In this episode, Sarah Kernan speaks to the principal investigator of Curious Cures in Cambridge Libraries, Dr. James Freeman. Curious Cures is an impressive project to digitize, catalogue, and conserve over 180 medieval manuscripts that contain unedited medical recipes in the University of Cambridge Libraries. Dr. Freeman talks about manuscripts, recipes, digitization, the Curious Cures team, and some of the challenges and rewards of working on such a large projects. To stay up-to-date with the project’s progress, follow #CuriousCures on Twitter, Instagram, and Facebook and the University Library on Twitter.
Frédéric Chopin, Etude Op. 10, no. 5 in G flat major
Provided by Musopen.org through Creative Commons Public Domain Mark 1.0
Mentioned in the Show
Medieval Medical Recipes in the Cambridge Digital Library
Culinary Recipes of the Middle Ages (CoReMa)
This Podcast Will Kill You (Episode 108 Gout: Toetally fascinating)
Sarah Kernan 00:08
This is Around the Table, a new podcast from the Recipes Project. I’m your host, Sarah Kernan. Together, we will learn about exciting scholars, professionals, projects, resources, and collections focused on historical recipes. Today I’m speaking to Dr James Freeman, principal investigator of Curious Cures in Cambridge Libraries, an impressive project to digitize, catalogue, and conserve over one hundred and eighty medieval manuscripts that contain unedited medical recipes in the University of Cambridge Libraries. James, thank you so much for joining me today.
James Freeman 00:48
My pleasure. Thank you very much for the invitation.
Sarah Kernan 00:51
Well, let’s start off with some basics. Could you tell everyone about Curious Cures, what exactly does the project encompass, and what manuscripts are part of this project?
James Freeman 01:01
Well, you summarized it very well in your introduction. So, as you say there’s over one hundred and eighty, I think it’s one hundred and eighty-six in total, medieval manuscripts, so handwritten books. They are predominantly late medieval manuscripts…fourteenth, fifteenth century, some into the early sixteenth century, though there are a few earlier. I think the earliest is eleventh century that we’ll be covering. And the one thing that they have in common is that they contain unedited, or uneditable even, medical recipes. So, these are recipes that have never been published in print before, or for the most part haven’t been published in print. They’re mostly English in origin and particularly the later medieval ones. Many of them contain recipes written in Middle English, but there’s a big mix of Middle English, Latin, there’s some in Anglo-Norman French, and there’s some, a few, in Old English. But it’s a fairly miscellaneous kind of corpus because, of course, the one thing it has in common is the unedited nature of the medical recipes that these books contain. So there were also manuscripts from Italy, in particular, but also France, I think there’s some from Germany, too, that are being covered by the project. But most of them are English and that’s a reflection also of the way in which the collections at Cambridge University Library and the Colleges and Fitzwilliam Museum that are collaborating with us have developed and what our collection strengths are.
Sarah Kernan 02:42
What kind of conservation and cataloging and digitization is actually taking place with these?
James Freeman 02:49
So, the conservation is, for the most part, focused on the conservation for digitization. So, this is the sort of routine stabilization work, repairing tears, that sort of thing, and where items are particularly fragile, providing the photographers in the University Library’s Digital Content Unit with sort of help in setting up, supporting, and handling the manuscripts. There’s a certain number within the project that will be able to do more intensive treatments which might involve in some cases doing a bit of rebinding or kind of more time consuming work that will ensure that these materials are accessible, not only digitally, but also in in the flesh, so to speak, to researchers for decades to come. All of the manuscripts that are being covered will be re-boxed or boxed, the ones at the University Library. So that’s the conservation sort of side of things.
