Expressing Complex Associations in Medieval Historical Documents: The Henry III Fine Rolls Project

This paper will focus on the use of technologies
traditionally associated with knowledge management and
ontological representation to express complex associations
between entities in historical texts that have been marked up
in XML according to the Text Encoding Initiative guidelines.1
In particular, we will describe our exploration of the potential
use of RDF (Resource Description Framework)/OWL (Web
Ontology Language) technologies and will reflect on the role
of an ontology in facilitating the interpretation of implicit and
hidden associations in the sources, examining its use and limits
in a digital humanities project in connection with editing tools
and delivery issues.
We will demonstrate our findings based on the Henry III Fine
Rolls project,2 where an RDF ontology is being developed to
make explicit information about person, place and subject
entities marked up as “instances” in the core texts themselves.
The Henry III Fine Rolls project and
the need for authority lists
The Henry III Fine Rolls is a three-year collaborative
project between King's College London and the National Archives (UK) that aims to represent the complexity of a
historical object known as the 'fine rolls' which chart offers of
money made to King Henry III of England in exchange for a
wide range of concessions and favours. A total of 64 rolls
containing around 800 parchment membranes, one for almost
all of the fifty-six years of Henry III’s reign from 1216-72,
survive in the UK National Archives.3 Each fine roll was
compiled in Latin by a handful of scribes and taken as a body
of documentary evidence, the rolls are of “prime importance
in the study of political, social, and economic history and of
government and administration at a local and national level.”4
The project will cover the first thirty-two years of Henry III’s
reign down to 1248 and will result in both print and digital
editions of the rolls, using as a model the format of the
traditional printed 'calendar' (an English summary of records,
plus a set of indices) in connection with the digitised images
of the rolls themselves. The digital edition will have a
sophisticated search interface and both print and digital editions
will provide a series of indices for people, places and subjects
that include complex associations between the various entries.
The core texts were encoded in XML using the TEI guidelines
and include mark-up related to some aspects of the physical
structure of the rolls as material artefacts.5 Particular attention
has been given in the mark-up to the occurrences of names of
persons, places and institutions. Furthermore, the project
researchers have identified and marked-up relevant subjects in
the fine rolls. Therefore, while the general mark-up in TEI XML
provides a framework for theoretical analysis and practical
encoding of the text of the calendar as it is being edited and
summarised in English, the need for additional modelling has
emerged so as to:
• associate textual instances of the same person, subject
occurrence or place to their correspondent logical authority
—whether that be a person (identified or anonymous), a
subject class or a place;
• express the mutual relationships between the relevant
authorities (e.g. individuals, locations, institutions and
subjects).
Why RDF/OWL?
In previous work on projects requiring an expression of the
associations between sources in core texts marked up using
TEI XML we have taken more traditional approaches using
XML structures or relational databases to solve the problem,
but have increasingly found these wanting. There has been
some research in the TEI community in this area lately,
particularly around 'biographical and prosopographical data'6,
but at the time of writing this was not mature enough for us to
make a commitment to using it and in any case our attention
was drawn to other standards whose main objectives are closer
to what we are trying to achieve.
After conducting a comparative evaluation of possible standards
to create authority lists structures that included research into
RDF/OWL, Topic Maps and MADS (Metadata Authority
Description Schema), we opted for RDF/OWL.
The main reasons for this choice were the following:7
• RDF/OWL models provide a logical organisation of data
together with the possibility of a flexible manipulation of
meanings (e.g. rich expression of relationships among
objects/entities mentioned in the source text, where an
object/entity might be a person, a place or a subject);
• RDF/OWL decreases ambiguity by identifying unique
entities independently from their instances in a decentralised
way.
