The theory and practice of electronic scholarly editing is in transition. Coming from the early use of the computer in concordance
making, textual analysis, and automatic collation, the techniques and tools to produce computer-assisted printed editions
have been well established and have been proven successful. The advent of SGML (ISO 8879:1986)[1] and the response by the Humanities to the Standard Generalized Markup Language through the creation of the Text Encoding
Initiative (TEI)[2] provided the Humanities scholar in general with a syntax and a methodology for representing text and the features within
text, and the textual critic in particular with guidelines for the transcription of primary textual sources[3] as well as the documentation of textual variation.[4] The renewed interest for hypertext, the improvement of digital imaging technology, the success of the internet, the world
wide use and development of graphical browser software, and the exponential growth of both the speed of computer processors
and digital storage capacity added a next logical step to the recent history of scholarly editing in the computer age: the
distribution of scholarly editions by means of electronic publishing. In the mid-nineties this resulted in a series of principles for and rationales of electronic scholarly editing,[5] and several classifying theories on the nature of electronic editions. Together with all this, new solutions were suggested
for the problem of visualization of textual variation or, as Peter M. W. Robinson puts it: »How does one read a text in 58
different versions?«[6] It may be clear that the simple linking of an electronic text to a traditional apparatus does not provide us with a viable
answer, and adds nothing but a click-experience to its hard-copy counterpart. Editions which mimic traditional editorial and
organizational conventions in an electronic paradigm mark the transition from the scholarly codex edition to the possibilities
created by the computer age.
Electronic editions, however, should essentially be organized differently. With »The Wife of Bath's Prologue on CD-ROM«[7] Robinson suggested a model in which a master text »simply provides a series of pegs on which the manuscript variants may
hang«,[8] thus focusing not on the presentation of an editor's text but on the documentation of the variants in all 58 witnesses at
any one point of the text. In an article with the self-reflecting title »Is There a Text in These Variants?« Robinson argues
that the full presentation of variation opens up new ways of reading Geoffrey Chaucer's »Canterbury Tales« and liberates the
reader from the restricted possibility to look at a text from one distance only.[9] In a recent article, however, Robinson announced his ›conversion‹ from the emphasis on the many texts to the suggestion of
the one text: »Rather than underplay the ›base text‹, or even remove it altogether, we will be doing the reverse: calling
the reader's attention to it, and even making this base text an edited, eclectic, text.« And he concludes: »There will still
be many texts in our publications; but there will also be the one text«.[10]
Provoked by this working model and confronted with the essential difference of an early 20th century Flemish novel we started
looking for a theoretically and practically sound paradigm for our planned electronic edition, back in 1998 when the Royal
Academy of Dutch Language and Literature in Belgium (Koninklijke Academie voor Nederlandse Taal- en Letterkunde, KANTL) commissioned
an electronic edition of the classic Flemish novel »De teleurgang van den Waterhoek« (The decline of the Waterhoek) by Stijn
Streuvels (1871-1969).[11]
The documentary and genetic history of »De teleurgang van den Waterhoek« is an excellent case for textual and genetic criticism.
The extant primary text sources include several authorial versions of the text, a pre-publication and thirteen print editions,
together with a corpus of relevant correspondence. After an incubation period of more than twenty years, Streuvels started
writing his novel and published it in installments in a Dutch literary journal (»De Gids«) in 1927. At the end of the same
year it was published in book form both in Flanders and the Netherlands. The novel tells the story, based on true facts, of
the intrusion of technology - the construction of a bridge over the Scheldt, a river in Flanders (in Belgium) - in a hamlet
called »the Waterhoek«, which causes the local people to revolt because of their belief that the bridge will destroy the typical
and peculiar character of the rural community and will bring about traffic and all kinds of strange people. But as soon as
they realize that the project will open up the region to the new world and that they can earn a living in their own neighbourhood
as construction workers the majority of the Waterhoek people is in favor of this enterprise. Into this plot, the author has
interwoven a love story between Maurice, the engineer supervising the construction works, and Mira, a voluptuous and frank
local beauty. They get married, but the marriage turns out to be a failure and it very soon comes to an end because of Mira's
unfaithfulness.
