The core task of the DiGA project is the digitization of two collections of Gandharan sculptures. But in doing so, it aims at advancing the research of Gandharan art in the digital space. We start from established practices in the field of Gandharan studies and bridge them with best practices from digital cultural heritage studies. One central building block for this is the Repertory of Terms for Cataloguing Gandharan Sculptures, a central reference work, that we aim to bring into the Linked Open Data space.
Art history increasingly embraces liked data, but existing vocabularies are mainly covering Western art. Gandharan studies, on the other hand, produced thorough reference works, but they exist only in printed form.
The Pelagios working group Linked Data Methodologies in Gandhāran Buddhist Art and Texts allows us to think beyond the DiGA project itself and place it into the broader landscapes of both digital humanities and Gandharan studies. For its final report, we looked at best practices in Linked Open Data, relevant reference works in Gandharan and Buddhist studies, as well as successful projects combining the two. One issue that we noticed quickly is that there is a significant gap: Art history increasingly embraces liked data, as established vocabularies like IconClass or the Getty Arts and Architecture Thesaurus show. But these are mainly covering Western art and are severely lacking in the field of Buddhist iconography. Gandharan studies, on the other hand, produced thorough reference works, but they exist only in printed form. A prime example of this is the Repertorio Terminologico per la Schedatura delle Sculture dell’Arte Gandharica by Domenico Faccenna and Anna Filigenzi. It contains detailed, thorough terminology covering a wide array of topics: architectural features, decorative motifs, people, fauna, flora, weapons, musical instruments, ceremonial objects, everyday objects, furniture, and means of transport. This terminology was assembled over the course of several years and finally published in 2007 in Italian and English. A new bilingual version in Chinese and English was recently published. One task we gave ourselves for DiGA is to build a linked data vocabulary from it.
With its over 250 pages, digitizing the Repertorio manually would be a painstaking task. On the other hand, its sections are structured very clearly, so it is a good case for semi-automatic conversion. Our final goal is to have a version of the Repertorio in SKOS format and link it to other existing vocabularies where possible.

The OCR itself is not too much of a challenge here, as we can work with clean print and acceptable scans of it. But in order to make most of the existing structure, we need a way to maintain layout information in the process. Then we will be able to extract the terms in structured form. Finally, we will convert them into SKOS format and manually go through all entries to check the results of the automatic extraction. We will discuss the approaches we took to each step and the experiences we made in future posts.
Albeit extensive in its scope, the Repertorio is not exhaustive. A panoply of additional terms is needed to ensure a standardized description of all types of Gandharan sculptures. Here, the Pelagios Working Group has identified further sources which would complement and enrich the Repertorio, notably with respect to the labelling of Buddhist narratives or of Persons (i.e. historical persons or literary persons mentioned in Buddhist legends). These sources are listed below but you’ll have to read through our Pelagios Group report to find out about their respective merit and value for the aim of the DiGA project. Eventually, our vocabulary will incorporate additional concepts from these sources and grow beyond just being a digital version of the Repertorio. In addition to extending the vocabulary, we also plan to align our vocabulary to external vocabularies such as IconClass or the Getty Arts and Architecture Thesaurus and when possible, add equivalent labels from Indic terminology. Why not avoid any semantic dichotomy between the terms used and the objects they describe while we’re at it?
Next in series: From print to LOD, part II: Extracting structure
Key reference works
Faccenna, Domenico. 1962. Sculptures from the Sacred Area of Butkara I, Swat, Pakistan. Vols. II, 2-3. IsMEO Reports and Memoirs. Rome: Istituto poligrafico dello Stato, Libreria dello Stato.
Foucher, Alfred. 1905–1951. L’art Gréco-Bouddhique Du Gandhâra: Étude Sur Les Origines de L’influence Classique Dans L’art Bouddhique de L’Inde et de L’Extrême-Orient. Ernest Leroux. Paris: Imprimerie Nationale.
Hinüber, Oskar von. 2016. “Some Remarks on Technical Terms of Stūpa Architecture.” Annual Report of the International Research Institute for Advanced Buddhology at Soka University 19: 29–46.
Zwalf, Wladimir. 1996. A Catalogue of Gandhara Sculptures in the British Museum. 2 volumes vols. London: British Museum.
OpenEdition suggests that you cite this post as follows:
Frederik Elwert (April 29, 2021). From print to LOD: A vocabulary for Gandharan art. Digitization of Gandharan Artefacts. Retrieved June 19, 2025 from https://digartefacts.hypotheses.org/14
One Reply to “From print to LOD: A vocabulary for Gandharan art”