Through the Reading Glass: Generating an Editorial Microcosm Through Experimental Modelling

Introduction
The Centre for Scholarly Editing and Document Studies
(CTB) is preparing a digital edition of De Trein der
Traagheid, a novella by the 20th century Flemish author Johan
Daisne. The project initially aimed at a print reading edition,
involving the constitution of a reading text based on a
text-critical analysis of 19 witnesses of the novella's print
history. However absent from the original project proposal, the
TEI markup scheme was adopted early in the project as the
means for digitally representing the edition. Its provisions for
marking up textual variation with the so-called
parallel-segmentation method informed the construction of a
single XML source text containing the transcriptions of all 19
text witnesses under consideration as well as the constituted
reading text, records of their mutual textual variation, and
editorial annotations. The subsequent development of the
electronic edition proved this unitary source text's potential for
modelling a microcosm of user-generated editions. This paper
will focus on characteristics, difficulties, and theoretical
challenges of this particular editorial constellation, as well as
the tools developed for probing into it.
Modelling
Refocusing the goal towards an electronic edition added
an experimental dimension to the project. The lack of well-established models for (creating) electronic scholarly
editions forced us to conceptualise the boundaries of this
particular electronic edition in the course of its development.
Initially, traditional notions of scholarly editing (explicitly
formalised as 'a print reading edition' in the initial project
proposal) provided a good starting point for the development
process. At first, development was guided by mimicking the
familiar print edition model, aimed at generating a reading text
with apparatus variorum from the XML source text. However,
this denotative model of the print edition (McCarty, 2004) soon
evolved to a guiding principle itself for conceptualising new
ideas, an exemplary model for electronic editions. The added
potential of an interactive edition framework allowing for
user-driven input opened up new ways of exploring possible
engagements of the user with the textual tradition. On a
theoretical level, this exemplary model for electronic scholarly
editions informed some challenging insights and rethinking of
the nature of this model's object (the edition).
Technology and tools
On the most basic level, the seminal potential of the XML
source text for our edition could be realised through the
use of several open source XML-related technologies and tools
that are currently being adopted as a standard amalgam for
accessing XML resources. Key technologies for deploying
XML texts like the Extensible Stylesheet Transformation
Language (XSLT) and XML Query language (XQuery) allow
for flexible manipulation and retrieval of XML encoded
information, commonly achieved through dedicated XSLT and
XQuery processors, and native XML databases.The advent of
XML publishing environments like the Cocoon web
development framework has made it possible to integrate these
functionalities in dynamic user interfaces for presenting and
querying XML content via easily accessible delivery
technologies such as a web browser. This integrative potential
stimulated the development of our XML text processing scripts
initially developed as a tentative instrument for a specific task,
to what we named the 'Morkel system', a generalisable suite of
XSLT and XQuery scripts for driving electronic scholarly
editions in an open source software environment.
Views on textual tradition
In the course of its experimental development the Morkel
system became a tool facilitating a multi-faceted user-driven
view on the textual tradition captured in the unitary XML source
text.The scope of this view can be adjusted from micro- to
macro-level. Users can have access to singular texts in the
tradition, by requesting specific versions of the text as
orientation version which presents itself as a faithful
reconstruction of this text version. A broader view on the
tradition can be accomplished by selecting a parallel edition,
in which different episodes in the textual tradition can be viewed
and contrasted, literally next to each other. This parallel
presentation mode of different text versions for visual
comparison resembles that of the Versioning Machine,
developed by the Maryland Institute for Technology in the
Humanities. Finally, the entire textual tradition can be taken
into account when a variorum edition is selected. In its ability
to compare any number of text versions with an orientation
version, this variorum edition is similar to the Juxta tool,
developed by the Applied Research in Patacriticism group at
the University of Virginia. The focal point of this text
comparison in the Morkel system is the contextual external
apparatus variorum containing only the relevant variants for
the selected comparison set and providing a locus for reorienting
the edition. This scope on the textual tradition can be further
refined on an intra-textual level. Complete text versions can be
compared, as well as separate text divisions (one of the 33
chapters or the dedication). Where applicable when comparing
different text versions, an entry to a generated apparatus is
provided both at chapter level and at paragraph level.
