Review

1The Shelley-Godwin Archive will provide all known and relevant manuscripts of Percy Bysshe Shelley, Mary Wollstonecraft Shelley, William Godwin, and Mary Wollstonecraft in order to bring the dispersed handwritten legacy of this family of writers together on a common platform. It was launched in October 2013 with an edition of the Frankenstein Notebooks by Mary Shelley. At present (in October 2014) the Shelley-Godwin Archive is still in the beta phase, indicated by a marker in the header. Further information about the project’s time frame is not provided on the website, with the last entry of the ‘Recent News’ dating back to October 2013. Despite the beta state, the Shelley-Godwin Archive at this point can still be considered relatively advanced in its presentation and technical features. Moreover, it is a good example of an edition-in-progress, publishing its content step by step.

2This review consists of two parts: (1) An evaluation of the conceptual and technical framework of the Shelley-Godwin Archive as of October 2014, and (2) a review of the first release of the platform, the edition of the Frankenstein Notebooks. Finally, the review will close with an overall summary.

The Shelley-Godwin Archive

3The Shelley-Godwin Archive is a collaborative project between various institutions, in particular the Maryland Institute for Technology in the Humanities (MITH), the New York Public Library, and the University of Oxford’s Bodleian Library. With major contributions from the Huntington Library, the British Library, and the Houghton Library, the Shelley-Godwin Archive aims to cover 90% of all known and relevant handwritten documents of the Shelley-Godwin family. The project started its work in January 2013 and was initially funded by a three-year Humanities Collections and Reference Resources grant from the Preservation and Access division of the American National Endowment for the Humanities and a grant from the Gladys Krieble Delmas Foundation. Although the funding will end in June 2015, the Shelley-Godwin Archive is designed to be open-ended. MITH intends to continue its work with Liz Denlinger, David Brookshire, and Neil Fraistat as editors and a larger team of volunteer contributors.¹ The list of contributors is exceptionally long (about 60 names) and reveals the edition as a collaborative product in which different tasks are shared between experts in the respective fields (philologists, librarians, software developers, consultants etc.). Despite the size of this group, the technical staff clearly make up the majority.² In the first phase of the project, two graduate seminars at the University of Maryland and the University of Virginia also worked on transcriptions and encoding, and future collaborations with classrooms are planned.

The archive’s content

4From the late 18^th century on, the Shelley-Godwin family produced a range of literary works, of which the most prominent are: An Enquiry Concerning Political Justice (1793) by William Godwin; A Vindication of the Rights of Woman (1792) by Mary Wollstonecraft; political writings and poetry by Percy Shelley; and Frankenstein (1818) by Mary Shelley. The launch of the digital edition of the Frankenstein Notebooks in October 2013 marks the start of a series of publications, which will unify the widely scattered literary work of the family members. The Shelley-Godwin Archive will provide digitized manuscripts in three different forms: (1) facsimiles with fully corrected transcriptions (TEI-encoded), (2) facsimiles with uncorrected transcriptions, and (3) facsimiles only. A color-coding system will denote the current status of each transcription and thus reveal their trustworthiness. No mention is made on the website of a future supply of critically edited or collated texts. Instead, the Shelley-Godwin Archive plans to establish a crowd-sourcing environment to engage users in collaboration and contribution to the Shelley-Godwin Archive‘s content.

Presentation and technical infrastructure

5 The modern interface and the clearly structured content of the Shelley-Godwin Archive make the website immediately very welcoming and easy to navigate for new users. To make the platform’s use even easier, a full description of the website’s functionality is provided under the menu item ‘Using the Archive’. Despite the suggestion to use either Google’s Chrome browser, the latest version of Safari, or the latest version of Mozilla Firefox, the full content and all the features of the edition (especially with regards to the different facsimile-text views) seem to work best with the latest version of Chrome. The ‘About’ page offers information about the project goals, its contributors, and its technical infrastructure as well as a short overview about the life and work of the Shelley-Godwin family. While the technical framework and the encoding guidelines of the transcribed texts are very well documented, the outline of the Shelley-Godwin family, their work, and their role for literary history is kept relatively short. A short summary of the benefits, which the Shelley-Godwin Archive brings to the research landscape of the Shelley-Godwin family, would also be a welcome addition.

6The technical infrastructure of the Shelley-Godwin Archive is built on the Shared Canvas data model. In line with linked open data principles, Shared Canvas is designed to support the description and presentation of digital facsimiles of physical objects in a collaborative environment. The data model is an abstract canvas, which allows for multiple annotations (e.g. transcriptions). Theoretically this entails that every user would be able to describe, rearrange and even reuse the facsimiles as well as associate additional annotations with the canvas. Thus, any single resource is no longer limited to one single purpose but can be edited by several persons both in the same project and in other contexts; and the interoperability between repositories that may hold related content is facilitated, too (Sanderson–Albritton).

