Almqvist on the Internet

av Johan Svedjedal

The story of how the edition of Carl Jonas Love Almqvist's collected works currently in preparation ever got started is presumably typical for many contemporary publishing projects. It begins as a mirage before taking shape as a dream and finally as a detailed campaign plan. But when the march begins it takes the form at times of a beggars' pilgrimage. The affair is not made any easier by the fact that all surrounding landmarks have during the passage of time been demolished. The solid old buildings from the turn of the century, rendered and with deep embrasures, have been levelled to the ground, and hi-tech houses have been erected where they once stood, equipped with lighting controlled by body temperature. Nearby stands the editor wondering what has actually happened. Where has his plan of campaign gone?

Doubtless the plan remains unchanged in the main. We live in the age of Internet and computerization, but as yet the conditions under which texts are edited and published have not changed drastically enough to stop one thinking of supplying the texts in book form. Despite the new technical possibilities the book remains the chief medium for publishing critically edited texts. What has happened is that computer techniques have come to exist which are useful for the preparation of such texts - both before and after publication.

In what follows the issue of Carl Jonas Love Almqvist's "Collected Works" (Samlade Verk) now being published will serve as an example of how computer techniques can be used in critical textual editing.¹ We are concerned in the first place with Internet as an outlet for publishing, but the growth and taking shape of the project of a complete edition is described as a background to this, along with discussion of the place of the digital version amongst Swedish digital publications. To publish such an account may seem audacious - development in this area is so rapid that decisions taken and principles established often have to be torn up or revised - but for the debate among experts it is on the other hand important that statements about the state of progress be published now and again. Such descriptions, moreover, become a part of the historical record of the developing significance of the digital revolution for the world of books.

It should be pointed out that the term "digital publishing" is used in what follows as a description of a particular type of "electronic publishing". A digital text is not stored as a picture but can be searched digitally; an electronic text, on the other hand, can be a picture, e.g. of handwriting or of a page of a book. It is not possible to search after or process the information given in such pictures using the tools for textual editing or searching found in word-processing programmes, for example.

(Åter till början av artikeln)

Carl Jonas Love Almqvist (1793-1866) is one of the most important Swedish writers from the nineteenth century. He was one of the leading fiction writers, as well as a controversial redical in the religious and political debate. In his youth, he was known as a romantic; in the 1830’s he turned to realism and journalism; and in the 1840’s, he published several sensationalist novels, written for both political and economical reasons. His aestheticism, radicalism and utopianism is a continuing influence in the Swedish literature.

No complete critical edition of Almqvist's collected works has ever appeared. The most comprehensive edition up to now is "Collected Writings" (Samlade skrifter), published during the 1920s and 1930s by Bonniers. The figurehead for this edition, as its official chief editor, was Fredrik Böök, Sweden's most illustrious literary researcher at that time. But the real driving forces in the work were his doctoral students Olle Holmberg and Algot Werin. Both defended their dissertations on Almqvist at the beginning of the 1920s, and their work on the edition formed part of their research activities. The edition was planned so as to consist of 32 volumes, containing in principle most of Almqvist, but not all. (Thus there were some exclusions from the various series of Törnrosens bok.) It was carefully modernized, mostly in fact through modernization of the spelling, but also through making certain loan-words Swedish and sometimes through normalizing the structure of Almqvist's sentences.

The issuing of the collected writings (Samlade skrifter) proceeded quickly at the beginning of the 1920s, but for various reasons it soon came to a stop. One reason was that Holmberg and Werin developed other research interests, another that the sales were so limited that the publishers lost interest. Samlade skrifter was finally broken off after 21 volumes. And despite the fact that several other individual volumes of the works of Almqvist have been published since then, yet far from all of his works have been published in satisfactorily scholarly editions.

Most of Almqvist's literary remains have been published, all the same. The reason is simple: not so very many of his manuscripts have been preserved. The manuscripts of the works which were printed during his lifetime seem almost without exception to have been destroyed. Certainly there are interesting autographs in the form of letters, drafts and manuscripts which remained unprinted during his lifetime. But they amount to barely a tenth part of the work he left behind him. In the other cases the printed versions of his work offer us the earliest preserved texts.

A complete edition of Almqvist's "Collected Works" has long been a dream of concerned researchers, but it took until the beginning of the 1980s for it to become a serious proposition. This happened when the relatively newly founded Almqvist society arranged a symposium for the 29th of November, 1985. This symposium resulted in the setting up of a working group to investigate the prospects for such an edition. The group's work bore fruit in the form of a report and in the gradual setting up of an editorial staff. This consists now of Bertil Romberg as editor in chief with Lars Burman and Johan Svedjedal as assistant editors.²

Samlade Verk will consist of 51 volumes. The first volume came out in 1993 (Det går an [It Can Be Done] and Hvarför reser du? [Why Do You Travel?] in the 1838 versions). Four volumes have appeared so far, and publishing is planned at the rate of two or three volumes a year.

