Internet Shakespeare Editions

Dancing Chips: Computers and Shakespeare's Text

This paper was originally presented as part of a session on "The Electronic Text as Tool in Research and Teaching" at the meeting of the Shakespeare Association of America in 1988. Other papers in the session were presented by Ian Lancashire, Hardy Cook, and Milla Riggio. image

Recent meetings of this association have offered some dazzling displays of electronic technology, featuring performances like the pioneering work of Peter Donaldson and his MIT project. Few of us, however, in our chronically budget-starved institutions, can afford the cost of the equipment and maintenance such projects require, and it is probable that even fewer of us are able to make the substantial investment of time required to learn how to use the gadgets. But in the last few years, computers have become not only more powerful, they have become easier to use, and most of us have, however reluctantly, chosen at least to switch from the quill to the keyboard. Our aim in this session is to demonstrate the increasing number of ways in which scholars and teachers with little more than a basic PC can take advantage of the electronic text.

The titles of academic papers in our discipline tend to conform to a number of well-established conventions: a good title should have a colon in it; it should also include, if possible, an apt quotation; and it can tease at least as much as it informs. My own title conforms to these expectations, with the additional bonus that the quotation involves a rather bad and anachronistic pun. I found the quotation as I was surfing my electronic concordance in search of inspiration. When I saw the phrase "dancing chips," I found the association of the joyous activity of dancing with an pun on the computer chip irresistable as a starting point for an enquiry into the effects that electronic media are having on Shakespeare studies: too often we forget that computer chips can dance, as they provide us with multimedia resources, games, and new ways of playing with the text. Under the influence of the title the concordance presented to me, this paper has taken a rather different shape from the one I had originally intended; thus part of the paper is a metanarrative about how the electronic media shaped it.

The origin of the quotation is not obvious: it comes from Sonnet 128, where the speaker describes his loved one, in this instance female, playing on a keyboard instrument. The central conceit is a familiar one: the speaker would like to change his state as observer with the favoured instrument that has such close contact with the object of his adoration:

How oft, when thou, my music, music play'st. . .
. . . . . . . . . . . .
Do I envy those jacks that nimble leap
To kiss the tender inward of thy hand,
Whilst my poor lips, which should that harvest reap,
At the wood's boldness by thee blushing stand!
To be so tickled, they would change their state
And situation with those dancing chips.
(1, 5-10)

Sonnet 128 has not figured largely in studies of the Sonnets; many critical discussions of them omit it completely, while others tend to dismiss it as a slight work, tempted by its subject-matter to describe it as a "five-finger exercise" in the tradition of courtly compliment. There are, however, at least two points of interest in the poem: the unusual use of the word that initially attracted my attention, "chips," and the play on the word "jacks."

Although it seems a common, workaday word, Shakespeare uses "chip" or words derived from it only four times. The Oxford English Dictionary cites this one use in Shakespeare's sonnet to justify a definition of a meaning of "chip" as "applied to the keys of a spinet or harpsichord" (sb. 1.4); the editors of the dictionary decided that Shakespeare recorded a "meaning," rather than using a metaphor.

It is at this point that I am able to invoke some of the new resources offered the scholar through the electronic medium, to see if they provide further enlightenment than is offered by the OED. I connect first to the on-line Early Modern English Dictionaries Database (EMEDD), [Note 1] edited by Ian Lancashire and available on line on the Internet from the University of Toronto. I submit a query to see how the word "chip" is used in these early dictionaries, and find that they use the word some fifty times, all in a context that makes clear that they refer to the simplest and most literal meaning of the word, a shaving of some material, or the action that creates it.

A similar search of LION -- Chadwyck-Healey's LIiterature On Line database [Note 2] -- reveals that chips were quite often mentioned by early poets: sometimes, of course, they are again simple, literal chips of wood, but more often chips are worthless irritants, not worth a chip, or differing not a chip. None of these chips is associated with a musical instrument. But the most interesting thing about the result of a search in the LION database emerges when we separate from the list the occurrences of the word when it is used as a rhyme. Of 50 "hits" in the database, nine are used as part of a rhyme scheme, and of those nine occurrences, six, like Shakespeare, rhyme "chips" with "lips." The evidence from the electronic record suggests that Shakespeare was not alone in looking for a rhyme for "lips."

The other play on musical words in the poem involves the "saucy jacks" that "that nimble leap/To kiss the tender inward" of the player's hands. Back in 1934, Edward Naylor argued [Note 3] that Shakespeare was minutely observant of the instrument, and that he is not describing his mistress playing the virginal, but correcting the mechanism, for a "jack" is the wooden device that plucks the string with the plectrum attached, rather than a key. In the LION database, only Margaret Cavendish plays musical jacks, in a poem called "Nature's Musical Instruments":

The Head unto an Organ I compare,
The Thoughts, as several Pipes, make Musick there;
. . . . . . . . . . . . . . .
The little Virg'nal Jacks, which Skip about,
Are several Fancies that run in and out.
LION: Cavendish, Poems, and phancies (1664) 1-10.