Edition formats
One end of the delivery spectrum features the dynamic
XHTML visualisation discussed so far. The versatility
of XML equally allows for the generation of a PDF visualisation
of the (different) edition(s), closely resembling a traditional
print view of the textual tradition. A PDF rendering consists of
an orientation version, either as an integral text or as a chapter
sample, possibly compared to any number of comparison
versions, as reflected in an inline contextualised apparatus
variorum. However dynamic this generative edition frame is
(Vanhoutte & Van den Branden, forthcoming), its boundaries
are still present. To cater for this limitation and to enhance
scientific independence, the other extreme of the delivery range
is offered as well: the Morkel system equally allows users to
generate pure XML renderings of the selected comparison
version texts or chapters, containing their parallel-segmented
inline record of the textual variation. These source texts can
then be used in completely different usage scenarios, perhaps
featuring completely different software environments.
Challenges
In short, the Morkel system enables users to generate their
own edition(s) along 3 axes (comparison set (19 text
witnesses and 1 reading text), textual scope (all or 1 of the 34
text divisions), delivery format (3 possible formats)), combining
to 58 different visualisation parameters. This generates the potential for 53.477.376 different views on the text, and
problematises some traditional text theoretical concepts, as well
as the defining role of tools for the electronic editions they
facilitate or constitute. An obvious consequence of this
generative edition paradigm (Vanhoutte & Van den Branden,
forthcoming) is the promotion of each text witness to a
candidate orientation version, instead of the adoption of one
text version as a base text for the edition against which all other
versions are calibrated. Instead, this calibration itself is made
relative by the possibility of restoring each different textual
witness as an autonomous landmark in the textual tradition,
thence allowing a forward or backward look into the tradition.
As a matter of fact, the constituted reading text itself has
become integrated as 'just' a (commented) view on the textual
tradition, against which all variant versions of the text can be
plotted. A dynamic selection of a comparison set not only
transforms the apparatus variorum to a dynamic, contextualised
rendering of the relevant textual variation, but equally promotes
it to a performative instrument for reorienting the edition to
another point in the textual history. Due to the dynamic selection
of comparison sets, the notion of variable classification becomes
relativised. Discerning different types of textual variants
becomes irrelevant: a variant can hold as a spelling variant in
one comparison set but can change classes and become a
semantic variant when compared to another version in the
textual tradition. The search capabilities of the Morkel system
even extend the view on the textual tradition from text level,
by allowing simple search operations inside one text version
(intra-textual), to collection level, by allowing complex search
operations over different text versions (extra-textual). To
conclude, this generative paradigm for electronic scholarly
editions seems to articulate the defining role of the specific
tools for accessing electronic texts more sharply. On its own,
the XML representation of text-critical research is a valuable
record of scientific labour, but it is the specific (generative)
interface which instantiates it as an editorial microcosm by
providing a range of user-driven access methods that enable
dynamic exploration of a textual tradition. The characteristics
and exact nature of this user-driven scholarly edition or
constellation of editions is strongly determined by the
boundaries this generative interface provides, allowing for a
microscopic, telescopic, stereoscopic, or kaleidoscopic view
on the textual tradition.
Bibliography
Juxta . <http://www.patacriticism.org/juxta
/>
McCarty, Willard. "Modeling: A Study in Words and
Meanings." A Companion to the Digital Humanities. Ed. Susan
Schreibman, Ray Siemens and John Unsworth. Oxford:
Blackwell Publishing Ltd, 2004. 254-70.
Sperberg-McQueen, C. M., and Lou Burnard, eds. TEI P4:
Guidelines for Electronic Text Encoding and Interchange,
XML-compatible edition. Oxford: TEI Consortium, 2002. <h
ttp://www.tei-c.org/P4X/>
Van Hulle, Dirk . Textual Awareness: A Genetic Study of Late
Works by Joyce, Proust & Mann. Ann Arbor, MI: University
of Michigan Press, 2004.
Vanhoutte, Edward, and Ron Van den Branden. "Describing,
Transcribing, Encoding, and Editing Modern Correspondence
Material: A Textbase Approach." Computing the Edition.
Toronto: Toronto University Press, Forthcoming.
Versioning Machine . <http://www.v-machine.org
/>