In retrospect it is easy to see how much of the work with Samlade verk has been affected by the rapid rate of technological change in book production since the beginning of the 80s. During this time there has been something of a revolution in the Swedish publishing industry. It is worth remembering how one went about things until just a few years ago - what the working day looked like when a scanner was at most something one saw in American TV serials and when Internet was not much more than a gleam in the eyes of research librarians.

The main aim of the Almqvist edition was to issue a new, modernized edition, something in keeping, that is, with the Böök edition. The idea was that such an edition would reach readers more effectively than one retaining the original spelling. In the background there was also the expectation of certain state guarantees, financial pledges big enough to induce some major publisher to take responsibility for the edition - rather as Bonniers had taken up the Böök edition, that is to say. The interest of both state and publishing world proved weak, however. In fact one could have stored pieces of ice in their commentaries on our suggestions.

For various reasons the idea was mooted that the Swedish Literary Society (Svenska Vitterhetssamfundet) could sponsor the edition. This decision proved felicitous inasmuch as it afforded the publishing project a milieu suited to textual critical analysis, freed from the pressure of time from cultural and political sources. In line with the other publications of the Vitterhetssamfund anything other than the original spelling for the edition was unthinkable. There were of course also weighty linguistic historical arguments for this: such an edition would be of very great interest for philology.

We were in this situation around the year 1990 and the publishing industry was humming and buzzing with computers. The use of disks for the type-setting of books was more or less routine, but scanners were still not a realistic alternative. How could one be sure of accuracy of reproduction? One method which the larger publishing-houses we talked to were using at the close of the eighties was that of recourse to cheap labour somewhere in Asia. Two people - who should preferably be ignorant of Swedish - would have to type-set the same text, and the computer files would then be compared against one another. Where they differed one would look for a misprint. We never seriously considered taking this course (which in the publishing trade was called "Taiwan-setting"). Instead the first volume was type-set through a student's feeding the text into computers.

When the first volume came out in the autumn of 1993 this method was already obsolete. The texts of the following volumes have all been scanned in. We have bought this service, including preliminary proof-reading, and we see no cause to take over the scanning work ourselves.³ In fact the editors have adhered strictly to the principle of buying the various types of computer services rather than working with one's own staff of computer experts. We have also tried to have recourse to acknowledged experts in the field rather than trying to get hold of these services as cheaply as possible.

The greater part of the preparation for publication is carried out by one of Sweden's more expert electronic publishers (Krister Gidlund, Gidlunds förlag, Hedemora). The method followed is that after scanning in the redactor of the volume concerned gets a disk-version of the text to work with, right from the beginning. After some time he or she delivers a disk to the publishers who then produce the print-ready version for proof-reading and subsequent printing. The publishers also produce from this disk a modernized issue of selections, called Skrifter (writings), for the bookshops. This has shorter prefaces and a much reduced critical apparatus. The modernization for Skrifter is done by Gidlund in cooperation with the Swedish Literary Society. Through this version many of Almqvist's works can reach libraries and bookshops in a manner scarcely open to Samlade Verk.

The editing work itself will here only be described extremely briefly. The task posed by Almqvists Samlade Verk is as simple as possible with regard to the quantity of materials involved, due to the lack of original manuscripts, but there are other problems to solve when it comes to the critical analysis of the received texts which each each volume demands. Like many writers of that time Almqvist used often to alter his work after printing (and sometimes during printing). It was not uncommon for him to tear out a page from the printed issue (the "cancelland") and replace it with a newly printed page ("cancellance") with altered text.⁴ Examples of variants, in other words, are to be found in the first impression of the original editions. It is a great labour for the editors to have to find and compare these examples. It makes great demands upon their competence in analytical bibliography, in a way that in a Swedish context implies that this publication explores new methods. So far as is known no textual critical edition in Sweden has ever before taken such perspectives into consideration. At the same time the fact that such regard is paid to these variant examples implies that Samlade Verk is a critical edition of the classical kind, an edition that is not only printed from a text but in which also, to use the bibliographer Rolf E. Du Rietz's fitting terms, a realtext (real text) is set forth which reproduces a work's idealtext (ideal text).⁵

As will have become clear the basic idea right from the beginning was partly to make as much use as possible of computers, partly to buy services to the greatest possible extent. When, therefore, the idea arose of also publishing the edition digitally the first step was to look for someone who had the right kind of competence. But it was also important to tie our publishing venture to some reputable institution - partly so that potential users would more easily come upon it, partly so that the digital version would be marked as being of high quality.

