The following paper on Open Shakespeare was given at The Open Knowledge Conference 2010. It explores the past, present and future of the project as of Summer 2010.
Thank you for inviting me to speak here today. My aim is to introduce the Open Shakespeare Project and to explore both the opportunities it provides and the issues it faces. I shall do this by first recounting the history of the project, then outlining our current position, before looking at the things we intend to do in the near and distant future.
A quick glance at the archives of our blog reveals that the project began as, and here I quote a post by Rufus Pollock, “a ‘hello world’ type open knowledge project, which would illustrate what we mean by open knowledge and why it is useful.” Shakespeare’s celebrity made him the ideal subject for such a project. That post was written in May 2006, and over the years since then the framework of Open Shakespeare was constructed until one day, I, waiting for a lecture in the English Faculty, came across a flier that declared a “marriage of text and technology”, a project in which you could:
Add your own annotations and critical notes, create summaries and introductions, track down sources and create your own critical edition.
This intrigued me. Upon visiting the website, I found tools that would allow me to search Shakespeare’s works, compare texts side-by-side, and perform rudimentary statistical analysis. Furthermore, all such tools were ‘open’ and thus, I learnt, free for anyone to use and re-use. Text and technology had certainly been married, but I admit that I was eager for more, and felt that the project had yet to realise its flier’s claim of providing “an alternative to traditional scholarship”.
To be fair, not all the tools the flier promised were in place at the start of this academic year, and the most notable absence was the ability to “add your own annotations”. This tool held, and still holds, the greatest fascination for me. It strikes me now that the website as it was in October 2009 contained a curious contradiction, and one that may be relevant for other Open Knowledge projects. The tools provided as part of Open Shakespeare were certainly open and easy for anyone to use. However, the results obtained from such tools were highly likely to remain in the private domain, since there existed no way of reinvesting such results back into the project. Annotation, then, understood as the capacity to write a comment on Shakespeare’s text, would provide this missing link and so allow the results obtained from such open tools as text comparison to become part of the Open Shakespeare Project. With annotation, the site would have the potential to be not just a tool box, but a living and growing entity that could truly provide “an alternative to traditional scholarship”. After all, we talk about open knowledge, but must ask ourselves whether the production of open tools necessarily results in open knowledge and, more largely, whether we sometimes pay more attention to the adjective than to the noun.
When I eventually walked into a very cosy kitchen not far from the Chemistry Department in Cambridge, I found Rufus and several other students already busy with the project. Amongst them, it was Nick Stenning who did much of the work on implementing the annotation tool, allowing a basic version to be launched in March 2010. In addition to working annotator, the Open Shakespeare Project of today is considerably different from how it was in October 2009. Much of what is different concerns the diffusion of information and the display of what can be done with some of the open tools we provide. This is one of the purposes of our blog, which, in addition to running the occasional meditative article, contains a regular ‘Word of the Day’ feature and also acts as channel through which new introductions to Shakespeare’s plays are published. These short introductions, which will hopefully one day lead to more ample collaborative critical introductions have been written by students from Cambridge and elsewhere, and represent another opportunity for people to engage with and add improvements to the Open Shakespeare Project. With the blog and a constantly improving website, we have attracted attention from university students of all levels in England, as well as interest from academics as far afield as the United States and France.
One fairly recent improvement was the addition of a French translation of Hamlet to the list of resources available. This text by François Guizot dating from the 19th century was one of the most influential foreign versions of Shakespeare ever published, and, as far as I am aware, it is only through our website that it can be electronically viewed side-by-side with the English original. However, as anyone who knows anything about Shakespeare scholarship will soon say, the idea of an ‘English original’ is deeply problematic. One of the most important issues facing the project today is not the creation and circulation of knowledge as it was in its infancy, but the more fundamental question of what Shakespeare texts we are using. Our website contains both the Moby Shakespeare, based on the Globe edition of 1864, and Gutenberg’s copy of a first folio. The very fact that this has become an issue should really be a cause for celebration since, in agonising over our texts, we have at least attained the level of traditional Shakespeare criticism. A brief example will demonstrate what is at stake here. Many of you will know Hamlet’s famous soliloquy, which begins
O that this too too solid flesh would melt…
Some of you may be surprised to learn that this reading occurs only in one early edition of the play. Another instead has Hamlet wish that “this too too sallied flesh would melt” (where ‘sallied’ means ‘assailed’) and some editors have even amended the line to read “O that this too too sullied flesh would melt”. Simply put, we can no longer be sure of what Shakespeare wrote. How then can the results produced by our tools have any value? This, notes Brian Murphy in a recent edition of the Shakespeare Quarterly, is the devastating flaw of the close cousin of our website, the Open Source Shakespeare, begun in 2003 by a US marine. He also used the Moby Shakespeare, the most widely available electronic text of Shakespeare, and so does not benefit from all the textual work undertaken since 1864 that has completely altered the way in which certain plays, such as King Lear, are read. More recent electronic texts, such as that provided by Wordhoard offer a solution to this problem, and we are currently in discussion with Northwestern University about whether we can use their Wordhoard text. The sticking point being that the establishment of this ‘Northwestern’ Shakespeare took a great deal of time, and its owners are understandably hesitant to let their work be used freely under an Open license.
