Queste del Saint Graal: Textometry Platform on the Service of a Scholarly Edition

In this poster/demo we will present an online
scholarly edition of
La queste del saint Graal
(
The Quest for the Holy Grail
) based on a
manuscript of Lyons public library (Lyon, BM,
P.A. 77) and built in the Textometry platform
(TXM). The particularity of this edition is that
it combines rich paleographic and philological
data (including digital photographs, various
layers of transcription, translation in modern
French and editorial notes) with advanced
linguistic search and analysis tools provided by
textometric software.
Despite some damage (torn miniature in the
beginning, missing folios in the end), Lyons
manuscript is considered to be one of the best
witnesses of the
Queste del saint Graal
, a
13
th
century French novel, part of the famous
"Arthurian" prose cycle. A well-known edition
by Albert Pauphilet based on this manuscript
was first published in 1923 and has been
regularly reprinted ever since. However, this
edition cannot be used as a trustworthy source
of linguistic data, as it contains multiple
corrections, and the readings of the primary
source are not always accessible.
In the late 1990s Christiane Marchello-Nizia
started working on a new edition, which was
supposed to be closer to the witness, explicitly
correcting only doubtless scribal errors and
equipped with multiple tools to assist the reader
in understanding the text and to explore various
features of its language. This work was to a
large extent inspired by the experience of the
Charrette
Project (Pignatelli & Robinson 2002).
The first prototype of the edition was published
on the web in 2002, and a completely new
version was released in 2009. The edition is still
under construction but its main components are
already in place and can be accessed on the Web
through an early TXM prototype.
These are the following:
-
Old French text (108 000 words);
-
digital facsimile of the manuscript (418 text
columns);
-
modern French translation;
-
scholarly introduction;
-
tools for textometric analysis.
The Old French text is established according
to rigorous editorial principles stated in the
Introduction. A particular feature of the edition
is the respect of scribal punctuation. It is
encoded according to the XML-TEI standard
with special attention to linguistic data. All
words are explicitly tagged and annotated
for part of speech using the Cattex2009
tagset (Prévost&al, to be published). The text
can be displayed and searched in several
presentation forms: a "traditional" view (
vue
courante
), close to the norms of French critical
editions (Vielliard & Guyotjeannin 2001),
a "diplomatic" view respecting linguistically
significant graphical features of the manuscript
(such as the absence of "phonetic" distinction
between
u
and
v
), and a "facsimile" view
where all noticeable graphical distinctions of the
primary source are represented. For instance,
medieval abbreviations are tacitly expanded in
the traditional view, they are expanded but
typographically highlighted (using italics) in the
diplomatic view and they are represented by
abbreviation marks in the facsimile view. The
concept of three different views of the source text
is based on the multi-level transcription model
of the Medieval Nordic Text Archive (Haugen
2004). MUFI character codes (Haugen 2009)
are used for "special" medieval characters in the
facsimile view (such as abbreviation marks or
letter variants). A MUFI compliant font (such as
Andron Scriptor Web) needs to be installed on
the system to display these characters correctly.
The three views of the Old French text, the
translation and the photographs can be browsed

2
individually or side-by-side in two columns. The
‘diplomatic’ and ‘facsimile’ views, as well as the
translation, are only available for a few folios at
present but they will be progressively edited for
the whole text.
The innovative aspect of this new edition
consists first of all in the integration of
textometric research tools. Those tools assist
the reader by offering qualitative services
like full text search engine to build KWIC
concordances and browse the text through
them, and quantitative services like comparing
statistically the occurrences of different
linguistic phenomena in various parts of the
text or analyzing statistically their collocational
attraction inside various contexts, such as
sentences.
One can open the textometric panel by clicking
on the
Outils
(‘tools’) button in the bottom of
the window. The platform makes it possible
to search for linguistic data (words, parts of
words, parts of speech, phrases, etc.) in any
presentation form using the powerful CQP query
language (IMS Open Corpus Workbench) and
to display KWIC concordances of the search
matches with customizable context size and sort
options. Concordances are displayed in a new
tab next to the search form. The corresponding
page (or column) of the edition is displayed upon
a click on a line corresponding to an occurrence
in the concordance. The words matching the
query are then highlighted (see the figure
below).
Figure 1
In the future, all the main textometric tools
will be interfaced in the edition. These
include specificity, collocates, lexicograms,
etc. The source XML-TEI files of the
Old French text and its modern French
translation will be downloadable under
a Creative Commons license (Attribution-
Noncommercial-Share Alike) as soon as the
edition is stable. The editorial and online
textometric research platform is distributed
with an open source license (GPL v3) and can be
used for publishing on the web various texts with
XML-TEI encoding. The platform can handle
any number of texts, up to several hundred
million words in total.
The poster will display the editorial principles
and several screenshots of the edition.
Interactive work with the edition can be
performed at a demo session.
References
Haugen, Odd E.
(2004). 'Parallel Views:
Multi-level Encoding of Medieval Nordic
Primary Sources'.
Literary and Linguistic
Computing.
19.1
: 73-91.
Haugen, Odd E., ed.
(2009).
MUFI character
recommendation. Characters in the official
Unicode Standard and in the Private Use Area
for Medieval texts written in the Latin alphabet.
Part 1: Alphabetical order. Version 3.0.
Bergen:
Medieval Unicode Font Initiative.
Pauphilet, Albert.
(1923).
La Queste del
Saint Graal. Roman du XIII
e
siècle.
Paris:
Champion.
Pignatelli, Cinzia and Molly Robinson,
eds.
(2002).
Chrétien de Troyes : Le chevalier
de la Charette (Lancelot) : le "Projet Charette"
et le renouvellement de la critique philologique
des textes, Œuvres et critique, 27.1.
Tübingen:
Gunter Narr Verlag.
Prévost, Sophie, Céline Guillot, Alexei
Lavrentiev and Serge Heiden
(To be
published).
Jeu d’étiquettes CATTEX2009.
Lyon: Équipe de la BFM.
http://ccfm.ens-lsh.
fr
.
Vielliard, Françoise and Olivier
Guyotjeannin
(2001).
Conseils pour l'édition
des textes médiévaux, Fascicule I, Conseils
généraux.
Paris: CTHS, Ecole nationale des
chartes.
Andron Scriptor font page.
http://www.mufi.in
fo/fonts/#Andron
.
Charrette Project.
http://lancelot.baylor.ed
u
.