For the second ›revised‹ edition (published in 1939), which the publisher requested for economic reasons (a shorter book means a cheaper book), Streuvels drastically reduced his novel by more than 25 percent. The author chose to cross out, amongst others, a
lot of passages which had been experienced by catholic Flanders of the 1920s and 1930s to be of a too erotic nature. Moreover,
the end of the story in the second edition is radically different from the first edition. In the latter version of the text
Mira can stay with Maurice's mother for a certain time after the divorce, while Maurice takes up a job in Congo. The author
suggests that the divorced couple can eventually reunite and there is an allusion to an unborn child as well.
In a letter of March 18th 1937, Streuvels' German publisher Adolf Spemann, draws the author's attention to the inconsistency
of the suggested possibility of a reunion; according to him, both characters are too different.[12] In Streuvels' answer of March 21st he points out that the characters of Mira and Maurice were based on real people and that
back in 1927 he still had to be careful. But now, after more than a decade, he could more easily modify the story. Indeed,
in the German translation of 1938[13] Mira's stay with Maurice's mother is no longer mentioned, nor is there an allusion to an unborn child. On the basis of the
German novel a reunion of the divorced couple becomes very unlikely. In the second Dutch edition the divorce is final. Mira
leaves the Waterhoek without saying where she's going. Up to the publication of the 13th print edition of the novel in 1987
this heavily revised second edition had been the basis for 11 reprints.
In our analysis of the three traditional national schools of textual theory - German, French, and Anglo-American editorial
theory - we did not come across a single ›off the shelf-out of the box‹ framework for what we wanted to do, that is to produce
an electronic edition which critically restores the text of the first print edition and provides the user with a method to
reconstruct the genetic history of the novel by means of versioning[14] rather than the inclusion of a conventional apparatus. Our edition of »De teleurgang van den Waterhoek« includes elements
of documentary, historical-critical, diplomatic, facsimile, study, and reading editions but is neither of them exclusively.
We believe that our electronic edition is in fact a new kind of scholarly study edition, in that it combines elements of the
three different ›editorial schools‹.
Indeed, in line with the developments in the Anglo-American editorial world, the basis of our edition is electronic, combining
an archive of six versions of the text with a critical edition of the first print in book form. As to the constitution of
the text in this edition we follow the German tradition in that only justified corrections of manifest mistakes are allowed
in the edited text. The fact that the digital facsimiles of the complete manuscript, the copies of the journal version and
of the first edition in which the author added and crossed out, plus 71 letters from Streuvels' correspondence, are included
in the edition, is in compliance with the requirements of genetic criticism. Our edition also contains a glossary, some articles
on the genesis of the novel and on the constitution of the edited text, and a list of the different manuscript sources and
printed versions. Together with the electronic edition on CD-ROM[15] we also published a reading edition in book form[16] which comprises the edited text with word explanation and the scholarly articles mentioned.
A couple of decades ago Anglo-American editorial practice definitively moved away from the Greg-Bowers-Tanselle school of
editing which was in support of the copy-text theory and which was in its rigorous form firmly rejected by the German editorial
school because of their production of an eclectic ›definitive‹ text. Meanwhile, concepts such as ›textual instability‹, ›processuality‹,
and ›versioning‹ have formed the core of the debate. A work develops from version to version and it has become the editor's
task to render this evolution as clearly as possible. These concepts go - at least theoretically - quite a bit in the direction
of German editorial theory, but contradistinctively their application to editorial practice does not necessarily result in
extensive editions in book form. Anglo-American scholarly editing has resolutely taken the road of electronic archives and
hypereditions.[17]
In a review article on our edition the Dutch editor Peter De Bruijn states that the Rossetti project is actually »nothing
less and nothing more than what Kanzog back in 1970 called an archive edition«.[18] In »Prolegomena zu einer historisch-kritischen Ausgabe der Werke Heinrich von Kleists« Klaus Kanzog differentiates between
four types of editions: the archive edition, the historical-critical edition, the study edition, and the reading edition.[19] The function of the archive edition is to amass all documentary source material such as manuscripts and authorized print
editions, and to provide a record for them by way of diplomatic transcripts. In order for the user of the archive to have
easy access to the complete genetic process of the work under study the archive must present the material in an orderly system
which - since documentation of the complete collection is its primary purpose - treats all documentary source material equivalently.