Another article from the same issue of The Shakespeare Quarterly as Murphy’s article, this time by Whitney Trettien, contains a revealing little phrase in which she points out that
…if Digital Humanities is, as imagined, to change the way we think about Shakespeare, it must embrace the web not simply as a content delivery platform, but as an expressive medium in itself.
The concerns about which text we should be using are concerns about “content delivery” and can, I feel, be circumvented by embracing the web as an “expressive medium”. Simply put, with the ability to annotate a text, Hamlet’s use of the word “solid” in act one, scene two, needs only a short comment in order to encompass all the variants currently posited by Shakespeare scholars. Should textual scholarship move further forward and invalidate even the more recent electronic texts, another comment is all that would be required on our site to maintain its relevance. The dynamic and reflexive space provided by annotation in the Open Shakespeare Project allows us to truly marry “text and technology” so that technology serves the text as much as the text serves the technology. In this user-led approach, not dissimilar to that offered by other web 2.0 initiatives, will the project provide an alternative to traditional criticism.
The obvious comparison in terms of user-generated content is Wikipedia, and this brings us to an issue with annotation that can never be fully resolved. When annotation went public, we took the step of hiding the button that would delete a comment in order to prevent vandalism. The ability to edit comments freely was left unchanged, and so far no spamming or other malicious activity has been noticed. Nevertheless, such vandalism or misinformation as that which plagues Wikipedia still remains a possibility, and solutions must be found. The most obvious response is a peer-review system which would harness the potentially destructive plurality of voices against itself, letting our readers judge a comment as useful of not. This could be further supplemented by giving each user the ability to filter annotations and thus block out those that they found misleading. Such filtering would be based upon a tagging system that could also include tags for things other than authors – romantic quotations, for example, or useful audition speeches. As well as blocking out undesirable comments, a system of filters would also be essential in the event of an explosion of user contributions to Open Shakespeare, since the site has the capacity to store far more comments, glosses and annotations than could ever fit inside even the weightiest variorum edition.
These ideas about tagging and filtering are just that, ideas only, and remain to be implemented in the hopefully near future. To do so we could use the aid of more programmers, for although the concept of Open Shakespeare appeals to many, it is a sad truth that many of those it appeals to, such as myself, have little in the way of programming skills. Finding dedicated volunteers to continue the project’s growth is our next great challenge, and we intend to do so by remaining flexible to all suggestions for possible improvements. One area for future expansion is the incorporation of audio or video recordings of Shakespeare’s works into our site; another possibility concerns the publication of an ‘Open Shakespeare Edition’ of Shakespeare’s works. Already, a basic version of Hamlet exists on lulu.com. However, we would like to create a full critical edition to print on demand, which would contain as many of the comments accumulated on the play so far as the user chooses. In such an initiative the project will have come full circle: having taken Shakespeare’s works and subjected them to computerised tools in a virtual world, it will then have produced a version of Shakespeare’s text to be read and enjoyed in the real.
This connection with real life is important, since it underlies the great interest many of us still have in Shakespeare, centuries after his death. Occasionally, when reading one of this poet’s plays, we recognise a part of ourselves, or something so alien that it negatively illuminates our own existence. The play becomes personal, and the effect is all the more dramatic when we take part in a production of it. The Open Shakespeare Project is also about making Shakespeare personal, and, more than this, recording that intimate relation. This poses problems, largely those of free speech and of data management, but it also inspires something vital to any study: the feeling that you are in a position to make a difference. As Lady Macbeth, one half of a rather different marriage and talking of a rather different task, has it, personal involvement is everything:
…screw your courage to the sticking-place,
And we’ll not fail.
Thank you for listening.