James Freeman 04:03
Digitization, so it’s a bit of a mixed picture. Because we’re collaborating with twelve Cambridge Colleges, the Fitzwilliam Museum, and doing some of our own manuscripts at the University Library, as well, the picture is quite, is a bit mixed. So, for instance, Trinity College, they’ve digitized all or almost all of the manuscripts that we’ll be covering there and they have their own set up. Some, but most of the colleges, don’t. Corpus is another one that has had all of its manuscripts already digitized, but most of them don’t, so they will be coming to the UL for conservation treatment and for digitization by my colleagues in the Digital Content Unit. That digitization is going to be cover to cover. So, all of the, you know, the bindings, spines, four edges, all of the end leaves, all of the pages, and I mean I couldn’t go into the technical specifications about photography. But if there are some photography enthusiasts among your listeners, they’re welcome to get in touch and I can put them in touch with Raffa Losito, who is the project photographer, and she could give them all the details about the lenses, and the photography cradle, the rig that they’ve got set up, because I know some people get really interested in this. So yes, so that’s the conservation and the digitization.
James Freeman 05:26
The cataloging, again, it’s kind of the usual sort of thing that you would expect with a digitization project. So, I’ve got a team of three catalogers, Clarck Drieshen, Sarah Gilbert, and Tuija Ainonen, who will be describing the manuscripts textual contents. So, what the texts are and the intellectual contents of the manuscripts, their material characteristics, so what they’re made of, the collation, the structure of the gatherings of leaves, the dimensions, the layout of the text, what the decoration is, style of handwriting, etc., etc., etc. And also the history of those manuscripts as objects, so what their date and place of their origin, to the degree that we’re able to know this either from explicit evidence like scribal colophons, which are quite rare in English manuscripts, or more generally paleographical or art historical evidence, and also their provenance. You know, their ownership. And then the point at which they’ve come to the UL, you know, and there’s a number of common ways in which they’ve been acquired or donated to us over the centuries.
Sarah Kernan 06:43
How have you determined, or how has your team determined, which manuscripts would actually be conserved and digitized in these collections, as well as what kind of transcription or contextualization work was actually necessary?
James Freeman 06:58
So, it’s a tricky one for conservation because, you know, we don’t have the capacity outside of a project, outside of an externally-funded project, to do the sorts of detailed condition survey work that is the step one, if you will, of the Curious Cures project. So we have to make a kind of guess and an informed estimate of what is likely to be required on the basis of our experience with previous digitization projects. So, we’ve just come to the end of a major project to digitize, catalogue, and conserve all of the manuscripts in Cambridge Libraries that contain texts in Greek, which was funded by the Polonsky Foundation. So, we’ve got a bit of experience there, so we can. My colleagues in Conservation Shaun Thompson, who’s the lead conservator, and his colleagues Marina Pelissari and Rachel Sawicki, can come up with a sort of per manuscript estimate of what materials they’re likely to need, you know, materials to make a box and you know, wheat starch paste, Japanese tissue paper, that kind of thing, for doing the repairs. But then can also make an estimate of what might be required for more intensive work, both materials and their time, for a subset of maybe a dozen, or a few more, that will require a bit more, a bit more TLC, let’s say. Though obviously they’re much better informed than I am, and they could again provide the kind of detail that me as a non-expert can’t. So, if there’s any, again, if there’s any of your listeners who are interested in what they’re doing, there will be blog posts, I can promise you that, but we could also give you further information if you’re interested.
Sarah Kernan 08:47
Well, you’ve mentioned a couple other individuals who are part of the team who are helping out with this project. How large is the team, is it is it all staff, are there faculty or students or members of the public, or who all is involved in this project?
James Freeman 09:04
So, it’s a combination of existing members of staff whose time the project has essentially bought out. So, part of my time has been bought out while I run the project so we’ve recruited somebody as backfill for my role. Although coincidentally, he is actually also working as a cataloger the rest of the time. So, for instance, the cataloging team, they’ve all been hired from outside and they’re on fixed-term contracts for the project. But members of Conservation and Digital Content Unit, for instance, they already worked at the UL, so the Digital Content Unit, for instance, has to be sort of self-financing. So, the wages of the photographers that are paid, that work there, are paid by digitization projects, commercial orders, that kind of thing, so it’s a little bit of a mix. So, I mean the team is pretty large. I think there’s, and a rough guess I think, there’s about a dozen or maybe fifteen people involved in one way or another. It’s larger if we also count all of the librarians in the college libraries who are liaising with my colleagues in Conservation about the transfer of manuscripts from their library to ours and so on. So, it’s a fairly big enterprise, you could say, so we have regular team meetings and there’s a steering group that meets quarterly just to kind of keep a handle on progress.
