Category Archives: Textual Communities

Creating and integrating a database – work in progress

By Marieke Hendriksen

As mentioned in the post introducing the ARTECHNE project at Utrecht University last month, we are in the process of creating a database containing recipes, artist handbooks, and art theoretical texts that can clarify the development of the use of the term ‘technique’, as well as related terms referring to processes of making and doing. The database is linked to Geographical Information Software (GIS), thus creating an online historical semantic map of ‘technique’. Such maps are more than merely nice illustrations; they can reveal connections that remain hidden otherwise.

We are in the fortunate position where we can build the database on an existing one, namely the excellent Colour ConText database. However, we do not simply want to offer a copy of the Colour ConText database, we want to integrate it with other sources on artisanal practices and theories, such as recipes, books of secrets, art theoretical texts, and artist handbooks, from the period 1500-1900, in Latin, Dutch, German, English, French, Italian and Spanish. The entire database has to be searchable with advance searches, allowing users to search full text for occurrences of terms, in particular geographical or linguistic areas and periods, or to trace the changing uses and meanings of particular terms over time, linking various forms of terms and different terms with a similar meaning in relational tables and glossaries.

Sneak preview of the new database website
Preview of the new database website

For example, if we want to know how instructions for making paint have changed in the low countries between 1600 and 1900, we want to be able to select all available sources from that period and area and search them for imperatives, nouns, measurements, and particular ingredients. We also want to be able to distinguish between printed and manuscript sources, artist handbooks and household recipes books, and ideally we want to be able to search annotations and marginalia too, as well as differences between various editions of the same work. That means a lot of different parameters have to be specified for each source, and everything we include must be carefully checked before it is added.

One of the problems we encountered when selecting the first sources we wanted to add was the low reliability of Optical Character Recognition software when used on early modern printed sources – a problem I have written about on my own blog before. We do not just want to combine existing digitized sources in our database, but also to add texts that have not been digitized and made searchable thus far. However, that often involves lengthy correction processes. One of the solutions we are considering is crowd sourcing such corrections. Fortunately, we can also add many digitized sources that are available under Creative Commons licenses (i.e. from the Digital Library of the Netherlands), and various researchers have contacted us to discuss adding their own datasets to the database.

The first edition of the database is now online – have a look at!  Please note that this is a first version, and that the database will be improved and continue to grow for the duration of the project. Over the next few months, we will focus on increasing searchability and adding new sources. If you have any questions or suggestions, please contact us at artechne[at]

Finally, we will also explore the options of maintaining the database after the end of the ARTECHNE project, for example by depositing the datasets or even the entire infrastructure with an archival institution.

The ERC ARTECHNE project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 648718) and is a cooperation of Utrecht University and University of Amsterdam.

Seiseinyū and secrets: problems of recipe attribution in early modern Japan

In my last post, I looked at the problems faced by early modern Japanese doctors trying to figure out how to manufacture a new mercurial drug called seiseinyū, which had first appeared in the Chinese doctor Chen Sicheng’s Secret Record of Syphilis (1636). Yet although Japanese doctors eventually found ways to produce seiseinyū, the complicated movement of books and ideas between China, Europe and Japan during this period meant that even doctors who knew how to make the drug could be confused about where their recipes had come from.

The syphilis doctor Tokujitsu Junchoku, hero of Funakoshi Kinkai’s “Illustrated Syphilis War Tales” (Ehon baisō gundan, 1838). Courtesy of Waseda University Library.

One of the most detailed manuscript recipes for seiseinyū was written down in the late eighteenth century by a doctor called Haruhi Gen’an. Gen’an claimed this recipe had been passed down in his family since the time of his ancestor Haruhi Genryō, who in turn had received it from a Dutch doctor called “Seirukettan” in Nagasaki in 1711. Gen’an believed that the Dutch themselves were reluctant to share this secret recipe, and he took great pride in his lineage’s knowledge of it; imagine his surprise, then, when he discovered that Chen Sicheng’s Secret Record of Syphilis, which had recently been reprinted in Japan, contained a recipe identical to the one his family had for generations been treating as secret.

After some thought, Gen’an came up with an explanation for what had happened: just as his ancestor had learned the recipe from the Dutch doctor Seirukettan, the Chinese had also learned about the recipe from European visitors in the early seventeenth century, and Chen Sicheng had tried to pass the recipe off as his own. Indeed, it is curious to note that Chinese and European doctors started using mercurial drugs to treat syphilis quite soon after the disease’s arrival, and difficult to rule out the possibility that maritime contacts might have allowed one medical culture to learn the idea from the other. However, European and East Asian mercurial therapies for syphilis differed in their details, and it seems safer to conclude that they developed independently.

