When trying to grasp what Constantine the African achieved in just two decades of labor at Monte Cassino in the later eleventh century, we might conjure up an image of the library in which he worked. Although bombed by the Allies in World War II, the monastery of Monte Cassino was subsequently rebuilt. Thus, when we enter the doors of its library today, we can imagine something of the great house as Desiderius, its most famous abbot after Benedict himself, conceived it in the 1070s.

We, can, moreover, repopulate that lost library with a number of the Latin medical manuscripts that Constantine would have had before him: a collection of texts attributed to Hippocrates and Galen from the tenth century; a copy of Book III of Paul of Aegina’s encyclopedia of medicine, perhaps just recently translated from the Greek; maybe even a copy of the text on medicine compiled by the ancient Roman encyclopedist, Celsus. But of the Arabic manuscripts Constantine used, we have nothing left at all.¹

Most likely, he had only one copy of each of his Arabic texts. If they suffered any damage, there could have been no easy way to obtain a replacement. And such a loss does seem to have happened. Writing in the middle of the twelfth century in nearby Salerno, the medical author Mattheus Platearius reports that, when Constantine arrived at Cape Palinuro, on Italy’s “boot” just south of Salerno, “a seastorm broke out, and a great quantity of water came into the boat and destroyed a certain part of the Pantegni, namely the Practica.”² Mattheus was referring, of course, to Constantine’s copy of ‘Alī ibn al-‘Abbās al-Majūsī’s Arabic Kitāb kāmil aṣ-Ṣināʻa aṭ-Ṭibbiyya, “The Complete Book of the Medical Art.”³ This is just the scenario of loss that Constantine would list as a cause for melancholy in his translation of Ishaq ibn ‘Imran’s treatise on that topic.

This story is supported, moreover, by every eleventh- and twelfth-century copy of the Pantegni we have. Manuscripts from the first century of the text’s existence fall into one of two groups. Some have only the Theorica, the first ten books covering basic physiology, anatomy, and disease causation. It was this section of the text that Constantine dedicated to his patron, the abbot of Monte Cassino, Desiderius, who would go on to serve a brief tenure as Pope Victor III, 1086-87.

Some manuscripts of the Pantegni, in contrast, have both the ten books of the Theorica, and also one, two, or three books of the Practica: Book I on general regimen for maintaining health; the first part of Book II on the types of medicines; and the first third of Book IX, on surgery. It is as Platearius said: “But of the Practica of the Pantegni he translated only three books, for it had been destroyed by the water.”

Yet the “puzzle” of the Pantegni Practica goes deeper than Platearius’s story suggests. When we cast our eyes a bit further, looking into the thirteenth century, we find copies of a complete Pantegni: not simply with the ten books of the Theorica and the three books of the Practica (two of them incomplete), but with twenty complete books, covering all the topics that al-Majusi had included in his Arabic original of “The Complete Book of the Medical Art.”

How do we account for this stunning discrepancy? Where did this “complete” Practica come from? Mattheus Platearius gives us another interesting detail: Alfanus (the archbishop of Salerno who had patronized Constantine early in his career and was the dedicatee of Constantine’s translation on stomach disorders) “wished to reimburse” Constantine for his expenses in completing the Pantegni. Maybe Constantine found a new copy of al-Majusi’s original on Sicily, where the Normans were continuing to claim the island from Fatimid and Zirid control. Or perhaps he obtained it through contacts in North Africa.

But Alfanus died in 1085. What happened to the project of completing the Pantegni thereafter? The answer is, “We’re still figuring it out.”

The first layer of detective work occurred in the 1990s, in the context of an international conference focused on al-Majūsī and his Latin translator.⁴ Working on Books II and X, Mary Wack untangled the peculiar histories of both of the Practica’s books on pharmacy, the former on simple (uncompounded) drugs, the latter on compound (mixed) drugs. Something odd was clearly afoot. At the same time, I worked on Book VIII, trying to unravel the sources for the sections on women’s medicine. In 2006, Raphaela Veit then did a similar analysis of Book III, on fevers.

What we found, coming from different angles, was pretty much the same. An anonymous editor had “re-created” the Pantegni Practica. That is, he (she?), feeling that they knew what the full text should have looked like, “re-created” it, using bits and pieces of Constantine’s other writings—the Viaticum, Ishaq al-Isra’ili’s Book of Fevers, etc.—to assemble a text that looked like a “complete art.”