James Freeman 10:39
There aren’t members of the public involved at this stage in what we’re doing, simply because the focus of the project. So, we’re funded by Wellcome by one of their research resources awards, and the focus, or what Wellcome’s kind of priority, are in their formulation health researchers in the humanities and social sciences. So, you know historians of medicine, historians of science, that kind of thing. So this is a sort of scholarly endeavor aimed at providing those researchers with better access to the collections that we have at the UL and other libraries in Cambridge. That said, there’s clearly great scope for public engagement and the response to the project so far to the initial publicity and the announcement of the launch has been kind of a bit mind-blowing, to be honest. Like I knew people would be interested, but I didn’t, I wasn’t quite prepared for how interested and what level of attention the project has received so far. I mean, it’s really encouraging and a lot of that, you know, that’s testament to the kind of skill of my colleague Stuart Roberts and his team in the Library’s communications office in getting the press release prepared and sent to the right places. So myself and other members of the team have been trying to keep the momentum going from that by doing podcast recordings like this. But also we’ve got blog posts that are forthcoming, and it’s in the very, very early stages, but after the project there is potentially the possibility that we might do a physical exhibition at the UL that can communicate some of the outputs of the project to a broader popular audience.
Sarah Kernan 12:37
Well, given all the excitement about the project, how long do people have to wait to see any of these digitized manuscripts? Will we start seeing any soon, or will we have to wait a couple years for anything to be posted?
James Freeman 12:51
No, no. For the sake of my sanity, we’re going to release them in batches rather than leave it all to the end. So, the first group of manuscripts, which I think might be sort of maybe around twenty or so, are going to, hopefully, all being well, unless there’s some major technical hiccup, will be appearing on the Cambridge Digital Library at the end of this month. It’s tricky, big digitization projects. There’s a lot of moving parts and eventually, obviously at the end everything comes together. But at the moment, we have a bunch of manuscripts that the catalogers have catalogued, a bunch of manuscripts that the photographers have photographed, but the two groups aren’t necessarily the same manuscripts, so where those two groups do overlap we’ll be putting those online as soon as we can. And so the plan is to do the first release at the end of this month and then to do quarterly ones thereafter; so end of March, end of June, end of September, and end of December.
Sarah Kernan 13:55
Digitizing these manuscripts obviously has great value for anyone studying medieval medicine, but what else is included in these manuscripts that other scholars might find useful? Are there culinary recipes, or literature, or poetry, or interesting household or scientific texts?
James Freeman 14:16
It’s a real mix, because the selection criteria, if you will, is that a manuscript contains an unedited medical recipe. We have got receptaria manuscripts, you know, compilations of dozens or even hundreds of medical recipes. We’ve got medical recipes that contain, medical manuscripts that contain recipes dotted in their, you know, texts or in their peripheries, but also non-medical manuscripts that have had recipes scribbled onto end leaves, or fly leaves, or margins, or blank space, or whatever. So, it’s a real mix. One thing I haven’t said so far is that in addition to the cataloging, digitization, and conservation, you know as if we’ve not got enough to do, I’ve also set us the challenge that we’ll be transcribing all of the recipes in full. I can say a bit more about that in due course. So, the idea is that what the cataloging will do is give researchers a sense or an insight into the intellectual contexts in which recipe medical recipe knowledge was being recorded and disseminated. And the digitization, kind of complementary to that, will show how that knowledge is being arranged and organized on the page.
