never occurs in the texts considered here.

Exhaustiveness, acknowledgment of the authority of earlier authors, and clear organization of the material around key areas represented important goals in Oribasius and Zosimus’ works, which reorganized recipes that we find scattered in “manuals” on papyrus. They tried to secure medical and alchemical practices against the risk of being fragmented and dispersed in a variety of recipe books, thus producing crucial writings in the study and transmission of these disciplines.


[i] Galen, On Avoiding Distress (De indolentia), §§ 32-33, trans. Vivian Nutton in Peter N. Singer, Galen: Psychological Writings (Cambridge: Cambridge University Press, 2013), 87.

[ii] Recipes 122 and 172 in Scribonius Largus, Compositiones, ed. Sergio Schonocchia (Leipzig: Teubner, 1983).

[iii] The extant fragments of this codex have been edited by the American papyrologist Louise C. Youtie in a series of articles for ZPE (Zeitschrift für Papyrologie und Epigraphik). Later on, these editions were republished in a single volume by Ann Hanson in Lousie C. Yountie, P. Michigan XVII, The Michigan Medical Codex (P. Mich. 758 = P. Mich. Inv. 21), ed. Ann Hanson (Atlanta: Scholars Press, 1996).

[iv] See, for example, Mark Clarke, “The Earliest Technical Recipes. Assyrian Recipes, Greek Chemical Treatises and the Mappae Clavicula Text Family,” in Craft Treatises and Handbooks: The Dissemination of Technical Knowledge in the Middle Ages, ed. Ricardo Córdoba (Turnhout: Brepols, 2013), 9-32.

[v] Greek text and French translation in Robert Halleux, Papyrus de Leyden, papyrus de Stockholm, fragments de recettes (Paris: Les Belles Lettres, 1981). Both papyri were translated into English by Earle Radcliffe Caley: “The Leyden Papyrus X: An English Translation with Brief Notes,” Journal of Chemical Education 3.10 (October 1926): 1149-1166 and “The Stockholm Papyrus: An English Translation with Brief Notes,” Journal of Chemical Education 4.8 (August 1927): 979-1002. A reprint of both translations (edited by William B. Jensen) is available here.

[vi] Oribasius, Medical Collections, introduction (CMG VI.1,1, p. 4 Raeder). English translation in Philip van der Eijk, “Principles and Practices of Compilation and Abbreviation in the Medical ‘Encyclopaedias’ of Late Antiquity,” in Condensing Texts – Condensed Texts, eds. Marietta Horster and Christiane Reitz (Stuttgart: Franz Steiner Verlag, 2010), 526.

[vii] For a French translation of extensive sections of these Syriac books, see Marcelin Berthelot, Rubens Duval, La chimie au Moyen-Âge, Vol. 2: L’alchimie syriaque (Paris: Imprimerie nationale, 1893), 210-266.

[viii] These are so-called medieval pratica, a well-organized description of series of procedures opened by a general introduction and often complemented by a theoretical part (theorica). See Robert Halleux, Les textes alchimiques (Turnhout: Brepols, 1979), 80-81.

[ix] Oribasius, Synopsis, introduction (CMG VI.3, p. 5 Raeder). Translation in Eijk, “Principles and Practices of Compilation,” 529.

[7] van Laer, Weg-wyzer, 134.


Blog Series: Learning by the Book

Join the conversation on Twitter with the hashtag #lbtb18. Tweet or email links to related discussions. Read more posts in this series, and check out the conference website.

Leave a Reply

Your email address will not be published. Required fields are marked *