These revelations were stunning, but we still couldn’t explain the genesis of the full ten books of the Practica. Because we were using the Renaissance edition as our reference point, we didn’t realize that there were additional layers we weren’t yet seeing. It has only been in exploring the full corpus of twelfth century manuscripts that it has become apparent how complex the puzzle of the Pantegni, Practica actually is.

In 2015, looking through a mid-twelfth-century manuscript in Pembroke College Library, Oxford, I realized that a hitherto unidentified text, called simply a practica, was in fact a section of al-Majūsī’s original Practica. The passages in the Pembroke manuscript were from Books VI and VII. These, I would later be able to confirm, corresponded with an early thirteenth-century manuscript in Munich that has these books. (The Munich manuscript had been known for some time but only became available for consultation recently). I then discovered a third manuscript, now in Berlin, and a fourth manuscript, now in Uppsala, that had these “lost” pieces of the Practica, too. (You can read about the Uppsala discovery here or, if you prefer Swedish, here.) Additionally, one mid-twelfth century manuscript now in Toledo, discovered by Iolanda Ventura, has a unique copy of the Book X of the Practica, on compound medicines. This, too, seems stylistically to be the work of Constantine.

Interestingly, the Munich and Uppsala manuscripts seem to have traces of what would have been the equivalent of Post-it® notes: that an author or editor had left to themselves to signal where different parts of the text were supposed to go. For example, there seems to be a “note” indicating where Book III, on fevers, was supposed to be added. Later copyists had copied these as if they were part of the main text, never realizing that the work they had in front of them was incomplete. Together, these clues suggest that, as Mattheus Ferrarius had noted, Constantine himself appears to have been engaged on the task of completing the translation he had earlier had to leave unfinished.

Pulling these pieces of the puzzle together, we can see that the creation of the Pantegni Practica fell into at least three stages:

• Constantine’s original translation of Books I, the first part of II, and the first third of IX from the undamaged parts of his Arabic manuscript⁵
• Constantine’s translation of other sections of the Practica, the second half of Book II, all of Book VI, most of VII, and all of Book X (from a new copy or “mended” copy of the damaged original?)
• An unknown editor’s reconstruction of the remaining sections of the Practica from bits and pieces of Constantine’s other works

Exasperatingly, we still can’t pinpoint when and where this last, unknown editor was working. No extant copies of the “re-created” Pantegni Practica have been found that date before the second quarter of the thirteenth century, well over 100 years after Constantine died. Nor does any twelfth-century testimony—either a scholarly book or a library catalog—witness the existence of a “complete” Practica. For now, we will just have to live with this mystery.

The “re-created” Pantegni gave the text a new lease on life. Fifteen of the thirty-five copies of the Pantegni that survive from the thirteenth century (there are fifty-three from the “long twelfth century”) present the “completed” Practica. These are often imposing volumes, running sometimes up to 200 leaves or more. However, the thirteenth century was also the point that Constantine’s Pantegni encountered its first real rival: the even more imposing compendium of medicine, the Canon of Ibn Sina (Avicenna, d. 1037). This had been translated into Latin before 1187 by Gerard of Cremona, working in Toledo, and although it experienced an initial period of invisibility, it emerged in the second quarter of the thirteenth century to a position of prominence, soon becoming the bedrock of medical curricula in later medieval universities and the writings of learned physicians.⁷

The Pantegni, retrieved from the sodden manuscript with which Constantine arrived in Italy, never lost its power to attract readers. The Latin encyclopedia was still being copied as late as 1452, and it was printed twice in the sixteenth century: once with just the Theorica and once with the full twenty-book “re-creation.”⁸ Constantine’s editorial efforts, and the Pantegni’s success, are a testament to the intellectual power of the “complete art” that al-Majūsī had first assembled in Persia, in Arabic, in the tenth century.