The archive edition, according to Kanzog's definition, should explicitly not contain any commentary section and should refrain
from any critical constitution of texts: it is intended only to present the ›facts‹ of the genetic history of a work.
The historical-critical edition is concerned with researching the history of the text - mostly in the form of a detailed apparatus
which takes all authorized versions, manuscripts as well as printed versions, into account. On the basis of one of the extant
(authorized) versions this type of edition presents an ›edited text‹ (edierter Text) in which emendations are allowed, as far as they are restricted to the removal of ›textual faults‹ (Textfehler) and with the obligation to justify them in the notes. Detailed chapters on the genesis and textual history and on the constitution
of the edited text have to be included, as well as - depending on the nature of the text - a certain amount of commentary
and of word explanation. The study edition - the third type of the German school - gives a reliable ›edited text‹ in the same
way as the historical-critical edition does, but without the presence of a detailed textual history in a lengthy apparatus
of variants. Chapters on text history and text constitution are necessary, as well as commentary on the text. The last type,
the reading edition meant for the general public, also gives a reliable ›edited text‹ with some commentary and word explanation
if necessary. Chapters or statements on the history of the text and on text constitution can be reduced to a minimum but must
be present in order for the edition to be of a scholarly quality.[20]
As said before in our electronic edition of »De teleurgang van den Waterhoek« we combine a critical text with an archive of
six versions of the text. This construction - call it a compromise between critical and documentary editing - has been suggested
by G. Thomas Tanselle in his instructive article from 1996, titled »Textual Instability and Editorial Idealism«:
Writers on hypertext editions frequently think of the goal as a kind of »archive« limited to documentary texts. The dissemination
in this way of manuscript and printed texts surviving from the past is extremely valuable, obviously, just as the publication
of documentary editions has always been valuable. But hypertext editions offer great advantages for the presentation of critical
texts as well, and one can argue that the potential of the electronic form is not being very fully exploited unless editors'
critical reconstructions are included along with documentary texts.[21]
We have taken the genetic orientation of our electronic edition from the French school of critique génétique as institutionalized in ITEM (Institut des Textes et Manuscrits Modernes) in Paris. Genetic criticism mainly deals with the
genesis of the work, the writing ›process‹, and not so much with the ultimate ›product‹, the printed text. Genetic criticism
of a text can result in a genetic edition, but this is not necessarily the case. In order to reveal the author's working method
and his poetical views, the emphasis is placed on the study of his notebooks, manuscripts, etc. (the avant-texte). Eventually this approach can lead to a better interpretation of his work.[22]
As genetic criticism is mainly interested in the writing process the aim of a genetic edition is not to establish a fixed
text but to describe the (genetic) development of the text (e.g. the edition of Flaubert's »Hérodias« 1991/95[23]). It is quite clear that the computer can be a most efficient tool in making the genetic edition possible even for lengthy
texts: »Since the concept of electronic hypertext allows every manuscript to be rendered in extenso without making the edition
unmanageable or even illegible, it is an adequate medium to present the avant-texte«.[24]
In the Archive and Museum for Flemish Cultural Life in Antwerp (Archief en Museum voor het Vlaamse Cultuurleven, now AMVC-Letterenhuis),
which is the central repository for literary manuscript collections in Flanders, several primary sources of »De teleurgang
van den Waterhoek« are preserved. The following ones are inserted into the electronic edition:
- - a complete holograph manuscript from 1927 (S 935 / H18)
- - a corrected copy of the prepublication in the journal »De Gids«, functioning as author manuscript (and printer's copy) for
the first edition in book form (1927) (S 935 / H17)