Our choice to begin with was between publishing on CD-ROM and publishing on Internet. To publish Samlade Verk on the Internet seemed for various reasons much more attractive. From the point of view of marketing one could scarcely launch a CD-ROM version before a large proportion of the volumes were published, whereas an Internet version could be made available at about the same rate as the books came out. In that way it could reach the users quickly - and be of great and continuous help to the editors while they were working on new volumes. Another reason for preferring Internet to CD-ROM was that we thereby avoided tying ourselves to one technical format already from the beginning. We have chosen the most popular and most flexible format, HTML coding, which makes Samlade Verk easily available on World Wide Webs hypertext milieu - and thereby able to be linked to other web sites and hypertexts.⁶

When we came to the choice of the place of publication for the Internet version we established cooperation with Språkdata at the Institute for the Swedish Language in Gothenburg. In the data base of Språkdata, viz. "Språkbanken", there are a number of texts already assiduously used by researchers: e.g. a concordance to the Swedish Academy's dictionary, concordances to Strindberg's Samlade Verk and to his letters, several medieval texts (inter alia St. Bridget's "Revelations").⁷

The choice of working with Språkbanken was soon amplified to a general decision within the Swedish Literary Society, viz. that publishing of the Society's new textual critical editions should in future take place simultaneously in book and digital form. Almqvist's Samlade Verk may thus be seen as a pilot project, but also other technical solutions can and should be used for the digital versions which the Swedish Literary Society is publishing of other texts. Available so far is Andreas Arvidi's Manuductio Ad Poesin Svecanam, published by Mats Malm (1996, in book form).⁸

(Åter till början av artikeln)

In practice the work on the digital publishing of Almqvist's Samlade Verk begins with the publishers sending a copy of the disk with the files containing the print-ready version to Språkdata. There the files are converted to the HTML format (which will make it easier in the future to attach them to the TEI standard [Text Encoding Initiative]), at the same time as explanations of individual words and textual critical commentaries are subjoined to the literary texts.⁹ The final product consists of hypertexts accessible via Internet. At present four volumes are available and during the coming years we expect to make two to four volumes available per year, i.e. to follow the publishing rate of the book version.

When we published the first digital works on Internet, Det går an and Hvarför reser du?, there were no critically edited Swedish hypertexts on the Net. Up to now we are still alone in the area. So far Strindberg's Samlade Verk still only exists on the Internet in the form of a concordance with restricted accessibility. It is extremely useful, but is something other than a fully fledged hypertext as regards its interest for readers and historians of literature.¹⁰

In other languages there are many well produced hypertexts on Internet. A large number are easily accessible via "Books on-Line", "Project Bartleby" or via Bibliotektjänst's link-listing "Internetkontakt" and in a number of other places. As an example of a successful publication one can point to the enjoyable and useful digital hypertext editions of Jane Austen's novels. Amounting almost to a digital hypertext library we have the "The Victorian Web", a comprehensive contribution to cultural history.¹¹

It should be emphasised, all the same, that most literary classics that are published on Internet are given out without any notable textual critical ambitions (even if some of them should build upon earlier textual critical editions in book form). The clearest example is the big Gutenberg project, which has made numerous texts available on Internet, but which is carried out with what Peter Shillingsburg calls "abysmal ignorance of the textual condition" - the texts are so unreliable that the project's leaves us with "a textual junk-yard".¹²

In Sweden several literary classics are available on Internet, also those without textual critical ambitions. One example is the Karin Boye Society's web site, containing some of her writings and material concerning it (handwritten manuscripts, essays, etc.).¹³ Another is Selma Lagerlöf's oeuvre, which has been scanned in by Autotext Ltd.¹⁴ A third example is the comprehensive Projekt Runeberg, started in December 1992 and based in Linköping.¹⁵ Here a wide selection of Swedish authors and of works (ninety so far) by such authors is presented, often in what are from the point of view of word-processing elegant and ingenious arrangements - but not textually reliable in the way that research and teaching demands. The editors very seldom state which version they have used, but in general the facsimiles seem to be based upon pocket or other recent editions, with all their inevitable textual shortcomings - add to which that a whole new batch of errors has come about or that changes have been made at type-setting or scanning in. Projekt Runeberg is a contribution which promotes literature and of which the participants have every reason to be proud, but for teaching or research these texts are scarcely usable.¹⁶ The problems with Projekt Runeberg, therefore, are the same as those of its international predecessor, Project Gutenberg.

In such a situation it is still more important that the scholarly community should offer good digital texts, i.e. texts without errors and unconscious distortions. One of the objectives of the digital Almqvist project is to contribute to the development of standards of working method in this area. For this reason the choice of forms for working on the project has up to now been kept very much open.

(Åter till början av artikeln)

Thomas Tanselle has persuasively argued that computerization and digital publishing make no difference to the possibility of bibliographically describing how texts are transmitted:

Computerization is simply the latest chapter in the long story of facilitating the reproduction and alteration of texts; what remains constant is the inseparability of recorded language from the technology that produced it and makes it accessible.¹⁷

From this starting-point one can try to describe in current textual critical terms the production of Almqvist's Samlade Verk.

