Introduction

1When attempting to judge a digital edition several questions arise immediately: Is this the optimal method to deal with the source material? Is there scholarly value and where does it lie? Does it achieve what it claims to do? Finally, does it belong to the digital world only by the broadest definition of being housed by computers or does it offer some truly digital enhancements? In case of The Correspondence of James McNeill Whistler (Whistler’s Correspondence) at the University of Glasgow it is quickly visible that while the appeal to the general public and scholarly value of such a resource cannot be disputed, the digital aspect of the edition is seriously lacking and will become even more obsolete as the users’ expectations rise with new technical developments. Nevertheless, the solid foundations of the project ensure that Whistler’s Correspondence still has the potential to overcome its current presentational deficiencies. In the subsequent paragraphs I will attempt to present the current state of the online edition of Whistler engaging with the questions mentioned above.

Whistler

2 James McNeill Whistler (1834-1903) — according to the popular myth a temperamental, witty dandy, endlessly feuding with the establishment — was a widely-recognized artist, adept in many media as he completed over 500 paintings and numerous etchings, pastels, watercolours, drawings, and lithographs during his lifetime. Initially influenced and inspired by many sources he eventually developed his own individual style. Leading the Aesthetic Movement he was promoting the ‘art for art’s sake’ philosophy. From the beginning of his artistic career he developed broad social contacts with numerous painters (Courbet, Manet, Fantin Latour and Degas among them) and important poets and writers like Baudelaire, Wilde and Swinburne, thus becoming a pivotal figure of the British and French art circles. Whistler’s flamboyance assured he spent all his professional life in the limelight of public attention, especially after his long libel case against Ruskin ended in Pyrrhic victory. Whistler was instrumental in establishing the credo of modern art when he preached in favour of the art for art’s sake and announced the end of naturalistic transcription in his famous Ten O’Clock lecture. His influence continues well after his death and the mass culture of the 20th century brought his fame to a new level of recognition. Born in America, both his face and probably his most recognized opus Arrangement in Grey and Black No.1 — colloquially nicknamed as Whistler’s Mother — have eventually made it onto US postal stamps. Continued popularity of his artworks combined with the Whistler’s legend and appeal of his social circles assure ongoing interest in online publication of his correspondence from both scholars and the general public.

The centenary edition

3 The Correspondence of James McNeill Whistler at the University of Glasgow is an online centenary (first published in 2003) edition of his correspondence that covers the period from 1855 when Whistler, aged 21, left America to study art in Paris until his death in 1903.

4The University of Glasgow owns the most comprehensive resources available for research on Whistler. The Hunterian Art Gallery belonging to the University of Glasgow, second of the two major public collections of his work — the Freer Gallery of Art in Washington being the foremost — includes 80 oil paintings, several hundred drawings, watercolours and pastels and numerous personal items. The largest single collection of the artist’s correspondence and papers, comprising some 7,500 documents, together with his own publications and a specialised reference collection of over 1,800 volumes is preserved in the Department of Special Collections of the Glasgow University Library.¹ The University holds Whistler’s copyright and has supported the publication of the correspondence as part of its commitment to make its collections available to researchers and the general public. These assets make the University of Glasgow, which in 1992 opened a dedicated Centre for Whistler Studies, in all likelihood the best-equipped place in the world to attempt a project such as an edition of Whistler’s vast correspondence.

Scope of the edition

5Though some small subsets of Whistler’s letters have been published earlier in print by Mahey (26 letters), Thorp (75 letters) and Merrill (89 letters), even combined they cover barely a fraction of the artist’s extant correspondence. Whistler’s correspondence with a scope of 10,000 letters vastly surpasses the scale of the earlier, printed editions and is the first project that undertakes a transcription of the complete body of these texts and their publication in electronic form. Extensive use was made not only of the Glasgow University Library material but also of the Pennell-Whistler Collection in the Manuscript Division of the Library of Congress, Washington, D.C., as well as other collections worldwide.²

6The online edition of Whistler’s correspondence consists of over 13,000 records of letters, accounts, legal documents and miscellanea such as drawings or even menus. Letters included in the edition date from 1855 until his death in 1903. It was in Paris in 1855 that Whistler first began his serious study of art, which explains why the editors from the Faculty of Arts in Glasgow chose this threshold. The edition includes all the letters written by Whistler and all the letters written to him; letters mentioning Whistler were included on a selective basis, though the editors do not explain the inclusion criteria. The collection contains also letters of Whistler’s mother, Anna Matilda Whistler.

