Manuscripts, Mapping, and Provenance

The Digital Orientalist has given us many examples of the uses of digital cartography or, more simply, mapping. Ranging from Geographic Information System (GIS) applications, used for instance for mapping ancient Indian urban spaces, to more out-of-the-box tools like Leaflet, the benefits of mapping are manifold.

Mapping is a powerful tool that allows us to interrogate, explore, and visualise geographical data. Here I want to primarily focus on the potential of mapping for data that we might not always think of in geographical terms, as well as some possible pitfalls and blind spots. While working on Wellcome Collection’s early Hindi medical manuscripts, it struck me that it might be helpful to explore how one could map their metadata. So what information is available to us?

Colophons

South Asian manuscripts frequently feature colophons in which the scribe offers some information about themselves and the context for the copying of the manuscript. If you’re very lucky, the colophon can give a wealth of information. Sometimes this scribal postscript can even be more interesting than the text itself!

Most colophons are, sadly, brief and perfunctory. What I like to refer to as ‘informal’ manuscripts, such as notebooks, random recipe collections, and similar, tend not to have any extraneous information beyond scattered marginalia. Wellcome Collection’s Hindi holdings are especially rich in these, giving us a high number of completely unique manuscripts but precious little metadata to go on. Out of nearly 200 medical manuscripts in early versions of Hindi, only 19 have colophons that provide definite information on the date and location of when and where they were copied.

TimeMapper

What can these 19 manuscripts tell us? I used TimeMapper, an exceptionally straightforward (and free!) mapping tool from Open Knowledge Labs, to place the 19 on both a map and a timeline. I first extracted the date and location from the pdf of Friedlander’s catalogue of Hindi manuscripts and simply inserted these into a Google spreadsheet compliant with TimeMapper’s functionalities. As such, the challenges were not technical, but rather in the interpretation of metadata, such as the conversion of Vikram Samvat-dates into the Gregorian calendar (unless the lunar month is specified, you’ll end up with a +/-1 error margin) and figuring out the modern-day sites of earlier place names.

The output, in the form of a map and a timeline, is easy to publish and embed on websites, and it lets you click your chronological way through the manuscripts as well as studying their locations on the map.

Looking at the distribution of the manuscripts across the map, it is oddly pleasing to see the pointers form a triangle between sites in Punjab, Rajasthan, and around Agra. While the former indicate the predominance of Punjab-based physicians and scribes in the production of early Hindi medical material, the two latter sites, Rajasthan and Agra, point to the some of the main trading routes in early modern North India, along which early forms of Hindi travelled, spread, and was adopted. Similarly, the lone Hindi manuscript from Tamil Nadu stands out more clearly.

Provenance

So far, so good, and in many ways the map confirms expectations we may have about the spread and circulation of these manuscripts. However, seeing the metadata laid out on the map like this can also give us a false sense of objectivity in our interpretation. For what the metadata from the manuscripts and the map fail to give us is a sense of how the manuscripts were collected in the first place.

This leads me to the essential question of provenance. Wellcome Collection’s manuscript holdings were gathered in myriad ways, over several decades. Many were purchased from UK-based auction houses, but large numbers were collected by purchasing agents employed directly by Henry Wellcome himself. Some of them, like the ever-fascinating Paira Mall, spent years travelling across South Asia and the Middle East and buying manuscripts.

We know that Paira Mall, himself a Punjabi, purchased great amounts of material from manuscript purveyors based in the Punjab and in Rajasthan. The neat triangle on our map may indeed not primarily reflect the spread and circulation of early modern medical material in Hindi, but rather Wellcome’s early 20th-century manuscript purchasing networks.

Accession registers

What can we do to remedy this situation? There are many aspects to consider, but an important starting point is research into the provenance of the manuscripts. In this, accessions information is invaluable as it gives what often is the very first record of an item’s history in an institution. If we have this information, we can begin to work backwards and uncover the histories that our map does not capture.

But accession registers can be difficult to track down, and even harder to navigate. Many of Wellcome Collection’s registers are digitised and available online, but still retain obstacles to access. If there’s no known accession number for the item you’re researching, you’re forced to trawl through thousands of handwritten pages.

Wellcome Collection is now working to address these issues. Through Transcribe Wellcome, a crowdsourcing project spearheaded by Dr Alexandra Eveleigh and other members of staff, several decades of accession registers are being transcribed and made available free and, crucially, searchable. The website is launching to the public in the future and will hopefully inspire maps that tell fuller stories of the lives of manuscripts.

Image: Sir Patrick Hehir, Malaria in India. Wellcome Collection, CC BY 4.0.