The original edition of Almqvist's works functions generally as copy-text for the critical edition. And it functions simultaneously as the printer's copy. But what is really the critical edition? In practice it consists of the redactionally processed computer files upon which the book version builds. These function as a type-setting of the text and thereby constitute in my understanding the edition in the original bibliographical sense. This edition is then produced in various versions, carried by various media. One product of the computer files is the printed issue of the book, the individual volume of Samlade Verk which goes into what I want to call the book version. Another form of disclosure is the Internet publication, the version which one sees on the screen and which is stored on a server. The Internet version should be called a digital version - other versions can of course be produced in the future, e.g. on CD-ROM or using other forms.

One can find unwelcome discrepancies between these two versions (book, digital). Small distortions occur, for example, at transmission of the text from computer file to book and from computer file to digital publication -we have had some problems with accidentals such as italicisings and dashes, but have not so far encountered any discrepancies where substantives (units bearing meaningful content) are concerned.

It is all the same essential to free oneself from the notion that the book version and the digital version must be exactly alike textually. In certain cases one quite simply does not want the Internet version to agree with the book version, since misprints and other oddities in the book version are discovered during re-formatting and linking. Therefore there is a special page whereon whatever corrections have been taken into the Internet version are recorded.

One should, in my opinion, see it as in principle inevitable and desirable that the book and the Internet publication do not agree perfectly. Why should a digital editor refrain from exploiting the limitless opportunity for textual correction which he or she enjoys? For a digital version is open and undetermined in a way that does not apply to the printed one, something which Jerome McGann, for example, has underscored in an essay on the critical editing possibilities where digital hypertexts are concerned.¹⁸

Will the Internet version become a competitor to the book version of Samlade Verk? Do people want rather to read texts from the screen instead of the book-page or do they store them on disk instead of buying the book? Certainly there is reason to ask these questions. But - "the town has no need to be nervous."

My experience is that a digital version functions excellently as an auxiliary source of reference, but never as an alternative. It is unbeatable for searching for or localising quotations, for various kinds of linguistic analysis, and so on. But it can never be especially attractive for anyone who quite simply wants to read the work. The fear that digital editions will replace books is exaggerated.

And at all events we must accept that the publishing of books can never be an end in itself. Books are documents which carry texts, nothing else. If other documents can carry the texts just as well or better then there is no reason to hold fast to books. But the digital publications which I have seen hitherto on Internet are no serious threats to books. The reasons are those which are always used in defence of books (digital texts are not portable, do not allow of marginal notation and are difficult to skim through), while as texts for reading the digital publications are still far from as technically advanced as are books.

Eventually, nonetheless, computer technology will probably influence the conditions for the actual printing of books. If the technique of print on demand keeps to what it promises then specialized academic literature will within a few years be printed in several small editions, rather than in a large initial edition intended to suffice for several decades. Should such a technology become generally accepted then textual dissemination will have to be radically rethought. In future the book could become a part of the publishing process, rather than its aim.

Such futuristic perspectives do not reduce the importance of working with digital publication. For the aim when publishing critical texts is in the last analysis to guarantee the text as authentically as possible, to comment on it and to see that it reaches as many readers as possible, either directly (through reading) or indirectly (as printer's copy for later issues). And nowhere does one have such a large potential public as on Internet.

It seems appropriate to offer Almqvist's texts in this way. For he dreamed, after all, of reaching a mass public - he wrote serials for popular magazines, worked as a journalist and made his contribution to the great library of novels. Part of the personality of the esoteric Almqvist was positively inclined to technology, which leads me gladly to imagine that he would have approved of Internet. And with that I don't intend to repeat the old cliché that the Romantics would have used Internet to create Gesamtkunstwerke, but rather that Almqvist wanted to make his texts available to as many people as possible. An author who lauded the democratizing influence of the art of book-printing as enthusiastically as Almqvist did ("The art of printing books is the prime means of universal education", he wrote in 1839) would scarcely have had any objection to being published on Internet.¹⁹ For what is a printing-press in our time? In principle it is every writer who is connected to a computer joined to Internet.

(Åter till början av artikeln)

As the major international debate about textual criticism and digital publication in recent years has made quite clear, there is scarcely any limit to the possibilities open to the publisher of digital texts. The storage capacity of computers makes it tempting to publish all the versions of any given work on Internet (and thereby perhaps transform the editorial task into a do-it-yourself shop), as well as the varieties of material relevant to the work (articles, commentaries, illustrations). In practice this changes the digital version into an archive or library of multimedial dimensions.²⁰ One possibility, attractive to many people, is that the traditional edition, with an established text plus an apparatus of variant readings from the other versions, be replaced by a digital layer upon layer edition. Here the material is structured so that the computer generates the various known versions of a work from the author's first draft, via the first published version through to the latest version as revised by the author. The user himself chooses which version he or she wants to read. Such a way of working, as Charles L. Ross has pointed out, spells the demise of the critical edition as we know it.²¹ More specifically it can entail that the kernel itself of editorial work (preferring one reading to another and reconstructing lost readings from surviving material) can become blurred and fall out of sight.²² Instead of the ideal text the reader is offered a series of real texts. This is the latest twist in the sociologically motivated questioning of the intentionalistic model of textual editing which derives from writers such as Jerome McGann and D.F. MacKenzie.²³