- - a corrected copy of the first edition, functioning as author manuscript (and printer's copy) for the second edition of 1939
(S 935 / H24)
Apart from the inclusion of 920 digital facsimiles of these three sources the electronic edition encompasses the complete
full texts of the prepublication in the journal (1927), and the critical texts of the first (1927) and the second (1939) edition
in which all emendations are explicitly documented by making use of TEI compliant SGML.[25]
All of these sources can be consulted on their own and in combination with the critically constituted text of the first print
edition. Taking this edited text as our orientation text, we visualize the textual history by applying a linkemic approach to textual variation.[26] A linkeme is the smallest unit of linking in a given paradigm. In our edition the linkeme of textual variation is the paragraph. By
making use of hypertext functionality (the edition includes 10.000 hyperlinks), the user of the edition can call up each and
every corresponding paragraph of every included version (figure 1). A HyTime-tool (ISO 10744:1997)[27] which is included in the edition enables the user to add personal annotations to the text, and to make cross-references from
one piece of information, be it a spot on a digital facsimile or a string of text, to another one, and thus between the different
versions. We believe this opens up promising perspectives for the use of electronic editions in the (university) classroom.
We will illustrate this by the following example.
Fig. 1. The simultaneous presentation of variant paragraphs from six documentary sources on the screen. Marcel de Smedt Edward Vanhoutte Open figure in new window [272.3KB]
At the end of the novel, after the construction of the bridge is finished, a large feast is being held in the Waterhoek. The
author remarks that the people of the Waterhoek do not realize that along with the new outlook of the place its typical character
and nature have been lost forever. This topic is in fact one of the constants in Streuvels' work; neither the author Streuvels
nor the person Frank Lateur liked changes in scenery and society.
The remark in question was lacking in the prepublication. It was added by the author to the copy of the »De Gids«-version,
which served as author manuscript for the first edition. In the electronic-critical edition of »De teleurgang van den Waterhoek«
it is possible to make an annotation to the paragraph in question, and moreover, to make a cross-reference directly to the
corrected copy of the prepublication which brings the addition in the prepublication on the screen (figure 2).
Fig. 2. The annotated reading text and the corresponding facsimile of the prepublication with addition. Marcel de Smedt Edward Vanhoutte Open figure in new window [143.8KB]
A couple of paragraphs further down in the text Mira is leaving the Waterhoek and Maurice. Whereas the first print edition
suggests that there is a chance of a reunion between Mira and Maurice, the second print edition is clear about a final divorce:
Mira takes the train and leaves without saying where she is going to. Therefore, Streuvels had to cross these passages in
the printer's copy of the first edition in which there was an allusion to Mira's staying with Maurice's mother and even to
the possibility of a pregnancy. A teacher can for instance annotate this passage, and provide hyperlinks to the corresponding
places on the digital facsimiles, so that the student can study the genesis of the end of the novel in great detail (figures 3 and 4).
Fig. 3. The changed ending of the novel. Marcel de Smedt Edward Vanhoutte Open figure in new window [124.0KB] Fig. 4. The deletions at the end of the corrected prepublication. Marcel de Smedt Edward Vanhoutte Open figure in new window [146.8KB]
The user specific annotations, hyperlinks (and bookmarks) are stored in a separate ›web‹ file on the user's hard disk and
can thus be exchanged between users via email. By opening another person's ›web‹, it is possible to follow his annotated reading
path. This way, the electronic edition corroborates the case for collaboration in literary studies.
NOTES
1]
ISO 8879:1986: Information processing - Text and office systems - Standard Generalized Markup Language (SGML). [Geneva]: International
Organization for Standardization 1986. Cf. Charles F. Goldfarb: The SGML Handbook. Oxford: Clarendon Press 1990.