Further Reading:

• Charles Burnett and Danielle Jacquart, “A Catalogue of Renaissance Editions and Manuscripts of the Pantegni,” in Constantine the African and ‘Alī ibn al-‘Abbās al-Maǧūsī: The ‘Pantegni’ and Related Texts, ed. Charles Burnett and Danielle Jacquart (Leiden: E. J. Brill, 1994), pp. 316-51. Our project has doubled the number of twelfth-century manuscripts reported by Burnett and Jacquart; their survey of later manuscripts, however, is, aside from a few new discoveries, largely complete.
• Monica H. Green, “The Re-Creation of Pantegni, Practica, Book VIII,” in Constantine the African and ‘Ali ibn al-‘Abbas al-Magusi: The ‘Pantegni’ and Related Texts, ed. Charles Burnett and Danielle Jacquart (Leiden: E. J. Brill, 1994), pp. 121-60.
• Raphaela Veit, “Al-Maǧūsī’s Kitāb al-Malakī and its Latin Translation Ascribed to Constantine the African: The Reconstruction of Pantegni, Practica, Liber III,” Arabic Sciences and Philosophy 16 (2006), 133-168.
• Mary F. Wack, “‘Alī ibn al-‘Abbās al-Maǧūsī and Constantine on Love, and the Evolution of the Practica Pantegni,” in Constantine the African and ‘Alī ibn al-’Abbās al-Maǧūsī: The ‘Pantegni’ and Related Texts, ed. Charles Burnett and Danielle Jacquart (Leiden: E. J. Brill, 1994), pp. 161-202.

1. An intriguing discovery was made of a tenth-century copy of Ishaq al-Isra’ili’s Arabic treatise on urines, now in the Vatican Library (BAV, MS Vat. ar. 310), that was produced in North Africa. Alas, there is nothing to connect it with Monte Cassino. Instead, it seems to have passed through Spain (and the hands of a Hebrew-literate user) before reaching the Vatican. Our thanks to Arianna D’Ottone for information on this manuscript.

2. The text of Platearius’ account, together with the biography of Constantine by his Cassinese contemporaries, Leo Marsicanus and Guido, can conveniently be found in translation in Francis Newton, “Arabic Medicine in Italy: Constantine the African,” in Mediterranean Passages, from Dido to Derrida, ed. Miriam Cooke, Erdağ Göknar, and Grant Parker (Chapel Hill: University of North Carolina Press, 2008), 115-121.

3. This work actually has two titles, both in Arabic and in Latin. al-Majūsī himself had called it both Kitāb kāmil aṣ-Ṣinā’a aṭ-Ṭibbiyya, “The Complete Book of the Medical Art” and Kitāb al-Malakī, “The Book of the King,” which Stephen of Antioch rendered as Liber regalis.

4. We should, in fact, say “translators” since a second translation of the Kitāb kāmil aṣ-Ṣinā’a aṭ-Ṭibbiyya of ʿAlī ibn al-ʿAbbās al-Majūsī was made around 1127 by Stephen of Antioch, a Pisan notary and philosopher working in the recently conquered Crusader state of Antioch. It is possible that the re-creator of the Pantegni Practica used Stephen’s version as a guide. But neither the chapter numbers or their titles correspond exactly, so any possible influence remains unproven.

5. The completion of the translation of Book IX, on surgery, took a different path and has been understood for some time. For a brief account, see Monica H. Green, “Crafting a (Written) Science of Surgery: The First European Surgical Texts,” REMEDIA, 13 Oct 2015, http://remedianetwork.net/2015/10/13/crafting-a-written-science-of-surgery-the-first-european-surgical-texts/.

6. Paris, Bibliothèque Nationale de France, MS lat. 6886 is also important in being one of three manuscripts that preserve an odd “fourth prologue” to the Pantegni, which connects the text, not with al-Majūsī, but with his contemporary, al-Razi.

7. Joël Chandelier, Avicenne et la médecine en Italie. Le Canon dans les universités (1200-1350) (Paris: Honoré Champion, 2017).

8. Constantine’s Pantegni has received no modern edition. It was printed twice in the Renaissance, and both editions are now readily available in electronic formats: [A. Turinus, ed.] Omnia opera Ysaac, 2 vols. (Lyons: Bert. Trot, in offic. Joh. de Platea, 1515), available at the Bibliothèque Interuniversitaire de Santé, http://www.biusante.parisdescartes.fr/histoire/medica/resultats/index.php?cote=00122&do=livre. This edition has the complete, 20-book version of the Pantegni. Another digital version can be found in the Wolfenbüttel Digital Library, URL: http://diglib.hab.de/drucke/ma-4f-35/start.htm?image=00463. The second edition of the Pantegni presents only the Theorica: Svmmi in omni philosophia viri Constantini Africani medici Operum reliqua … (Basel: H. Petrus, 1539).