James Freeman 14:51
We know where the recipes are because of the great work that contributors to the Index of Middle English Prose or Thorndike and Kibre’s index of incipits or Linda Voigts and Patricia Kurtz’s work on Middle English medical texts which are now in the kind of eTK and eVK online resource, the fantastic work that they’ve done. We know where this stuff is, but we don’t necessarily know what the, you know, the kind of granular detail. You know, where can we find recipes that concern toothache, or headache, or difficult pregnancies. Or where do we find instances of, I don’t know, herbs like rosemary, or sage, or rue being used. And in combination with what other sorts of ingredients. It’s much harder to get at the at the nitty gritty. So, the idea behind the transcriptions is that we can give researchers that sort of access. So yes, so the other contents can be very, very varied. They’ll be the kinds of medical texts that you would expect to see, texts by, I don’t know, Matthaeus Platearius and others, or Johannes de Sancto Paulo, but also nonmedical texts. You know, so we have bibles, books of hours, liturgical books. I think we have literary texts as well. Legal manuscripts.
James Freeman 17:08
I have to say I was when we were thinking, when I was thinking about the project and putting things together and considering the scope for the project, I thought, you know, can we really justify all the energy and effort and expense that digitizing a non-medical manuscript that might just contain a few recipes on its end leaves when we could focus on more medical, more medically-focused, or you know, medical-specific manuscripts. Now, my rationale was that, well, manuscripts that contain more stable medical texts are, if they haven’t already been edited, are more likely, or have a greater chance than tiny little recipe texts which are anonymous and very mutable and often have very small variations between them. They have the more stable medical texts have a much greater chance of being edited in a more traditional sort of scholarly edition whether it’s printed or online. And, I was looking at some of the manuscripts, as well, just thinking about the non-medical contexts in which you find these recipes, and there’s a legal manuscript at the at the UL that I was looking at as part of the project. I think it’s called Additional Manuscript 2994. And it contains copies of statutes, legal formularies, that kind of thing. Interesting if you’re into that kind of thing. And I looked at the recipes which are on the end leaves, and they are, you would expect, they’re going to be miscellaneous, they’ve been scribbled in by a number of different hands, they don’t have any kind of rationale. It was just, you know, convenience, or there’s the most easily to hand bit of spare parchment. What I found here, was that there’s a couple of dozen or more recipes, all for the same complaint, that have been deliberately gathered by one person. They’re all copied by the same hand or thereabouts. Most of them have been copied by one hand, there might be a couple of others that have been added on by different hands. So, there’s a sort of deliberate organizing, you know, intention behind this. I mean they’re all for gout, so insert obvious joke about medieval lawyers and their rich diets here. So yes, so the contents of the manuscripts is going to be very varied. Now that fulfills sort of our own institutional objective, which is to get as many of our manuscripts digitized and cataloged and available to people virtually as possible, but it also fulfills Wellcome’s requirements and the needs of these historians of medicine, health researchers in the humanities and social sciences, in being able to access this kind of stuff.
Sarah Kernan 20:07
Well, going back to when you were coming up with the idea for this project, what was your intention? What makes now the right time to do such a large project like this?
James Freeman 20:21
Well, we were, I mean it’s been long in gestation, I have to say. I think that our first kind of conversations about it, and I have to acknowledge the help of Peter Jones who’s a fellow at King’s College here in Cambridge and he was very helpful in these sort of early stages, thinking about medical manuscripts at the at the UL. It was not a part of the collection, I have to confess, I knew very well at that time. But I think we were, I think we were first talking about it in certainly pre-pandemic times. So I think maybe 2018. And we were at that stage just at the start of the Polonsky Greek Manuscripts project. But of course, you know how it is. You’re already thinking about what the next thing might be, and we knew that Wellcome offered this research resources award to enable libraries like us to make our collections more accessible to researchers. And we also knew that and we’ve got a kind of proven track record of cross-collection collaborative digitization projects. So, we thought well, how could we best leverage, if you will, what we’re doing with the Polonsky Project to the kind of next thing.
James Freeman 21:34
One of our collection strengths as well as medical manuscripts, are manuscripts containing text in Middle English, you know, and literary texts, you know, Canterbury Tales, works by, you know, William Langland, John Gower, and so on, are very frequently used for undergraduate and postgraduate teaching, as well as being sort of internationally significant for research. The medical ones are as well, but I have to say I hardly ever get those out for teaching. And I thought students would really take to this because, of course, you know, well, they would take to them for the reasons that they take to Chaucer, because it’s full of kind of bodily functions and you know very kind of visceral descriptions of human life and death. So, I thought these will have a great role to play in student teaching in the future. So, then I sort of started to look at what kind of resources existed and think, well, how could we improve on all the brilliant work that the IMEP or eTK and eVK have done.