The many and swift linkings of which the digital hypertext is capable allows the digital critical edition to point beyond itself in a different way to the book edition. Ultimately the development here is analogous to trends within postmodernism and post-structuralism, such as the view of works of literature as all being bound together with one another in an endless web (but not necessarily with a genetic connexion) or the idea that the reader's associations have at least as much relevance for interpretation as do whatever intentions the author may once have had.²⁴

Accordingly, the publisher of a digital text can welcome an open structure where the users, ideally, can continually introduce new links. The digital edition thus resembles a poststructuralist utopia - a web of digressions and associations, a non-linear system rather than the story or poem laid out in one direction to which we in our previous reading culture have become accustomed. In this way an Internet edition becomes the really decentered art-work.

This all means, furthermore, that the digital hypertext will in all probability distance itself from the effects which the author strives to produce in the reader, the artfully calculated hold which builds upon the simple thought that the reader begins at the beginning and reads through the narrative to the work's end. Reading is seldom like that in the Internet world's ad lib zapping between different pages. The non-linear technology of digital hypertexts therefore often tends to militate against the aesthetic qualities of the works which it is otherwise able to mediate to us. How far this is desirable or regrettable is for the user to decide - but the logic of the digital edition seems to be that the author's intention becomes marginalized while the reader's own choice becomes upgraded.²⁵

(Åter till början av artikeln)

The digital version of Almqvist's Samlade Verk can be very briefly described. The simplest way to grasp how it works is to sample it oneself.²⁶ The basic idea is that the digital version shall contain the same material as the book version, and the layout is such that to every volume there corresponds a web site, a "net volume".

A more radically conceived digital version would have freed itself from the layout of Samlade Verk and opened the possibility of freer combinations - perhaps a web site with all of Almqvist's works in alphabetical order, rather than the web site we now have which structures the contents according to a list of the volumes in Samlade Verk.

That we have kept the division into volumes as in Samlade Verk depends not a little upon our having retained the pagination of the book version in the digital version. Whoever reads on the screen the scene from Det går an where Albert and Sara eat "Hallon och grädda" (raspberries and cream) in Strängnäs will find the figure for page-number 31 at the bottom and so can easily find the right place in the printed book. For the same reason we have preserved all the original page numbers in our annotations. The digital version is thus an aid to reading the book version, not a totally independent form of publication.

In addition to the edited texts every volume in Samlade Verk contains an introduction (concerning the origin of the work, its reception, etc.), a critical commentary on the text, two lists of variant readings (accidentals and substantives respectively) and a section with annotations, glosses and so on. All this is included in the digital version. There are no links in the introduction (except in regard to its own endnotes), or in the textual critical commentary. The list of variants and the notations, glosses etc. on the other hand, are linked to the text. They are, however, independent of one another, which means that only one of these categories can be consulted at a time: the connection is made by means of emphasis in blue of the relevant parts of the text. On a special menu one can choose which category one wishes displayed (e.g. annotations, substantives, accidentals). That one can choose between a sheer reading of the text and reading it with all the references and comments is a clear advantage; that one can only display one category of these at a time is a clear disadvantage.

To the practical difficulties with the Internet version there belongs still today the fact that the text in the HTML format is stored in files which are often substantively smaller than the work concerned (the files correspond as a rule to around ten to fifty book-pages). This makes the handling of texts of greater length more difficult. Whoever wants to work more intensively with the text has of course the option of copying it onto his own computer.

The digital Almqvist's Samlade Verk is a micro-hypertext within a macro-hypertext, for Internet is really nothing else but the most comprehensive hypertext that exists. The problem at the moment is that its layout does not allow very long documents - with present techniques the loading times would be far too long. The digital world is built up on short documents, the world of books upon longer ones. One can also express the difference by saying that the digital text is set up for consultation, the book text for reading.

The aid essential to our digital version is the search function attached to the digital issue - a tool allowing the user to choose whether he or she wants to consult the whole body of the Almqvist texts stored, a single volume or certain divisions common to the texts as a whole (introductions, variants, annotations or Almqvist's works), or, say, whether the consultation shall be directed to whole words or to parts of words (truncation). It is in its possibilities for searching and cross-references via links and in the corresponding technology that the digital edition has its main usefulness, rather than in the option of reading the works on the screen.

