The use of imperial media in representing India and its people was an important aspect in the consolidation of colonial rule. This poster examines how the analysis of the representation of Indian individuals’ links to colonial consolidation of power using highly detailed encoded metadata in sources. I will be demonstrating how the development of an overlapping and layered approach to metadata in encoding can better represent and therefore aid research into media representations.
The portrayal of the Indian nobility exemplifies a powerful, exotic, yet ultimately submissive form of Imperialism to the British, and the presentation of the dynasty of the female rulers of Bhopal is a prime example of this. My work charts the ways in which these female rulers were represented and misrepresented as a means of demonstrating Indian animosity and consolidating imperial power. As favoured rulers of a ‘native state’, the Begums are frequently discussed in newspapers or as the subject of photographs and while authoring letters and other personal papers.
Shaharyar M. Khan,
The Begums of Bhopal: A Dynasty of Women Rulers in Raj India, (London: I.B. Tauris) p. 102, p. 171
Combining colonial textual and visual sources allows for a comprehensive view of how these rulers were envisioned to live. Yet, the overall image of the Begums is highly contradictory; for example, in text, these women are typically portrayed as being veiled or in purdah while photographs show them to have almost masculine characteristics, embodying features of any great Indian ruler.
Bourne and Shepherd,
Sultan Shahjahan Begum, Begum of Bhopal: Prince of Wales Tour of India 1875-6, 1875-6, 27.3 x 23.3 cm, Albumen print
Sydney Prior Hall,
The Reception of the Begum of Bhopal, 1877, 16.0 x 24.3 cm, Pencil and watercolour
The contradictory nature of representation has been a central theme in my research. How were the Indian nobility represented by the British? Were these contradictions central to the way in which Indian nobles were represented, or were they discrete mistakes, demonstrating inconsistency in the understanding of these figures by various British and Indian commentators? Answering these questions required me to uncover patterns in the depiction of these individuals, and their world, to develop a model of the perceived colonial Indian aristocrat. Moreover, it requires understanding how these patterns can transfer from textual to visual sources.
To contrast physical and conceptual elements within a multi-modal dataset in a robust and reproducible manner requires detailed and nuanced hermeneutic encoding.
This poster will, therefore, explore the use of digital metadata analysis in the field of representation. It will demonstrate the cross-referencing of digitalised newspaper material from the British Library, Trove (Australia), Papers Past (New Zealand) and Chronicling America (USA), with biographies, autobiographies and personal letters held by the British Library and the India Office Collection and digitalised and physical photographs and drawings from throughout the Commonwealth in various combinations and at multiple resolutions. Current text encoding practises, TEI specifically and XML more broadly, do not allow for two overlapping elements of metadata, nor does annotation of visual material do so without damaging or destroying the integrity of the image.
Tufte, Edward, Beautiful Evidence, Graphics Press LLC, 2006
Although multiple layers of analysis within XML documents, namely multi-rooted trees or using XML linkages and then subsequent transformations to fully represent multiple hierarchies are possible, this is technically complex and requires both a definitive and transformed version of the document for recording and analysis. My system allows for straightforward annotation of a single text document with multiple overlapping segments in the same manner as it annotates full or segmented images. This unity of input and analysis is more appropriate to my specific case, comparing visual and textual representations of a common event or group. My poser will demonstrate a custom-designed database system that allows for consistent layering of metadata on textual and material historical objects, digital reproductions, and enhanced fragments – sections of textual or visual objects that have been transcribed, cropped or otherwise computationally transformed. Building upon concepts from the semantic web, the system allows the user to apply metadata and provenance details recursively between the original object and discrete fragments as well as an encoded controlled vocabulary on the content of those objects. This layering of metadata allows for a much deeper analysis of historical sources, as well as being able to maintain the integrity of the original object in the project’s documentation.
The database, written in Python 3.6 and stored in a plain text JSON file, functions as both an input and retrieval mechanism, allowing users to input data sources either manually or through templated input files. In addition to standardised MARC and Dublin Core provenance metadata, the database will encode both the aesthetic and artistic properties (source medium, structure, composition), and the historical characteristics (subject matter and historical relevance). This will allow for consistent cross-referencing that may be difficult if not impossible to determine by human eye alone. Using metadata allows for a much more robust comparison between textual and visual sources, highlighting trends that may otherwise never be brought to light. Finally, although this level of metadata analysis requires a significant temporal commitment to data entry, the deep data created will inform not only the present research regarding representation but provide consistent training for unsupervised computation analysis in the future.
Khan, M. Shaharyar, (2000)
The Begums of Bhopal: A Dynasty of Women Rulers in Raj India, London: I.B. Tauris
Beals, M. H., (2016)
The Platonic Ideal of a Newspaper Article, mhbeals.com/the-platonic-ideal-of-a-newspaper-article
Tufte, Edward, (2006),
Graphics Press LLC, ids-pub.bsz-bw.de/frontdoor/index/index/docId/5878
If this content appears in violation of your intellectual property rights, or you see errors or omissions, please reach out to Scott B. Weingart to discuss removing or amending the materials.
Hosted at Utrecht University
July 9, 2019 - July 12, 2019
436 works by 1162 authors indexed
Conference website: http://staticweb.hum.uu.nl/dh2019/dh2019.adho.org/index.html
Series: ADHO (14)