While the more practical advantages are that:
• as a set of standards which are at the heart of the Semantic
Web, RDF/OWL is internationally recognised and supported
- its models could facilitate the interconnection between
humanities computing projects in general and between
directly related data in particular;
• there is a relatively good selection of tools for RDF/OWL
compared to the other technologies that we have explored;
• RDF/OWL can be expressed in XML format, thus allowing
the re-purposing of data for web delivery (e.g. creation of
indices through the use of XSLT).
Ontology: Structure, Tool and
Delivery
The RDF schema and OWL are used to define the
knowledge domain around the core materials on the
project (the fine rolls) using classes and predicates. Already
existing predicates are used as extensively as possible (e.g.
CIDOC-CRM, Dublin Core, Friend of a Friend, Simple
Knowledge Organization System).
The connection between the TEI source files and the ontology
play an important role in our model.8 Indeed the XML files not
only feed the ontology with some data (e.g. variants of a certain
person name), but they are themselves part of the ontological
model. The fact that they exhibit names and assert facts related
to those names can be modelled in the ontology together with
the statement that a person name (proper name or in general
reference string) is associated to a specific person as an instance
of the entity person (e.g. that has a gender, a possible status
etc.). Quoting Eide,9 we include “explicit statements of the
sources of the assertions exhibited in the text”: The concrete connection between the XML files and the
ontology is maintained through the use of identifiers, as in the
example below:
<persName key="arsic_robert"> Robert Arsic
< / p e r s N a m e > → < f r h 3 : M a n
rdf:ID="arsic_robert"> … </frh3:Man>
Evidently, the value and granularity of the ontology depends
on various factors, which include the richness of the source
files themselves (e.g. how well do the fine rolls express the
reality of the roles/occupations in the thirteenth century?), the
specific interests of the researchers (e.g. in the types of fines
rather than in the amount of money they keep record of) and
the definability of the knowledge domain itself (as shown in
the contrast between the identification of individuals and the
much more blurred identification of subjects).
In addition to expressing complex associations in an abstract
intellectual sense, it was also crucial to the project to create a
system endowed with sophisticated facilities for editing and
final publication. We will describe our experience in creating
such tools, in particular taking into account such project
requirements as:
• the facilities for managing/editing data: the addition of new
information, new classes/predicates/literal values; the
definition of relationships between entries, the possibility
to merge instances of classes or subclasses (e.g. while
editing we may realise that person A and B are actually the
same);
• the connection and synchronisation between the ontology
structure and data on the one hand, and the TEI source files
on the other (e.g. links from classes in authority list to actual
references to those classes as exhibited in the mark-up);
• the facility to browse information alphabetically or
categorised in some way (e.g. by date, county, role, kind of
relation) and to search for particular associations as
expressed in the ontology (e.g. all the locations connected
to the person called 'Robert Arsic' or all the relatives of
Robert Arsic).
Conclusions
In the case of the fine rolls (as is true of many projects
involving complex primary sources) it is not enough to
mark-up an occurrence of a person name if you wish to create
complex indices or to create structured search functions.
Moreover, an authority-based approach is essential in order to
make the resource interoperable; external authority lists are
needed to record information in a systematic, comprehensive
and possibly standardised way, so as to:
• store additional information related to persons, places and
subject (exhibited and marked up as corresponding instances
in the TEI XML files);
• make explicit the multiple connections between places,
persons and subjects;
• merge or disambiguate identifications and eventually correct
the original mark-up of the rolls.
Our paper will describe how, for the Henry III Fine Rolls
project, an RDF/OWL ontology was used to model complex
associations and how this has assisted the project researchers
in the interpretation of the material they are editing and
facilitated the production of a digital resource that will allow
future users to browse the material under different perspectives,
to explore the relationships among mentioned individuals,
locations and subjects, and to foster new understandings.
1. <http://www.tei-c.org>
2. <http://www.frh3.org.uk>
3. <http://www.nationalarchives.gov.uk/>
4. Dryburgh, Paul. “Henry III Fine Rolls Project” (paper presented
at the Institute of Historical Research, London , February, 9, 2006).