The intention in the long run is that the digital version shall be supplemented by materials not found in the books - concordances, vocabularies, references to articles, illustrations, facsimiles of book-pages, a bibliography, examples of music, etc. The digital version can then be widened to become an Almqvist Web - a digital archive of which the open way of functioning will probably conflict with the meditative way of reading in communing with the spirit of the author which Almqvist himself advocated.²⁷

(Åter till början av artikeln)

"Information wants to be free", it is often said in connection with Internet. With that is meant that the idea of Internet itself is that information should be able to circulate freely: the private operator should be able to copy computer files and programmes from Internet. This is an idealistic and democratic principle which in the space of a few years has made Internet the most important international channel for the spreading of information. But it is also a principle which collides with copyright legislation, the law's way of protecting the author's right to his works, for example. Internet is really an anomaly in an industrial society: the promise to give something for nothing, to give information for free. "Information wants to be cheap" is really the furthest that an commercial Internet publisher could stretch himself.

Internet publishing raises two problems for the editor of scholarly texts. One is to do with copyright, the other with the preservation of quality. Both can be reformulated as specifically technical problems. Neither of these problems is unsolvable, but both are difficult to handle in practice.

The problem of copyright is at present a nightmare for those producing books in Sweden. Publishers and authors have for a long time been locked in negotiations about who really should have the right to publication of the digital version of a work. The authors appeal to their rights as authors and demand extra payment for giving up this right to the publishers, while the publishers fear that the authors may organise digital competitors to the book versions unless such authors make over their right to digital publication along with the right to book-publication. Behind the publishers' standpoint, in other words, lies the idea that digital publishing is a threat to the book as a means of communication. But in principle the legal situation, as always, is that the author has the right to publication and reproduction of his own work.

Copyright law in Sweden implies, for a scholarly edition, that in practice also the publisher is adequately protected legally. A responsible editor of a text has of course the right of authorship to introductions, commentaries and so on (as with all other authors' rights the protection covers the presentation as such, the linguistic formulations, while the factual contents are not protected in copyright law). But also the edited text itself is covered by author' rights, namely when it is a text which the editor has so altered, through active scholarly labour, that it differs markedly from the version printed earlier. Authors' rights as in force include protection for "Whoever has translated or revised a work" (para. 4).²⁸

Where the boundary between reprint and revision should in practice be drawn can here be left open, but the important thing for the digital publisher of scholarly editions is to remember that in general the material is legally protected so as to accord rights to the author. In practice, nonetheless, this scarcely creates any problems for a scholarly edition - the editorial work is seldom done mainly for payment. Even if our labours on Samlade Verk are remunerated per printed page, yet we have not managed to bring about such a productivity in the editors of our volumes as the payment for his pages aroused in Almqvist in the 1840s.

An author's rights apply for seventy years after the author's death and cover the author's economic and personal interests. With that is meant that whoever would publish a work may not do it without the author's permission. The work may not be distorted either. Here things become critical for those working on Internet. "Information wants to be free" - this principle means that anyone whatever should be able to copy information further, maybe even alter it. This is permissible - also with respect to the legal rights of authors - if one does it for simple use. But to make information public without its originator's permission is an invasion of the author's copyright - and to change such information can be a further invasion in cases where the changes are so major and insensitive that they also strike at the author's personal rights.²⁹

All this means that, in theory, editions published digitally ought to be thoroughly secure legally. People can use them as they think fit, type links to them from their own web sites, make them known in different ways. But they may not copy the material for their own publishing, whether digitally or printed on paper. Nor may they change the text without the author's permission. The problem is only that in practice the technical possibilities for copying and changing texts are now so much more varied than before.

Certainly our own digital version is covered against changes in the sense that no unauthorised user can gain entrance to the computer files at Språkdata and make modifications to them. In this sense the digital Almqvist edition is secure. Nevertheless it is a simple matter for anyone who wants it to get access to a digital copy of Almqvist's Samlade Verk. In the space of a few minutes one can copy a whole literary work from Internet to a disk. And during a further few minutes, by means of a text-editor in a word-processing programme, one can make sweeping changes to the text. Thus one can modernise older spelling forms half automatically - using a series of macro-commands linked to one another.

In this situation it is small consolation for the publisher of critical texts to know that the possibilities for making swift alterations are practically just as great even if the text is not on Internet. The way round via the technique of scanning need not be long. With a good OCR programme the computerization of a printed book can be relatively swift and sure (if not yet perfect letter by letter). And then henceforth it exists in digital form, exposed to changes and revisions. However we publish Samlade Verk the texts can never be shielded from mechanical manipulation.

All this is well known for anyone working in a modern academic milieu. "The electronic library" is already now a reality - the amount of texts on Internet is now so great that not a few literary quotations can be identified just as quickly on Internet as through using dictionaries of quotations. But what will happen when the supply of digital Swedish classics becomes yet greater? Is there a danger of pirate editions? The possibilities are in any case so multiple that, prudently enough, those who give out critical texts regularly search through Internet to see if there are any alternative versions there of these texts. For it would be a nightmare if a faulty version of a text should start to be spread through repeated copying. In that case Internet would become a Pandora's box with the worst pains reserved for the editors of critical texts - a swarm of corruptions.