7The general spirit of the edition was to provide mass-accessibility and accurate annotation of the contents of the correspondence, fit for the broad and not necessarily scholarly audience: students, historians (art and otherwise), collectors and curators.³ Thus, the edition is heavily annotated when it comes to persons, places, events and works of art mentioned in the texts — all of high interest to a historian and especially a historian of arts and very helpful to those with less experience in this area: students, collectors and family researchers. Auxiliary records with biographic notes for persons mentioned in the letters and English translations of letters written in French further facilitate the use of this resource.

Digital Trancripts

8 A detailed view of a single letter (see fig. 2) consists of the transcription of the letter with a basic metadata header: the system reference number, author, date and place of sending, recipient and place of receiving, repository, call number and document type. The range of textual phenomena delineated in the transcript is very sparing and hardly any use has been made of features that could not be found in classical print. Notes and references to the people are conveniently made into hyperlinks that connect with the corresponding biographical or editorial notes, but otherwise the user only encounters underlining and occasional strikethrough to mark crossed-out passages.

9This simplicity reflects the customary form of printed editions and indeed makes the edition very straightforward to use; but the single viewing mode does not facilitate reading nor does it fully exploit the range of possibilities available to a web based edition. Probably the greatest disappointment arises from the absence of facsimile images. (This lack is certainly a result of the temporal and organisational limitations of the project; ⁴ nevertheless, the very simple addition of a link to external resources would still have been possible in those cases in which the editors already mention in a note that a digital facsimile is available from another institution.) This drawback has serious negative consequences: it is impossible to compare the transcription with an image of the original in cases where it is not available anywhere on the web, and even if it can be found online the reader is not provided with a hyperlink, but only with a note that it might be found elsewhere (see fig. 3 and 4).⁵ Thus, on the basis of the online edition alone it is neither possible to judge the faithfulness of the transcriptions, nor to pursue research questions that depend on the original layout or authoring process.

Editorial practice

10The editors followed a common practice in traditional, print editions of modern letters. As is symptomatic from the brief amount of space reserved for the explanation of the actual editorial policy they were more concerned with presenting a reading-text of the correspondence than with diplomatic representation of the documents. The edition refrains from in-depth representation of the layout or transcriptional features.

11 Nevertheless, when the editors state their editorial practice in just one paragraph, they are describing only a handful of features explicitly. ‘The transcriptions reproduce the text as written, including punctuation, capitalisation and errors of spelling, grammar and foreign accents; deletions and insertions are also marked. Occasional use has been made of ‘[sic]’ to confirm that the text is indeed as written. Editorial insertions are marked by square brackets.’ Such conciseness obviously cannot document every editorial intricacy of the thousands of transcribed letters. The editors do not make their policy clear even if they must have had one — e.g. the underlining from the sources is preserved in the transcription but there is no mention of it in the short introduction. Therefore the user could not be sure what the underlining means nor rely on it being consistently marked and would have to refer to sources. Conversely, very small-scale substitutions (e.g. overwriting of a small number of letters) seem to be sometimes left unmarked while in other cases are treated as deletions.

Website organization

12The website hosting the digital edition of Whistler’s correspondence follows a similar layout as other Whistler-related pages⁶ hosted by the University of Glasgow. Everything that is needed to know about the project’s participants and funding is clearly presented on the project’s website along with other vital information like copyright notices, citing guidelines and lists of abbreviations used throughout the edition. Finding contact information is similarly easy through the website’s dedicated subpage, providing e-mail addresses and a special enquiries form.

13 A simple menu allows the user to navigate to one of many subpages describing the project itself, the persona of Whistler, contacts and other resources, while the online edition section is further divided to allow jumping to specific search forms, (e.g. Persons, Date, Subjects). At all times the current position in the structure of the website is displayed to aid navigation.

14Results of the search (see fig. 5) are presented as a list of letters ordered chronologically. Each letter entry in the list shows the basic metadata summary. Letters published elsewhere are further marked as published. Other documents (e.g. accounts or drawings) are also clearly indicated.

15The navigation structure and simple searches are very straightforward even for readers unfamiliar with digital editions. Yet this simplicity also has a negative side: it is impossible to carry out more advanced queries that combine basic search criteria or that use the full metadata that is evidently stored in the system (like a search for the particular type of document or for the published letters only). Even simple browsing of letters can be cumbersome as there is no ‘next item’ button or a similar solution. The closest one can get to viewing the entirety of the edition is browsing through the letters year by year.

16 The edition lacks separate indices, though the search facility can partly fill that gap; for example searching for all persons enables the user to browse Whistler’s correspondents by the first letter of their surname; similarly searching for all places allows the user to see places grouped by country. In the case of persons, the editors always provide at least a short biographical description and more detailed entries for more prominent individuals.

