Written by Rosie Maxton.
Wellcome Collection comprises close to 1000 Arabic manuscripts, showcasing a vast array of genres, historical periods and geographical contexts. Around half of these have been digitised and are freely accessible on the Wellcome Collection website. To add to this richness, it was recently discovered that a tiny fraction of the Arabic collection – three manuscripts, to be precise – is transcribed in Karshuni, a method which uses the Syriac alphabet to write Arabic texts. I was tasked with cataloguing this mini subcollection using the Text Encoding Initiative (TEI), a tool currently being used to develop metadata for the entirety of the Wellcome manuscript collection. Below I share some brief observations on cataloguing this fascinating and hitherto unknown dimension to Wellcome’s manuscript collections.
“Are you talking Karshuni?!”
Though the term ‘Karshuni’ (or ‘Garshuni’) can refer to the transcription of other languages in Syriac letters – such as Armenian, Kurdish, Malayalam, Turkish – it is most commonly used in the Arabic context. One of the earliest known examples of Arabic Karshuni dates to the twelfth century, in a manuscript from Mount Lebanon. However, this writing system was most prolific within the context of Eastern Christian communities living in Ottoman territories during the seventeenth and eighteenth centuries. Amusingly, I was informed on a recent visit to Lebanon that the modern-day usage of ‘Karshuni’ in the local Arabic dialect is for unintelligible, nonsensical talk (something akin to gibberish), hence the expression “Are you talking Karshuni?!”
Due to the Covid-19 pandemic, I was unable to access physical versions of the Karshuni manuscripts for cataloguing. Instead, I relied on a series of photos which captured as best as possible the physical, textual and paratextual attributes of the manuscripts.
Although a compact set, the Karshuni manuscripts strongly encapsulate the thematic diversity of the Wellcome Arabic collection. While one manuscript is – not uncommon within the Karshuni literary tradition – a Christian theological treatise (a translation of the Hexaemeron of Jacob of Edessa (d. 708)), the other two consist of a medical treatise (the Kitāb al-Iqtiḍāb of Ibn al-Masīḥī (d. 1259/60)) and an anonymously authored text on astrology and physiognomy (entitled Kitāb al-Abrāj).
The manuscripts are all leather-bound codices, in differing states of preservation, and copied in Serto (West Syriac) script. While a rough estimate might place them in an early modern Ottoman context, and probably – owing to their script – a Syriac Orthodox, Syriac Catholic or Maronite scribal milieu, establishing their provenance with any certainty is difficult. Certain paratextual features of the manuscripts do, however, offer intriguing snippets of information. The colophon of Kitāb al-Abrāj, though undated, reveals that the manuscript was copied by a student called Mīkhāyīl in Batroun, a town in northern Lebanon; and a note of ownership on the final folio of Kitāb al-Iqtiḍāb is dated 1875, implying its creation predates this time. Quite wonderfully, we get a glimpse into the usage of this particular manuscript from copious notes scribbled throughout it, almost all of which are in Arabic script (rather than Karshuni): aside from recipes for ailments such as swelling and shortness of breath, we learn that after the death of its owner, the doctor Yūsuf son of Mūsá Maqṣūd, the manuscript was passed on to his brother Bishārah.
Thoughts for the future
The Karshuni manuscripts in Wellcome Collection have much to offer researchers. Notwithstanding their value to studies of codicology or even acquisition history, the texts they comprise are in themselves extremely significant: the Hexaemeron of Jacob of Edessa, for example, marks a new contribution to the handful of Karshuni versions of the text known today. Moreover, while TEI provided an excellent platform for conveying the detail within these manuscripts, I observed that the XML editor I used did not support display of the two types of scripts – Syriac and Arabic – at once. Given the frequency with which these scripts coincide in Karshuni manuscripts, solutions to this technical issue may be worth consideration. At any rate, I hope this partnership between developing digital humanities and exploring forgotten materials will long continue on its exciting path.
 Syriac being a dialect of Aramaic originating in Edessa, current-day southeast Turkey, during the first century CE.
 That is, in the Rabbula Codex: see Alessandro Mengozzi, ‘The Evidence of Garshuni as a Writing System: Evidence from the Rabbula Codex’, in Frederick Mario Fales and Giulia Francesca Grassi (eds.), Camsemud 2007: proceedings of the 13th Italian Meeting of Afro-Asiatic Linguistics (2010), pp. 301-2.
 Francisco del Rio Sanchez, ‘Arabic-Karshuni: An attempt to preserve the Maronite identity: The Case of Aleppo’, The Levantine Review 2.1 (2013), pp. 7-8.
 Similar observations are made in Mengozzi, ‘The Evidence of Garshuni as a Writing System’, p. 297, note 1 and Sanchez, ‘Arabic-Karshuni’, p. 4.
 Full title is given as al-Iqtiḍāb al-majmūʿ ʿalá ṭarīq al-masʾalah wa-al-jawāb.
 I am grateful to Dr Salam Rassi for his help in deciphering this partially erased text:
 McCollum, Adam C., ‘Remarks on Recent Cataloging Efforts among Syriac Manuscripts Preserved at the Hill Museum & Manuscript Library’, Hugoye: Journal of Syriac Studies 15.2 (2012), pp. 367-8.