So the question of authors' legal rights shades off into one of security. How does the editor protect himself against distortions?

For the present my answer must be that she or he should not shield any other than his own publications (which obviously should be strictly guarded) - not because it would not be technically possible to do more, but because this would be in conflict with the main idea of digital publication. "Information wants to be free" - and whoever makes his work public on Internet should accept Internet's conditions. Of course one can protect oneself against widespread copying of one's texts. One can for example store them in a format requiring a programme to which most people have no access. One can require a password or some form of authorization. And one can store the texts in the form of pictures instead of as text-files.

All such measures would still be in conflict with the very raison d'être of digital publishing: making the text quickly and easily accessible to as many as possible. As editor of a scholarly text one should certainly not have any illusions about the breadth of one's public (a sale of over a hundred copies on the open market of a critical text makes it a bestseller), but just for that reason one should meet the users' wishes as far as is possible.

The digital Almqvist edition's public consists in the first place of university-based researchers, but hopefully it will also be able to attract other users: high school pupils, pupils of popular academies, others who write about literature. Then it will not be enough to work with a format requiring a password, help from computer expertise or all too complicated retrieval of programmes. One should contrive to have texts which are usable and able to be found in the same way as most other texts on Internet. This entails that the texts lay wide open to retrieval and subsequent publication in another form. In this respect the question of security is for the present unsolved, even if one would like to hope that the development will lead to ways of solving the problem.

Until then one has to remember that the topic of security should chiefly be seen as a special instance of the question of quality. Internet's great and unacknowledged problem at present is that so much of the information on it is false, out of date or distorted. But information wants to be certain. Otherwise it is not information.

(Åter till början av artikeln)

(An earlier version of this article was published i Swedish as "Almqvist på Internet. Om publicering av en textkritisk edition som digital hypertext", Tidskrift för litteraturvetenskap 26(1997):2, s. 60-74.)

(Åter till början av artikeln)

Om författaren

Johan Svedjedal är professor i litteraturvetenskap, särskilt litteratursociologi vid Uppsala universitet sedan 1993. Föreståndare för Avdelningen för Litteratursociologi. Biträdande huvudredaktör för C.J.L. Almqvists "Samlade Verk" (1993-), bl.a. med huvudansvar för den digitala versionen.
Pågående forskning : Birger Sjöberg. En biografi och Emilie Flygare-Carlén, Pål Värning. Textkritisk utgåva för Svenska Vitterhetssamfundet

(Åter till början av artikeln)