An army of antisyphilitic drugs defeats the syphilis demons. Funakoshi Kinkai, Illustrated Syphilis War Tales (1838). Courtesy of Waseda University Library.

But why did the Haruhi lineage attribute their seiseinyū recipe to the mysterious Dutch doctor “Seirukettan”? It isn’t possible to offer a definitive answer to this question, but a little background knowledge concerning the circulation and uses of medical texts in eighteenth-century East Asia is enough to suggest a plausible hypothesis.

Patients seeking treatment for syphilis tended to seek out specialists in “wound medicine” (yōka), a field that included the application of topical cures as well as surgical techniques. Eighteenth-century Japanese doctors were highly receptive to learning about European surgery and anatomy, which were much more developed than their East Asian counterparts at that time; it would thus be unsurprising for a Japanese practitioner of wound medicine to think of a cure for syphilis as more likely to be European than Chinese.

Nevertheless, Japanese practitioners of wound medicine would also have known about the Chinese surgical tradition and probably would have kept some classic Chinese treatises on surgery in their library. One such treatise was the Complete Book of Proven Remedies for Wounds and Ulcers (Chuangyang jingyan quanshu 瘡瘍經驗全書): the first edition of this book was produced in 1569, and an expanded edition published in 1717 included an additional chapter entitled “The Secret Record of Syphilis” – this was none other than Chen Sicheng’s book, incorporated into the new edition without attribution.

Early modern Japanese doctors often made their own manuscript copies of such medical treatises, especially if the only available printed versions were expensive editions imported through Nagasaki. Sections of multiple books could be copied into a single manuscript volume or a single book could be copied into several separate volumes; as a result, authorial attributions and the status of chapters as sections of larger works could easily become unclear. If something like this had happened to the Haruhi family’s copy of the Complete Book of Proven Remedies for Wounds and Ulcers, the original source of their recipe for seiseinyū could easily become forgotten, giving rise to a family legend that the recipe derived from a Dutch doctor in Nagasaki.

Quantities in Recipes

In my post last month, I discussed an online course in which students contributed their transcriptions of Lady Frances Catchmay‘s recipe book to the Textual Communities site.  Two of these students from Fall 2013 continued their work in the spring, and this post and the next are the results of some of their research.  – Rebecca Laroche

By Kayla Perkins

While transcribing a rather long recipe, “A good Receyte to make matheglin” from Lady Frances Catchmay’s manuscript, I came across a peculiar phrase, “. . .toe hundred gallans of fayre water or more. . .” Two hundred gallons?! Even now that would be quite a feat.  Why so much water? This question drove me to discover what Metheglin is and if using such amounts of water was common practice.

Metheglin is a spiced or medicated variety of mead, originally popular in Wales. Starting with the Wellcome’s amazing cache of recipe manuscripts, I first compared Catchmay’s recipe to six others. Each recipe I read through made Catchmay’s preparation more unique, and they were all variants of each other. Despite the variants, however, there are aspects that each of the recipes share.

For instance, in preparing Metheglin, the two main ingredients are equal parts honey and water boiled together. They are continually boiled until the mixture can ‘bear an egg.’ Catchmay does not include this step within her recipe, which places hers in the minority. For each recipe, the quantities of honey and water vary quite a bit. Where Catchmay calls for two hundred gallons of water, the recipes of Mrs. Carr call for just five gallons, and another recipe calls for just a few quarts.

An assortment of herbs and spices is then added to the brew, the most common among the recipes being ginger, nutmeg, and cinnamon. Catchmay is again differentiating here; her preparation does not call for just a handful of herbs. She has the longest, most detailed list for ingredients required. Not only does she advise when it is best to gather these herbs;[1] she also provides directions on how to clean and prepare said herbs.

While the Metheglin is fermenting one has to periodically skim barm off the top. The liquid is then stored in a variety of containers with directions on how long it keeps and when it should be consumed. Thanks to the honey, Metheglin keeps for quite a while. I believe that is why Catchmay chose to prepare such large batches.