2]
Guidelines for Electronic Text Encoding and Interchange (TEI P3). Ed. by C. M. Sperberg-McQueen and Lou Burnard. Chicago,
Oxford: Association for Computers and the Humanities (ACH) / Association for Computational Linguistics (ACL) / Association
for Literary and Linguistic Computing (ALLC) / Text Encoding Initiative 1994. Cf. also the paper of Tone Merete Bruvik in this issue of »Sichtungen«.
3]
E.g. Peter M. W. Robinson: The Transcription of Primary Textual Sources Using SGML. Oxford: Office for Humanities Communication
1994.
4]
Viz. respectively chapters 18 and 19 of Guidelines for Electronic Text Encoding and Interchange (note 2).
5]
E.g. David L. Gants: Toward a Rationale of Electronic Textual Criticism. Paper. In: Association for Literary and Linguistic
Computing / Association for Computers in the Humanities Conference (ALLC / ACH) Paris, 19-23 April 1999, http://parallel.park.uga.edu/dgants/ach94.html [7 June 2002]; Jerome McGann: The Rationale of HyperText. In: TEXT. An Interdisciplinary Annual of Textual Studies, 9 (1996),
pp. 11-32; Peter Shillingsburg: General Principles for Electronic Scholarly Editions (1993) [Distributed at the Modern Language
Association (MLA) meeting in Toronto, December 1993], http://sunsite.berkeley.edu/MLA/principles.html [7 June 2002]; Peter Shillingsburg: Principles for Electronic Archives, Scholarly Editions, and Tutorials. In: The Literary
Text in the Digital Age. Ed. by Richard J. Finneran. Ann Arbor: The University of Michigan Press 1996, pp. 23-35.
6]
Peter M. W. Robinson: ... but what kind of electronic editions should we be making? Abstract. In: Association for Literary
and Linguistic Computing / Association for Computers in the Humanities Conference (ALLC / ACH), Bergen, 22-29 June 1996, http://gonzo.hd.uib.no/allc-ach96/Panels/Finneran/robinson.html [7 June 2002].
7]
Geoffrey Chaucer: The Wife of Bath's Prologue on CD-ROM. Ed. by Peter M. W. Robinson. Cambridge: Cambridge University Press
1996.
8]
Peter M. W. Robinson: The Collation and Textual Criticism of Icelandic Manuscripts (l): Collation. In: Literary and Linguistic
Computing, 4 (1989), issue 2, pp. 99-105, here p. 102.
9]
Peter M. W. Robinson: Is There a Text in These Variants? In: The Literary Text in the Digital Age. Ed by Richard J. Finneran.
Ann Arbor: The University of Michigan Press 1996, pp. 99-115.
10]
Peter M. W. Robinson: The One and the Many Texts. In: Making Texts for the Next Century. Ed. by Peter M. W. Robinson and Hans
W. Gabler (= Literary and Linguistic Computing 15 [2000], issue 1), pp. 5-14, here p. 7.
11]
Stijn Streuvels is a pseudonym for Frank Lateur.
12]
Letter held in Archief en Museum voor het Vlaamse Cultuurleven (AMVC, Archive and Museum for the Flemish Cultural Life), Antwerp,
S 935 / B2.
13]
Stijn Streuvels: Die große Brücke. Stuttgart: J. Engelhorns. Nachf. Adolf Spemann 1938.
14]
Donald H. Reiman: Romantic Texts and Contexts. Columbia: University of Missouri Press 1987 (esp. Chapter 10: »›Versioning‹:
The Presentation of Multiple Texts«, pp. 67-180).
15]
Stijn Streuvels: De teleurgang van den Waterhoek. Elektronisch-kritische editie / electronic-critical edition door Marcel
De Smedt en Edward Vanhoutte. Gent: KANTL / Amsterdam: Amsterdam University Press 2000 (CD-ROM).
16]
Stijn Streuvels: De teleurgang van den Waterhoek. Tekstkritische editie door Marcel De Smedt en Edward Vanhoutte. Antwerpen:
Manteau 1999.