17 Whistler’s correspondence is easily accessible without paying and requires no registration. Yet it is copyrighted in its entirety and thus does not allow for the easy reuse of data and does not supply any kind of technical interface for viewing or harvesting the source data. Nor does it integrate with social media or allow for discussions linked with particular parts of the source. However, each document has a unique reference number and can be easily cited separately. The construction of website URLs similarly allows linking to a single document, though it needs to be manually copied from the browser’s address bar as there is no feature to assist in extracting a link or generating the citation.

Technical Background

18 Whistler’s Correspondence, very much like other enterprises of a similar scale, spans many years, outliving the tools and workflows it started with. Decisions that were perfectly reasonable in 1997 when the editors employed an SGML mark-up based on a custom DTD with only some elements in common with the TEI Guidelines are no longer considered good practice nowadays. The files were edited and viewed with the SoftQuad Author-Editor and the Panorama SGML browser and later adjusted to make them compatible with XML processing tools for XSLT transformation into HTML. The project also made use of relational databases that changed along the way: from Paradox/FoxPro through MS Access and finally to MySQL for use on the website.⁷ This technical legacy affects not only the data structures or potential for future publication interface enhancements but also the possibility of continuing the present editorial work (as it is increasingly difficult to run old SGML-based software on modern systems). This aspect of the technical history of the Whistler’s Correspondence clearly illustrates an inherent problem with digital projects: during their lifetime they face the need to cyclically switch to new, better methods, software and hardware environments while always preserving and migrating the digital knowledge-base, which requires continuous funding, maintenance and re-creation of the web-based publication in line with current technology. Presently, the intention for Whistler (funds permitting) is to convert SGML fragments to TEI P5 files that could be edited with modern XML-based software and integrated into the eXist-based publishing environment currently used in Glasgow.⁸

Conclusion

19Editorial projects should be judged considering their individual circumstances: financial and human resources, the length of the project and pre-existing material (e.g. printed editions) to build upon. In some aspects the online edition of Whistler’s letters seems not to fulfil the expectations set up by more recent similar projects like The Letters of Vincent Van Gogh.⁹ The slightly dated layout and simplistic interface, the single fixed visualisation, and the unavailability of facsimiles may be hard to accept for the contemporary reader. Yet we need to take the context into account: Whistler’s edition went online in 2003 and what nominally is only a decade, in this case is an era apart when it comes to the web design, the availability and quality of tools and the permeability of standards such as TEI into the wider community of editors. Whistler’s Correspondence is a no-nonsense, extremely well-commented, freely accessible collection of an enormous number of letters that can be used equally easily by the general public interested in Whistler as by professional historians. This is the result of more than a decade of scholarly work that is actually of interest not only to academic researchers but also attracts a significant non-scholarly audience as well.

20Obviously there is room for improvement. In my opinion the scholarly aspect of the edition would benefit in particular from two things: the addition of digital facsimiles and improvements to the search engine. The former, however, requires substantial additional effort to create and publish images of appropriate quality and perhaps poses even greater challenges in acquiring the necessary copyright licenses. Enhancements to the interface, though not necessarily of tremendous scholarly value, might, on the other hand, increase the appeal to the general public and improve the using experience for all readers.

21To conclude: Whistler’s correspondence is definitely a valuable scholarly source of historic material. The edition is a freely accessible resource of impressive size with mostly otherwise unpublished material. It can be used to answer multiple research questions for a broad range of disciplines. Patrick Sahle very pragmatically defined the digital edition as ‘a concept that is not restricted to the technological limitations of print technology but that realizes a digital paradigm’ (Sahle 2014). This sense of ‘digital’ does not seem fitting for Whistler’s correspondence, since, save for its size, the edition could very comfortably be presented on printed pages. For now, then, let us just stick with the editor’s choice and call it an ‘on-line’ edition. Nevertheless, as I argued elsewhere (Turska) the long-lasting value and the prospects for scholarly use and re-use lies in the quality, consistency and adequate depth of the underlying digital modelling of the information, not in its presentation that has to be considered ephemeral by its very nature. For Whistler’s correspondence the potential to become truly digital, in the sense of achieving far more than print could ever hope for, is there, waiting to be unleashed. The intimate knowledge and laborious research of the editors are preserved in the mark-up, never to be lost as long as there is the will and the resources to maintain, re-pack and re-paint it with latest web technologies to enhance the raw data with a set of digital tools to explore it.