Each recipe shows the different levels of what the compiler expects from her reader. Catchmay is extremely detailed with her ingredients, how they should be prepped, and gives an almost exhaustive description of Metheglin’s preparation. The Jacob recipe, however, is not overly detailed. While, one would like to assume that Catchmay was able to read through and gain inspiration from the others. This is not the case as Catchmay’s manuscript dates to around 1625 and the earliest (Elizabeth Jacobs) of the other recipes is dated at 1654. The six recipes in the Wellcome catalog that I read through share too many similarities for it to be coincidence. I believe Catchmay’s recipe was either isolated from or improved on by others. Either way, it is fascinating to see the uniqueness of the recipes and how each preparer made it her/his own. After becoming familiar with the ingredients, I can see Metheglin being a comforting beverage for these coming winter months. Bottoms up!

[1] Catchmay advises the herbs be picked around the time of Michaelmas and Lammas. Lammas takes place around the 1st of August and is celebrated by baking bread taken from the first harvest. Michaelmas takes place near the end of September, for the feasts of St. Michael. I can see this period of the year being used to stock up for the cold months ahead.


Teaching Recipes as Literary Practice and the Practice of Transcription

By Amy L. Tigner

Students from the University of Texas, Arlington cook using early modern recipes.  Photo courtesy of the author.
Students from the University of Texas, Arlington cook using early modern recipes. Photo courtesy of the author.

As a founding member of Early Modern Recipe Online Collective (EMROC)—an international group whose main objective is to transcribe early modern receipt book manuscripts and upload them onto an open, searchable database—I, along with my EMROC compatriots, wanted to create a labor force to increase the number of transcriptions available to the public.  As many of us teach at universities, we thought it would be a great idea to conceive of classes in which a portion of the workload would contribute to this digital humanities project.  I have to be honest, that at first, I thought of my students as a kind of slave labor, but Rebecca Laroche (see previous blog post in this series) quickly reminded me of the value for students of what we were trying to accomplish as scholars ourselves.  After all, we are in the process of building a new source of knowledge and making viewable to the masses what had previously been seen only by the lucky few. Also, such a process would in fact teach the students valuable skills that they could use if they continued in the profession, or, if not, they could parlay into other pursuits outside academia.  From a traditional point of view, students would need to learn paleography, a necessary skill for any archival work. As more and more libraries are digitizing their manuscripts, there will be quite a lot of this kind of scholarly work for years to come.  Additionally, students would learn how to code their transcriptions into XML in order to upload their work to the database, Textual Communities, which is hosted by the University of Saskatchewan; this skill is clearly transferrable to the outside world.

The first class I taught that incorporated this project was a senior seminar for English majors at the University of Texas, Arlington, in spring of 2013.  The course was entitled “Recipes for Literature/Literature of Recipes.” I have a previous post that details the course, and students in the class also posted blogs about their discoveries in transcribing Wellcome MS108, a receipt book written by Jane Barber.  As this course was really successful (students generally loved working on the project), I thought I would take it to the next level and have graduate students contribute.  So the following year, I taught a class entitled “Food, Women, and Manuscript Culture.”  The course was designed to change students’ conception of the “literary” by broadening the meaning to include household writing, and in particular recipes. The graduate students (like the undergrads) first had to learn how to decipher seventeenth-century handwriting, so I (like Rebecca Laroche) had students use the Cambridge online paleography course, a site that has progressively difficult handwriting examples from the period.

Students from the University of Texas, Arlington cook using early modern recipes.  Photo courtesy of the author.
Students from the University of Texas, Arlington cook using early modern recipes. Photo courtesy of the author.

I found in both classes group work was key, as students get easily frustrated with the kind of work, especially as most were inexperienced with difficult and archaic handwriting. As some students naturally have a greater aptitude for transcription than others, they were to become teachers—good practice for many whose future plans include teaching. When it came to coding, which was a completely new world for me, students had to use problem-solving skills or often relied upon other students to help them. This collaborative atmosphere in the class created a greater sense of intellectual community—and fun.  Another interesting thing happened, however, in the work of transcription and later of coding: students had to slow their reading down, often to the level of deciphering individual letters, in order to read the text.  Thus transcribing and coding turned into close reading, causing students to dig deeper to find meaning.  We often ask students to read a text slowly and carefully, but more and more students are losing the patience to read in this manner. Our experience with the Internet, and with media more generally, trains us to read quickly and to skim the surface but rarely to be contemplative. This digital humanities project brings together traditional scholarly pursuits with twenty-first century technology to contribute to a growing a body of knowledge about the past.