7At present, the Shelley-Godwin Archive has implemented the Shared Canvas model such that each manuscript page is associated with an URI. Thus, the facsimiles and aligned transcriptions can be placed in different contexts, as for instance in the ‘Volume I-III Drafts in Chapter Sequence‘, which combines manuscript pages from witness A/B and C1/C2. As far as it concerns the possibility of adding multiple annotations to one canvas, the technical foundation is already built on RDF/JSON manifests. Before users will be able to target the canvases with annotations, the Shelley-Godwin Archive needs to realize its extension to a participatory platform. While the website does not reveal much about how or when this extension will support the open and interactive character of the platform, an article in the Chronicle of Higher Education suggests that the Shelley-Godwin Archive aims to be open to public participation at a large scale by 2016 and that the collaborative environment will be realized by implementing a framework called ‘Skylark’ (Howard).

8The idea of ‘sharing’ content extends to the technical infrastructure of the Shelley-Godwin Archive: all software applications, libraries and encoded transcriptions are available under open licenses (the Apache License, Version 2.0 and the Creative Commons Attribution license) on GitHub³ and can be reused by third parties. GitHub also contains the detailed technical documentation of the Shared Canvas model and of the whole edition. The only thing missing from the documentation – and which due to the open-ended character of the Shelley-Godwin Archive would have been particularly interesting – is a statement about the long-term preservation of the content and the role of the participating institutions in this issue.

Benefits of the Archive

9The literary work of the Shelley-Godwin family in multiple genres – treatise, novel, drama, and poetry – shows influences of historical events and movements such as the French Revolution, early feminism, and the Napoleonic wars. Bringing the (nearly) complete handwritten legacy of the family together in a digital form contributes not only to the research of the historical events they touched but also and primarily to the studies of English Literature in general and British Romanticism in particular. Furthermore, a common platform for all documents allows revealing possible connections to develop between the literary work of different family members. From a digital humanities perspective using established standards like the TEI as well as working with emergent standards of linked data makes the technical infrastructure of the Shelley-Godwin Archive exemplary for a new generation of open and interactive editions.

The Frankenstein Notebooks

10This central section of the review concerns the Shelley-Godwin Archive’s first editorial release: a digital edition of the Frankenstein Notebooks by Mary Shelley. More precisely, this release is the digital adaptation and incorporation (supplemented with a few corrections by the editor himself) of Charles E. Robinson’s edition The Frankenstein Notebooks: A Facsimile Edition of Mary Shelley’s Manuscript Novel, published in 1996.

11Mary Shelley’s epistolary novel Frankenstein (or The Modern Prometheus) was first published anonymously in London in 1818 when the author was only twenty. Mary Shelley had written the ur-story two years before while travelling in Europe with her future husband, Percy Shelley, and Claire Clairmont, during a stay in nearby Geneva in Switzerland where they joined Lord Byron and his physician John Polidori. In the rainy summer, the group entertained themselves by reading ghost stories, which lead Lord Byron to suggest that they each should try to write their own story. Mary Shelley’s novel is about the scientist Victor Frankenstein who, experimenting with the reanimation of dead bodies, creates a monster. Frankenstein became one of the most popular and most reprinted works of British Romanticism.

The edition’s content

12 The digital edition of the Frankenstein Notebooks is introduced by a short description of the transmission of Mary Shelley’s original notebook drafts, denoted as A, B, C1, and C2. These manuscripts were purchased by the Bodleian Library in 2004 from Mary Shelley’s descendant, Lord Abinger. Notebooks A and B (Bodleian MS Abinger c. 56 and c. 57), the oldest witnesses represented in the edition, were written in 1816 in collaboration with Percy Shelley as a recast of the lost ur-text of the same year. Together, these two notebooks contain approximately 87% of the novel as it was published in 1818. The notebooks C1 and C2 (Bodleian MS Abinger c. 58), fair-copies for prospective publishers of which only 12% have been preserved present the second witness. In the digital edition, the transcription of the notebooks and the attribution of hands are taken from Robinson’s print edition, as are key parts of his editorial introduction and his chronology of Frankenstein’s composition history. A third representation of the notebooks, ‘Volume I-III Draft/Fair Copy in Chapter Sequence’, might be at first confusing as the provenance of this (non-)witness is not explicitly explained in the introduction. The incorporated introduction of Robinson’s edition gives further insight and reveals that these volumes are an arrangement of the two drafts (A, B) and the two fair-copies (C1, C2) in order to construct the chapter sequence in which they later were meant to be published.