James Freeman 22:43
And in September 2019, I went to a conference in Austria in Graz about TEI, the Text Encoding Initiative. So, this is about the kind of marking up of text with custom tags to essentially make it machine readable, categorizing what the content of the text is. And I met a researcher called Helmet Klug who is working on the Culinary Recipes of the Middle Ages project, CoReMA, which I’m sure your listeners will be familiar with, and his team were transcribing and marking up these cooking recipes using TEI tags, so marking up what words described ingredients, preparatory techniques, measurements of weight or quantity or time, equipment, that kind of thing, and I thought, well, this would map brilliantly onto medical recipes. I don’t feel embarrassed about saying that I thought, well, I’ll just nick this methodology, it’s brilliant! Why change it? You know, and in keeping some consistency with what CoReMA were doing, then of course you could bring the corpus of culinary and medical recipes together potentially, which would be really valuable for research so to see what kind of crossover there is in in these two texts. Now, first of all, you need the transcriptions, so we’re not going to be able to do what CoReMA are doing at this stage. We may actually have to leave that to health researchers in the humanities and social sciences to do that more involved kind of compact marking up and then comparative analysis. But, the development of handwritten text recognition technologies as well, means that we stand a good chance of being able to generate the full text, the quantity of full- text transcriptions, that previous initiatives, cataloging and indexing projects, haven’t been able to do.
James Freeman 24:51
So, we’ll be using Transkribus as a platform to do the, you know, line division and transcribe the text. We’ve got some developers here at the UL, Mary Chester-Kadwell is one of them, and she is going to help us kind of get up and running with Transkribus and improve on the TEI export that Transkribus generates. Because it’s a TEI description that will be produced, that the catalogers are producing, so we can bring that data together and put it onto the digital library. So yes, so that’s, those sorts of all of those sorts of things came together in terms of the timing of the project. We’ll have to see, there’s going to be a bit of, you know, trial and error and testing with Transkribus now, because of course. So, I went to the latest conference in Innsbruck last September and it’s clear that the machine, if you could call it that, requires much less manual training than it used to, say, five, six, seven years ago, in order to produce a model that comes up with an acceptable or a workable character error rate. And some people who were presenting their research were getting you know, really quite incredible resources. You know, character error rates of not just less than ten percent, less than five percent. So, and I was assured by people who are more familiar with this than I am, which was basically everybody else at that conference, that it’s much quicker to run things through Transkribus and then correct, than it is to do it manually. But, we’re not dealing with beautifully organized, neatly laid out texts with most medical recipe manuscripts. We’re dealing with scruffy layouts, multiple hands, writing in in different directions, you know, texts on the margins and the peripheries. So, we’re going to have to just sort of see how things go. Some manuscripts undoubtedly will respond better to the to the Transkribus treatment than others, but I think it might be quite interesting and push the boundaries a little to see what the capabilities of this software is with this somewhat un-standard, un-standardized style of manuscripts.
Sarah Kernan 27:28
The digitized images that come out of this project, those will obviously be able to be freely viewed by anyone. Will researchers be able to use those images freely?
James Freeman 27:40
Sure. So, we’ll be making them available as you say on the Digital Library using the Creative Commons license CC BY-NC 4.0. So, people can, they’re free to share them and adapt them so long as attribution is given and that the purposes are non-commercial in terms of publication. So, you know, you could use things on your blog or in teaching or in slideshows. Go for it. But for publications, they would have to apply for licenses to reproduce. Now, there might be room for negotiation about terms on that, I’d have to leave that to my colleague Domniki Papadimitriou in the DCU. She could advise, but I hope that seems kind of reasonable. The images will be available in one form or another, I mean, I would say in perpetuity to the degree that, you know, any of us know what’s going to happen in five, or ten, or twenty years’ time, but certainly the descriptions will be uploaded as a data set and individually to the university library’s repository, Apollo. So, it will be possible to access them both there and through the Digital Library. So, the idea is that there’s a kind of stable version. Each catalog entry has its own DOI and the collection as a whole has a DOI, so that there is a record of each of the catalogers’ work that they can point to in their future research and whatever. But also, then we can, we may as the years go by want to enhance and augment and revise and expand those descriptions in the Digital Library.
