You are here:

“Roland HT”, a hypertext and corpus study

, Brown University, United States

Brown University . Awarded


RolandHT is centered around the protagonist of the 11th-century French Song of Roland. In the past 900 years Roland has been appropriated by many Western cultures. He has often mutated drastically in the process of cultural transmission, becoming a multifaceted, often self-contradictory archetype. Roland is a cultural hypertext, an entity composed of segments interlinked in such a complex way that they are impossible to faithfully represent on paper.

The works under consideration form a corpus held together by recurring themes, imagery and characters. In collaboration with a programmer I present a representative set of semantically enoded corpus excerpts, with a web interface that provides a means for readers to acquaint themselves with the corpus following thematic threads that they find compelling.

Heuretics—a heuristic approach to theory—guides this experiment in humanities computing scholarship. Semantic markup is apposite to expressing concepts that are poorly served by linear prose. RolandHT contributes to the study of cultural transmission and adds to an understanding of art's function in social history. I take up the issues of objectivity and originality, and their role in this dissertation's intellectual contribution to cultural studies. The electronic project (which is a more versatile entity than are extant physical and digital archives) is a pedagogical tool; some suggestions for teaching using RolandHT are provided.

RolandHT makes a case for working in ways—and with tools—unfamiliar to most humanities scholars. Use of computation to make a scholarly argument are described in detail. Strengths and weaknesses of such an approach are addressed. I continue a line of research advanced by researchers of social computing and propose that humanists stake out plots of the electronic noosphere. Networking—not only interpersonally but also computationally—will allow us to conduct research more quickly and effectively.

This dissertation is a proof of concept: the corpus is far too large to process in the time allotted for research and writing. In the final chapter I discuss some preliminary statistical results, and propose ways to address the inevitable incompleteness of the current project by offering likely future directions for RolandHT.*

*This dissertation is a compound document (contains both a paper copy and a CD as part of the dissertation). The CD requires the following system requirements: Internet browser.


Zafrin, V. “Roland HT”, a hypertext and corpus study. Ph.D. thesis, Brown University. Retrieved March 24, 2019 from .

This record was imported from ProQuest on October 23, 2013. [Original Record]

Citation reproduced with permission of ProQuest LLC.

For copies of dissertations and theses: (800) 521-0600/(734) 761-4700 or