13Comparing the content of the digital edition with the previous printed version by Robinson, it should be mentioned that the content has not been transferred in its entirety. The 1996 edition provides, alongside the transcription, a transcript of the 1818 first print edition to offer insight into the transformation process between manuscript and print. This synopsis is lost in the digital version as the Shelley-Godwin Archive wants to represent only manuscripts and no printed volumes. Instead, the editors have chosen to provide a representation of the manuscripts according to the chapter sequence in the later published print. This method of constructing a non-existing text is debatable from a philological point of view. Moreover, without a clear explanation, the constructed volumes may come across as a third witness. Finally, it must be said that the introduction to the edited notebooks is relatively short and does not reveal much about the broader topical context of the edition. Although the Shelley-Godwin Archive suggests consulting the incorporated parts of Robinson’s edition – on which the whole edition is built in the end – due to their arrangement in the right column they seem to hold just a secondary role.

The digital remake

14The digital edition of the Frankenstein Notebooks has a modern, well-designed user interface that allows the user to navigate the content intuitively and to use offered features easily. The presentation of the edited notebooks can be divided in three parts: the metadata section, the toolbar, and the view of the edited notebook as facsimiles and corresponding transcriptions.

15 In the header section of the edition, the metadata about the respective folio is supplied. This includes author, hand(s), date, folio number, and bibliographical information of the source. The state of the encoded transcription and the provided metadata is assigned through color-coded symbols of three levels (red, yellow, green) to indicate how much work has already been done on them. Information about the editors who transcribed and encoded the respective page is not provided, however.

16 Below the metadata section, the digital edition provides a toolbar with several buttons that offer different options for viewing the digitized manuscript. These include zooming features for the image, the diplomatic or the reading-version of the transcription, its XML-code, and a view of the respective hand (in Frankenstein Notebooks, Mary Shelley’s or Percy Shelley’s hand). The user can also search in each page of the transcription and refine the results by choosing the hand and/or filtering by added and deleted passages. Searching through all available transcriptions at once, containing additional filters, refining and sorting options are offered under the menu item ‘Search’.

17 The third and main part of the edition consists of the transcription and corresponding facsimile. The default representation of the edition is a diplomatic transcription, generated from an underlying XML, which itself follows the TEI P5 customized module for genetic editing. Instead of capturing and presenting the linear structure of the document’s text, the genetic module of the TEI aims to trace the composition process. To this end the physical structure, namely the topological features of the text on the document, are captured and encoded. Changes of hand are encoded as well as deletions, additions, substitutions and similar editorial phenomena. In the representation the parts of the transcription, e.g. marginalia or additions over/under lines, can be displayed both, at their original position (diplomatic view) or integrated into the linear text (reading view).

18Below the main view of the edited notebooks the Shelley-Godwin Archive suggests for each folio the appropriate citation. For instance the first folio of Notebook B is cited with ‘Shelley, M. (1817) “Frankenstein – Draft Notebook B”, in The Shelley-Godwin Archive, c. 57, fol. 19r. Retrieved from http://shelleygodwinarchive.org/sc/oxford/frankenstein/notebook/b#n=1’. Such a citation suggestion can be very useful as it prevents incorrect references to the source. Nevertheless it would be preferable if ‘speaking’ URLs were used. The connection between ‘b#n=1’ and ‘c. 57, fol. 19r.’ is not evident, neither to a human being nor to a machine. To make the connection comprehensible, the citation should follow a canonical reference system.

19 Regarding the usability of the edition of the Frankenstein Notebooks, apart from temporary browser problems, only two minor points can be criticized: (1) the slider along the right-hand side of the transcription, supposedly for moving across a large number of pages, is confusing due to its unusual vertical form and would be easier to understand if horizontally designed, and (2) the position of the marginalia is shifted in the transcription and therefore the user might have difficulties in finding the corresponding line.⁴ This issue could be easily resolved with a text-image-linking and a mouse-over function between the marked zones on the facsimile and the transcribed text.

20 Regarding the encoding scheme of the transcription, it should be positively mentioned that the Shelley-Godwin Archive is absolutely transparent about its encoding guidelines in the project’s documentation. Besides explaining the encoding structure behind genetic editing, it offers a table with all employed elements and attributes⁵ as well as sharing the transcription files, the stylesheets and the schemata on GitHub. Nevertheless, the reasons for employing a relatively complicated and time-consuming transcription model like the genetic one are not evident, as few of the possibilities such a detailed encoding model offers have been used here. What appears on the edition surface – a diplomatic transcription – could have been realized with a much simpler encoding scheme.

Achievements of the digital edition

21First, it should be stated that no further scholarly value has been added to the Frankenstein Notebook’s edition by the content of the edition. In fact, rather the opposite has happened: the transcription and the attribution of the hands are taken directly from Robinson’s printed edition of 1996. Meanwhile, the genetic synopsis between the Frankenstein Notebooks and the first published volumes of 1818 has not been adopted. While it could be argued that the Shelley-Godwin Archive aims to include only manuscripts and no printed volumes, the editors excluded existing scholarly work from Robinson’s edition, without any convincing explanation for the decision. The result is a digital edition that is, in this respect, poorer in scholarship than the source from which it borrows.