Fotnoter

One has to use the term "work", since Almqvist did not limit himself to writings. His music and pictures which he may have painted will also be included in the edition. Åter till texten
For a review of earlier editions and of the guiding principles applied to Samlade Verk see Bertil Romberg and Johan Svedjedal, "Carl Jonas Almqvists Samlade Verk. Planer och principer", Samlaren 114(1993)(tr. 1994), pp. 15-26. Åter till texten
The scanning in is carried out by Autotext Ltd. in Norberg. Åter till texten
Several examples are instructively described in Lars Burman, "Kancellanserna i C.J.L. Almqvists Amalia Hillner (1840)", Tidskrift för litteraturvetenskap 24 (1995):2, pp. 47-54. Åter till texten
See Rolf E. Du Rietz, "Böcker, texter och editioner - av alla de slag", Fiktionens förvandlingar. En vänbok till Bo Bennich-Björkman den 6 oktober 1996. Ed. Dag Hedman and Johan Svedjedal, Uppsala 1996 (Skrifter utgivna av Avdelningen för litteratursociologi vid Litteraturvetenskapliga institutionen i Uppsala = Publications from the Section for Sociology of Literature iat the Department of Literature, Uppsala University, 33), [pp.89-107], p. 96 ff. Åter till texten
For detailed argumentation along these lines cf. Peter L. Shillingsburg, Scholarly Editing in the Computer Age. Theory and Practice, 3rd edn., Ann Arbor 1996, p. 166ff. Åter till texten
Språkbanken's address:
http://svenska.gu.se Åter till texten
The address of the Swedish Literary Sociey: http://svenska.gu.se/vittsam.html.
Arvidi's address: http://hum.gu.se/~litmm/Arvidi.html Åter till texten
Yvonne Cederholm at Språkdata has chief responsibility for this work, while from the editors' side Johan Svedjedal has the scholarly care of the digital hypertext version. Åter till texten
The following section recalls part of my article "Det skönlitterära nätet. Internet och svensk skönlitteratur", in Medialiseringen av Sverige, ed. Anders Björnsson and Peter Luthersson, Stockholm 1997, pp. 29-40. Åter till texten
The address for Books on-line:
http://www.uni-sb.de/z-einr/ub/ebooks/bookauth.html
The address for Project Bartleby:
http://www.columbia.edu/acis/bartleby
The address for Btj:s länklista:
http://www.btj.se/btj/saburl/saburl.html
The address for Jane Austen's novels:
http://curly.cc.utexas.edu/~churchh/janeinfo.html
The address for The Victorian Web:
http://www.stg.brown.edu/projects/hypertext/landow/victorian/victov.htm
There are useful pointers to literary hypertexts on e.g.
http://library.lib.binghamton.edu/english.html
http://etext.lib.virginia.edu/uvaonline.html Åter till texten
Shillingsburg 1996, p. 161. Åter till texten
The Karin Boye Society's web site
http://www.ivo.se/kboye/dikter.html Åter till texten
The address for Selma Lagerlöf's works:
http://wwis.upnet.se/selma/indexswe.htm Åter till texten
The address for Projekt Runeberg:
http://www.lysator.liu.se/runeberg/katalog.html Åter till texten
Projekt Runeberg is included in "Länkskafferiet" (http://www.ub2.lu.se/skolverket/), a "virtual school library" which was commissioned by Skolverket and Svenska skoldatanätet. "Länkskafferiet is a data base with 895 Internet resource units structured according to subject and of guaranteed quality. Skafferiet is a means of seeking information on Internet which is intended in the first instance for pupils and teachers in Swedish schools." According to the presentation Internet resources are included which "just now are the best we know of in this area." It would seem to follow that no first-hand evaluation of textual quality has been done. Åter till texten
G. Thomas Tanselle, "Printing History and Other History", Studies in Bibliography 48 (1995) [pp. 269-289], p. 288. Åter till texten
Jerome McGann, "Radiant Textuality
http://jefferson.village.virginia.edu/public/jjm2/radiant.html Åter till texten
"Boktryckerikonsten och Folkuppfostran" (originally Aftonbladet, 5 January 1839), Journalistik. Selection, introduction and commentary by Bertil Romberg, I: 1839-1845, Hedemora 1989 [pp. 56-63], p. 57f. Åter till texten
Cf. e.g. Peter Shillingsburg, "Principles for Electronic Archives, Scholarly Editions, and Tutorials", The Literary Text in the Digital Age, ed. Richard J. Finneran, Ann Arbor 1996, pp. 23-35. Åter till texten
Charles L. Ross, "The Electronic Text and the Death of the Critical Edition", The Literary Text in the Digital Age (1996), pp. 225-231. Åter till texten
Cf. e.g. G. Thomas Tanselle, "Textual Instability and Editorial Idealism", Studies in Bibliography 49 (1996) [pp. 1-60], p. 59f. Åter till texten
For a survey of the debate between "intentionalists" and "historical text-sociologists" see e.g. Johan Svedjedal, "Textkritisk lietteraturteori. Några linjer i svensk och anglosaxisk textkritisk debatt", Textkritik. Teori och praktik vid edering av litterära texter. Föredrag vid Svenska Vitterhetssamfundets symposium 10-11 september 1990, ed. Barbro Ståhle Sjönell, Stockholm: Svenska Vitterhetssamfundet, 1991, pp. 42-78. Åter till texten
This line of development within textual criticism is described in e.g. G. Thomas Tanselle, "Textual Criticism and Literary Sociology", Studies in Bibliography 44 (1991), pp. 83-143 and D.C. Greetham, "Editorial and Critical Theory: From Modernism to Postmodernism", Palimpsest. Editorial Theory in the Humanities, ed. George Bornstein & Ralph G. Williams, Ann Arbor 1993, pp. 9-28. Åter till texten
See several of the contributions to Hyper/Text/Theory, ed. George P. Landow, Baltimore & London 1994: Espen J. Aarseth, "Nonlinearity and Literary Theory", pp. 51-86; Gunnar Liestøl, "Wittgenstein, Genette, and the Reader's Narrative", pp. 87-121; "'How Do I Stop This Thing?': Closure and Indeterminacy in Interactive Narratives", pp. 159-188. Cf. also e.g. Philip E. Doss, "Traditional Theory and Innovative Practice: the Electronic Editor as Poststructuralist Reader", The Literary Text in the Digital Age (1996), pp. 213-224 and Johan Svedjedal, "Internet och den svenska skönlitterturen", Nordisk kultur under internationell press, København (Tema Nord 1997:523), pp. 50-58 (for an earlier version, see above, note 10). Åter till texten
The address of the publication is:
http://svenska.gu.se/vittsam/almqvist.html Åter till texten
Cf. Johan Svedjedal, "Om två slags lässätt. Om receptionen av C.J.L. Almqvist", BLM 66 (1996):6, pp. 21-25. Åter till texten
For the law concerning copyright for literary and artistic works, SFS 1960:729. Cf. Ulf Bernitz et. al., Immaterialrätt, 5, revised edn., Stockholm 1995, p. 33. Åter till texten
Cf. Bernitz et. al. 1995, p. 49ff. Åter till texten

Åter till Human IT 3/1998