Getting to grips with early medieval medicine can be hard. If you were drawn to study the period by the complex characters and funny stories, dry, technical medical handbooks probably wasn’t what you were looking for. The field has not exactly been neglected, but it definitely falls into the category of one in which there is still much more to be discovered because the labourers have been few.

There are, for instance, plenty of early medieval manuscripts in which there are texts that haven’t been properly identified or studied. There is still a great deal of ongoing detective work by Cloudy Fischer and Aresenio Ferraces Rodríguez to make sense of the material and produce new editions. There is a new generation of scholars including Meg Leja and Claire Burridge too who have exciting things to say about the wider culture.

How tricky is it? I offer a simple example: the curious, fragmentary 8 pages of the manuscript Paris BnF nouv. acq. lat. 203.

In the grand scheme of things, this is actually one of the earliest surviving manuscripts of Latin medicine. It is written in what E.A. Lowe called ‘partly careful minuscule’ and ‘partly a careless and late uncial’, which he dated to the decades either side of AD800. It is possibly related to the centre that produced Glasgow Hunter 96, a more well-known early medical manuscript. Where it was produced is a matter of relative speculation. Lowe thought somewhere in southern Gaul or northern Italy with Visigothic influences. Bischoff narrowed the Glasgow manuscript to ‘Wahrscheinlich Narbonensis’.[1] The Paris fragments came to light after the pages were removed as binding in another book and presented to a judge in Le Puy in 1875 as a gift.

The challenge of the manuscript comes on three fronts: (1) it really is a mess (do have a look at the digitised microfilm at Gallica), (2) some of the contents fall into the overwhelming genre of short anonymous recipes, and (3) the titles that are given to the texts are misleading.

It is (3) in particular that intrigues me. It is a well known problem with early medical manuscripts that cataloguing efforts based on incipits (the title and/or opening sentence) can generate lots of problems because incipits are unreliable guides to content. We have seen this before on my blog of course with apocalypse and computus texts. Often, everything is at it seems, and everyone can relax. But the lesson is that researchers really have to check things for themselves.

Let’s look at the Paris manuscript in this context. On f. 3r, there is a text introduced with the words ‘Incipit de po/n\deribus Geronimis’, ‘Here begins the “On Measurements” of Jerome’. Most cataloguing efforts will therefore tell you that that is what it is. Except it is actually two chapters from Eucherius of Lyon’s (d.c.449) Instructions (II.13-14).

Following that text on f. 3v is a text introduced with the words ‘Incipit epistola Ypogratis de naturas humana’, ‘Here begins the letter of Hippocrates on human nature’. Hippocrates wrote a work by that name (edited here) and there is evidence of a late antique Latin translation of it. Our text, however, is not that. The work is generally referenced with caution, such as adding that the attribution is ‘uncertain’.

We can do better than noting caution. It is actually a variation of a text published by Valentin Rose as the Gynaecia of Vindicianus.

Vindicianus was a fourth-century physician who wrote a now-lost medical digest for the overrated Emperor Julian (d.363) and a letter to his nephew Pentadius offering a precis of Hippocratic theory on the humours. He was one of a number of late-antique North African medical writers who worked between Latin and Greek traditions and who, really, provided the foundations of much early Latin medicine.

The Gynaecia is itself a headache of a text. The title is perhaps misleading (and probably not original) as the work itself is a short introduction to anatomy that has a couple of long chapters on the conception and development of babies. Rose found the manuscript tradition of ‘the text’ so problematic that he eventually decided to publish the five versions he knew about in parallel columns in 1894. A few years later Karl Sudhoff had found two more version and in 1915 printed them in parallel too.[2] A couple more have been found since then but not to my knowledge this one (although I would not be surprised if someone had noticed it and I had missed it – sorry if that is you!).[3] The wording and contents continued to vary too much to produce a definitive text. Even the attribution might be shaky: sometimes the work was copied without attribution, only in one version (Rose’s II) was it attributed to Vindicianus, a couple of times it was attributed to Accius Iustus, plus there is another version which, like ours, was attributed to Hippocrates. Perhaps the multiple versions and mixed attributions should not be a surprise. Most versions start, as indeed ours does, by claiming to have been translated from a Greek source. North African Latin medical writers did this often because command of Greek among their audiences was poor. We may well be dealing with more than one translation or appropriation of the same source which might itself have existed in different versions.

Such matters probably made it harder to identify the alleged ‘letter of Hippocrates’ for what it was. Just compare the transcriptions of the opening words and that of its nearest relative (Rose’s III, from the early-ninth-century Dijon manuscript Paris BnF lat. 11218, where it is attributed to Accius Iustus).

The heading in NAL 203, f. 3v

NAL 203: Nunc et Gregorum liber certatu(s) sum et in Latino eloquio (t)hensauro ab conditit pertulit, a quibus conpaginibus fo(r)mam a materie contenemur in utero materno vel quibus ossibus aut quibus nervis coniu(n)cti sumus. Nunc a principio membroru(m) incipiat ad omnia…

Lat 11218: Nunc in hanc epistula exponere ex libris Grecis in Latino sum certatus, quibus ossibus vel quibus nervis vel quibus conpaginibus hominis conteneantur vel quomodo in uterum maternum coneneamur vel furmamur. Huius nos anathomi racionem exponere…

You can see that it is more or less the same, but it is pretty loose. The loose parallel continues throughout (to c. 12 at the bottom of 4r, c. 18 by the end of 4v, and then you have to flick back to folio 2 where you get the rest of the text). And that is compared to the closest version, at least as far as one can check from an attic by the Northeast Fife coast during a coronavirus lockdown.

In addition to the Eucherius and Vindicianus, there are some short anonymous texts that might in future illuminate medical knowledge at the centre that produced the manuscript. A short text on bloodletting that encourages one to keep one’s mind on Christ is a nice example of the gentle way that Christianity crept into early medieval Latin medicine. One cure for ergotism, meanwhile, is identical to one in the famous Lorsch Medical Book of c.800 (on f. 34r). In both cases, the material in common is accompanied by completely different other cures for ergotism. This should remind us again that our earliest surviving medical manuscripts are built up from material which had been under development over time in lots of different ways and in lots of different places, before we get to our manuscript ‘evidence horizon’ of c.800. It is not quite a case of ‘absence of evidence is not evidence of absence’, but rather that the evidence we do have implies that there must once have been more copies of things that we have lost.

One could put too much weight on eight scrappy pages. I like them because they might stand as a reminder that early medieval Latin medicine drew extensively on late antique North African developments. I like them because they might show some ways in which Possibly-Narbonne in c.800 had texts related to others in Dijon or in Lorsch. But, as ever, I probably like them most of all because they show that there is still work to be done.


[1] Bischoff, Katalog der festländischen Handscriften des neunten Jahrhunderts, 1, no. 1396.

[2] Sudhoff, ‘Zur Anatomie des Vindicianus’, Archiv für Geschichte der Medizin 8 (1915), 414-23.

[3] It is not, for instance, listed in L. Cilliers, ‘Vindicianus’s Gynaecia: Text and Translation of the Codex Monacensis (Clm 4622)’, Journal of Medieval Latin 15 (2005), 153-236 at 155-6… but knowledge does move on.