22Even apart from the lost print-manuscript synopsis, there are questions regarding the data model the editors of the Frankenstein Notebooks opted for. Due to the fact that the notebooks display different redactions, genetic editing would suggest itself for the notebooks. And indeed the editors chose a TEI data model fit to capture genetic features and encoded the notebooks to such an extent that the topological structure of the text in the document and the changes of hands are documented. However, these two features – despite being crucial for a genetic edition in this case – cover only little of the potential which could have been exploited by using the genetic module: on the level of micro-genesis, in addition to the attribution of hands to the passages written by Mary or Percy Shelley, further layers reconstructing the chronology of the writing process would be necessary in order to reveal the genesis of the final text. Such layers are of particular interest and importance in those cases in which Mary modified the text again after Percy’s intervention.⁶ A second missing feature is the text-image linking, which would help the user to find the corresponding parts to the image-zones in the transcription.⁷ On the level of macro-genetic analysis, a synoptic presentation of the composition history through all witnesses (there are only two) to reveal the variations of the text in its different stages would be extremely useful; otherwise, the user has to go back-and-forth to connect the corresponding parts in the different witnesses.⁸ It could be argued that the editors renounced a deeper genetic investigation of the manuscripts, because the Frankenstein Notebooks do not even aim to be a genetic edition, but this would raise the question why the editors chose the genetic module at all and not a more compact and simple data model.

23Both criticisms – the loss of the manuscript-print synopsis and the ambivalent genetic character – lead to a bigger question: what is the focus of the digital edition of the Frankenstein Notebooks? The edition’s introduction claims to present ‘for the first time in digital form all the known manuscripts of Frankenstein‘,⁹ but this is not an editorial statement. To understand the edition’s ‘identity’, a more extensive introduction containing information on the editorial methodology and its scope, including insight about the motivation to choose the genetic encoding scheme would be necessary.

24In summary, the positive aspects of the digital edition should be underlined: the digitization of the Frankenstein Notebooks will allow scholars to gain deeper insights into the history of the origins of the Frankenstein story. This is of particular interest because of the long lasting debate about Mary Shelley’s authorship and the contribution of her later husband Percy Shelley to the novel. The digitized manuscripts reveal that very significant changes were made by Percy Shelley in words, themes and style, but they show at the same time that Mary Shelley should be credited with the whole narrative framework of the novel, which was written by her hand. The high-resolution facsimiles and the attribution of the hands in the transcription will greatly contribute to this important debate and will allow scholars to refute, confirm or modify the theses and theories regarding the origin of Frankenstein.

25Last but not least, and this is a crucial outcome of the transformation from Robinson’s edition into the digital form, the Frankenstein Notebooks goes much further than a simple digitization of a print. The edition has completely shed its book-character, being entirely adapted and transformed into the digital medium. Following the digital paradigm, the Frankenstein Notebooks has obtained even the advantages that go along with it: the storage format is flexible, and as data and presentation of the notebooks are separated, the data can be transformed in multiple ways to create different outcomes of the edition (e.g. a reading or a diplomatic view of the transcription). Further, the content is also searchable, and the machine-readable format even allows for refining both search and search results.

Conclusion

26The first release of the Shelley-Godwin Archive, the edition of the Frankenstein Notebooks, has a few minor weaknesses, including the loss of the missing manuscript-print synopsis from the print edition. Furthermore, the complex genetic encoding scheme coupled with the lack of genetic analysis represents a paradox. A more extensive overview of the principles and goals of the Frankenstein Notebooks will be a necessary addition to resolve these inconsistencies. Nevertheless, a rare and very expensive facsimile print edition has been made available for everyone in a way that goes much further than a simple digitization project. The digital edition of the Frankenstein Notebooks completely follows the digital paradigm and, as a result, obtains the advantages that go along with it. In the future it will be interesting to see how further material will be integrated and if the contents will be somehow interlinked.

27The Shelley-Godwin Archive is certainly a great resource for students and scholars. Even in beta-state, the project’s web presentation, its technical framework and its documentation represent a highly elaborate and accomplished project. By collecting the dispersed literary legacy of the Shelley-Godwin family and re-unifying it on a common platform, the Shelley-Godwin Archive is a great win for the international research landscape of the family of writers as well as of English literature and British Romanticism. Finally, I expect that the Shelley-Godwin Archive will be – with its linked data principles, its philosophy of ‘sharing technical achievements’, and its envisioned collaborative environment to empower ‘citizen scholars’ – a future contender for the state-of-the-art in a new generation of open and interactive digital editions.¹⁰