Sarah Kernan 29:19
Yes. Well, final question. Do you have any favorite recipes yet that you have come across in these manuscripts, or are there any really outstanding features that you are excited for researchers to discover?
James Freeman 29:35
Well, yeah. That’s a hard question to answer because there’s so much good stuff in there, and if people, you know, and your listeners, you know, work on this stuff they’ll be, they’ll know this already. I mean, I like one that’s a cure for headaches which basically involves grinding pepper and boiling it in white wine, and then, “as hot as thou might suffer hold thereof in thy mouth and when it is nigh cold spit it out and take new and so do till the ache be away.” So, basically deal with your headache by getting drunk again. I mean there are some really interesting recipes deal with sort of female health and particularly childbirth. And it’s interesting. You know that the head-to-toe organization of recipes, such as it is, struggles, I think, to know where to place these treatments. You know it does start off with the head and you know works through headaches, and toohaches, and sore eyes, and, you know, coughs, and, you know, aching arms, and then it sort of gets to the gets to the middle regions and then sort of loses its way. But there are charms in amongst these medical recipes, of course. You know, there’s one in Additional Manuscript 9308, which is at the UL, very common, it’s one of the so-called peperit charms and it’s for a woman that travailth of child. So it, and peperit being the, you know, Latin verb give birth to or bear a child, you know, invokes all of these female saints who, you know, gave birth. So, invokes, you know, Saint Mary peperit Christ, Sancta Anna peperit Mariam, Sancta Elizabet peperit Johannem, Sancta Cecilia peperit Remigium. So, it’s invoking all of these important female saints, and you’re supposed to write this on a strip of parchment and bind it around, you know, the arm, or the leg, or the belly of the of the woman who’s in a kind of having a bad labor. And in fact, I think the Wellcome have a surviving example of this.
James Freeman 31:54
So that’s the kind of charm side of it, but then you also see the medical side of it, as well. And there’s a recipe for a woman in a similar predicament and it instructs you to take the juice of vervain, so the kind of common plant verbena, and give this to her to drink in cold water and she shall soon be delivered with the grace of Jesus. So, I mean, I wouldn’t dare to presume how effective or not that might be. I suspect not very effective at all. But I don’t know what the juice of vervain tastes like, maybe it’s utterly disgusting and the idea is that it takes your mind off things. But yes, there’s so many to choose from. There’s, if anybody here listens to the podcast This Podcast Will Kill You, you know they had one on gout recently. And the ones for gout are absolutely brilliant. Ah, you know there’s a very elaborate recipe which I couldn’t resist mentioning in the publicity for the project, about roasting a puppy stuffed with snails and sage, or the one that involves taking an owl, gutting it, salting it, baking it in an earthenware pot and then grinding it up and mixing it with boar’s grease to make a salve that you then rub on the affected area. None of which will have worked. But yeah, the elaborateness of some of these recipes. And there’s a researcher here in Cambridge, Hannah Bower, who works on recipes and magic tricks, as well, and she’s sort of suggested, you know, what’s the kind of value of these, is there a kind of entertainment dimension to them, were some of them not actually intended to be taken seriously, which I think is really interesting. So, hopefully what the project is doing will provide her and other researchers in this area with the material for new discoveries and new insights.
Sarah Kernan 33:47
Yeah, I’ve no doubt that this project is going to yield a lot of incredible research with all the information that researchers will now have access to freely. James, thank you again so much for joining me today to talk about Curious Cures.
James Freeman 34:04
My pleasure. Thanks very much for the invitation, Sarah.
Sarah Kernan 34:07
Thanks to everyone for listening today. Please remember to subscribe to this podcast so you never miss an episode! I’ll see you again next time on Around the Table.