17]
In this yearbook it's quite superfluous to remind of McGann's »The Rationale of HyperText« (note 5) and of his Rossetti Hypermedia
Archive (The Complete Writings and Pictures of Dante Gabriel Rossetti. A Hypermedia Research Archive, http://www.iath.virginia.edu/rossetti/ [7 June 2002]).
18]
Peter De Bruijn: ›De eerste brug volgens het nieuwe systeem‹ oftewel: een nieuwe evolutie in de editiewetenschap? Naar aanleiding
van de ›elektronisch-kritische editie‹ van Stijn Streuvels' »De teleurgang van den Waterhoek«. In: TNTL Tijdschrift voor Nederlandse
Taal- en Letterkunde 117 (2001), issue 1, pp. 59-72. Our translation.
19]
Klaus Kanzog: Prolegomena zu einer historisch-kritischen Ausgabe der Werke Heinrich von Kleists. Theorie und Praxis einer
modernen Klassiker-Edition. München: Hanser 1970, pp. 15-44.
20]
On the basis of the German editorial principles, voluminous editions of the works and letters of authors such as Johann Wolfgang
Goethe, Friedrich Schiller, Friedrich Hölderlin, Heinrich Heine, Friedrich Klopstock, Conrad Ferdinand Meyer and Annette von
Droste-Hülshoff have been published or are in progress. The Dutch Constantijn Huygens Instituut (The Hague) basically follows
German editorial practice. Editions of e.g. the 19th century novel Max Havelaar or of the poetical works of Jan Hendrik Leopold,
Martinus Nijhoff and Gerrit Achterberg - just to mention these - are illustrations of this approach.
21]
G. Thomas Tanselle: Textual Instability and Editorial Idealism. In: Studies in Bibliography 49 (1996), pp. 1-60, here p. 54.
22]
In an instructive article in Yale French Studies, Pierre-Marc de Biasi makes a distinction between the »avant-texte stage«
and »manuscript genetics« on the one hand, and the »text stage« and »textual genetics« on the other. Changes in the former
stage take place in the private sphere of notebooks and manuscripts, whereas changes in the latter are public: the reality
of the published book cannot be denied. In the »avant-texte stage«, de Biasi distinguishes the »exogenetics« (i.e. all kinds
of notes and paralipomena) and the »endogenetics«, with sketches of the work, schemes of chapters, etc. on the one hand, and
(the) rough draft(s) on the other; the latter also called the »compositional« or »textualizing« rough draft(s), the draft(s)
in which the composition of the text has taken place. Pierre-Marc de Biasi: What is a Literary Draft? Toward a Functional
Typology of Genetic Documentation. In: Drafts. Ed. by Michel Contat, Denis Hollier and Jacques Neefs (= Yale French Studies,
89 [1996]), pp. 26-58.
23]
Gustave Flaubert: Corpus Flaubertianum. [Établi par] Giovanni Bonaccorso et collaborateurs. Vol. 2: Hérodias. Pt. 1: Paris:
Nizet 1991; pt. 2: Messina: Sicania 1995 (= Les textes français).
24]
Dirk van Hulle: Textual Awareness. A Genetic Approach to the Late Works of James Joyce, Marcel Proust, and Thomas Mann. Antwerp:
University of Antwerp 1999 (Doctoral Dissertation), p. 379.
25]
Whereas the critically constituted texts of the first and the second edition document editorial corrections by making use
of the <corr> tag with a sic attribute, the text of the prepublication marks corruptions by a <sic> tag with a corr attribute, thus documenting the same elements but applying a different approach.
26]
Edward Vanhoutte: A Linkemic Approach to Textual Variation. Theory and Practice of the Electronic-Critical Edition of Stijn
Streuvels' »De teleurgang van den Waterhoek«. In: Human IT. Tidskrift for Studier av IT ur ett Humanvetenskapligt Perspektiv,
1 (2000), pp. 103-138.
27]
ISO 10744:1997: Information technology - Hypermedia / Time-based Structuring Language (HyTime), 2nd edition. [Geneva]: International
Organization for Standardization 1997.
|