5. For a more detailed presentation of the mark-up model see Ciula,
Arianna. “Searching the Fine Rolls: A Demonstration of the
Electronic Version” (paper presented at the International Medieval
Congress, University of Leeds, July 10-13, 2006).
6. See “Biographical and Prosopographical Data”. In
Sperberg-McQueen, C. M., and Burnard, Lou, eds. "TEI P5
Guidelines for Electronic Text Encoding and Interchange". <ht
tp://www.tei-c.org/release/doc/tei-p5-d
oc/html/ND.html#NDPERS> (accessed 15 November
2006). 7. OWL is a language developed on top of RDF by W3C to write
ontologies. For W3C RDF specifications on RDF see <http:
//www.w3.org/RDF/>. For an overview and further
resources on OWL see <http://www.w3.org/2001/s
w/>
8. For a similar approach see Eide, Øyvind and Ore, Christen-Emil.
“TEI, CIDOC-CRM and a Possible Interface between the Two”
(paper presented at Digital Humanities 2006, Université
Paris-Sorbonne, July 5-9, 2006) <http://www.allc-ac
h2006.colloques.paris-sorbonne.fr/DHs.p
df>.
9. Eide, Øyvind. “The Exhibition Problem. A Real Life Example
with a Suggested Solution” (paper presented at Digital Humanities
2006, Université Paris-Sorbonne, July 5-9, 2006) <http://w
ww.allc-ach2006.colloques.paris-sorbonn
e.fr/DHs.pdf>.
Bibliography
Boot, Peter. "Decoding Emblem Semantics." Literary &
Linguistic Computing 21(Supplement 1) (2006): 15-27.
Ciula, Arianna. "Searching the Fine Rolls: A Demonstration
of the Electronic Version." Paper presented to the International
Medieval Congress 2006, University of Leeds, July 10-13,
2003.
Dryburgh, Paul. "Henry III Fine Rolls Project." Paper presented
to the Institute of Historical Research, London , February 9,
2006.
Eide, Øyvind. "The Exhibition Problem. A Real Life Example
with a Suggested Solution." Paper presented at Digital
Humanities 2006, Université Paris-Sorbonne, July 5-9, 2006.
Abstract available at <http://www.allc-ach2006.co
lloques.paris-sorbonne.fr/DHs.pdf> (accessed
15 November 2006).
Eide, Øyvind, and Christian-Emil Ore. "TEI, CIDOC-CRM
and a Possible Interface between the Two." Paper presented at
Digital Humanities 2006, Université Paris-Sorbonne, July 5-9,
2006. Abstract available at <http://www.allc-ach200
6.colloques.paris-sorbonne.fr/DHs.pdf>
(accessed 15 November 2006).
Poupeau, Gautier. "De l’index nominum à l’ontologie. Comment
mettre en lumière les réseaux sociaux dans les corpus
historiques numériques?" Paper presented at Digital Humanities
2006, Université Paris-Sorbonne, July 5-9, 2006. Abstract
available at <http://www.allc-ach2006.colloque
s.paris-sorbonne.fr/DHs.pdf>(accessed 15
November 2006).
Spence, Paul. "The Henry III Fine Rolls Project." Paper
presented at Digital Humanities 2006, Université
Paris-Sorbonne, July 5-9, 2006. Abstract available at <http:
//www.allc-ach2006.colloques.paris-sorbon
ne.fr/DHs.pdf> (accessed 15 November 2006).
"Biographical and Prosopographical Data." TEI P5: Guidelines
for Electronic Text Encoding and Interchange. Ed. C. M.
Sperberg-McQueen and Lou Burnard. Accessed 2006-11-15.
<http://www.tei-c.org/release/doc/tei-p5-
doc/html/>
Tuohy, Conal. "Using XML Topic Maps to Present TEI." Paper
presented at the 5th TEI Meeting, Sofia, Octobery 28-29, 2005.