We may perhaps be forgiven if we suspect that Cavendish knew more of the technical details of the virginal than Shakespeare. For all the ingenuity of Naylor's explanation, Shakespeare's sonnet does seem to assume an equivalence between the jacks and the chips or keys on which the player's fingers are "walking." Interestingly enough, there is some non-electronic support for this assertion: Sonnet 128 is one of several that exist in variant versions; there is a manuscript in the Bodleian Library which records what is either an earlier draft of the sonnet, or a later, poorly remembered, version. Whatever its provenance, the Bodleian copy substitutes "keys" for "jacks" on both occurrences of the word. [Note 4]

So where do we stand after this brief electronic exploration? It has produced some interesting, if inconclusive, evidence of the way that two words in the poem were used by others in the period. If we wish to argue that the sonnet is mechanical, there is evidence that Shakespeare may have chosen the image of the chip for the sake of the rhyme, as so many others did. If we wish to argue that Shakespeare was forcing an overly-ingenious play on words by making the jacks in the virginal a synonym for the very different keys on the instrument, we can do so, since the sonnet appears to be alone in suggesting an equivalence. But if, in contrast, we choose to argue the originality of the sonnet, we can point out that, apart from the musically exact lines of Cavendish, Shakespeare is the only poet who makes any play at all with the technical terms of the virginalist, and his are the only chips that dance.

My point in this deliberately rather basic use of the power of an electronic database for pattern-matching is equally basic, but no less important: the power of the computer, linked to extensive and increasing resources of machine-readable texts, permits us to enhance the critical techniques we already have: the OED is extended by the resources of the EMEDD and LION, and it becomes possible for the researcher to assume the role of the first readers for the OED, with more opportunity to compare and extract meaning from context. The opportunity is suddenly there for the reader to explore and to play with the text, using LION as a kind of computer game, a linguistic MYST or RIVEN, with the distinction that there is no final goal and that all explorers are winners. It is also important to realise that the resources are as readily available to our students as to scholars, since the texts can be searched without the benefit of research grants for travel to major libraries.

We are conditioned to think of the computer as rigidly binary, mechanical beyond anything produced by the most minor of the sonnetteers; what I am demonstrating here is that the computer is capable of providing an element of serendipity, as we are given the opportunity to play word games with the text, and to read works we would not otherwise have ready access to. The question remains: What games shall we play? How shall we ask the chips to dance? My brief exploration of the sonnet indicates that we can ask all the old questions editors have tried to answer in their commentary, with the opportunity for a fuller response because the computer remembers everything it has read, and allows us to retrieve that memory far more efficiently than even the most widely-read human.It remains to be seen whether the computer will allow critics and textual scholars of Shakespeare to ask genuinely new questions.

At this point, I am reminded of the remarkable changes in scientific thinking about a host of seemingly unconnected phenomena that have been brought about by the mathematics of fractal ("chaos") theory, a mathematics that was effectively impossible to work with before the advent of the computer. A recent cover story in the New Scientist was headed "Why earthquakes are like businesses are like magnets are like flocks of birds are like bacteria are like traffic jams are like ants..."; [Note 5] to this list we can add new ways in which the functioning of the brain is being understood, and yet more theories to explain the extinction of the dinosaurs. The computer has made it possible for scientists in many disciplines to ask questions that previously were not only unanswerable, but effectively unthinkable.

Shakespeareans of all persuasions are likely to associate computer analysis of texts with the kind of work that Donald Foster has done with his Shaxicon database, specifically with his attribution of the "Funeral Elegy" to Shakespeare. But attribution studies ask a very old question indeed; Foster's conclusions and methods have both been challenged, but whether we agree or disagree with his findings concerning the "Funeral Elegy," or his hyphothesis concerning the parts Shakespeare acted in his own plays, the fact remains that in asking how Shakespeare used what for him were "rare words," Foster has been able to ask a question that would have been beyond the industry of even a Caroline Spurgeon to answer. As machine-readable texts of non-Shakespearean literature become available, scholars will increasingly have the opportunity to use the power of the computer in tracing vocabulary and lexical patterns through a much larger corpus than any one reader can master; and as this additional material becomes available, it may become possible to radically reshape the nature of traditional source studies, learning not only what books Shakespeare read, but more accurately when he read and re-read them, and when others were reading or listening to Shakespeare. There is a whole network of interconnections to be discovered, of trails to be followed, of games to be played with the text, of dances to be danced as we discover links that are potentially trivial, potentially profound.

If all this begins to sound cloyingly glowing and evangelical, I can assure you that I am about to apply a partial corrective. In my quest to see how chips and jacks were employed in early modern texts, I referred to two quite different kinds of resources, the EMEDD and LION. Apart from the fact that it is presented in electronic format, the EMEDD is very much like a traditional scholarly resource: it is edited by a distinguished scholar, and published at a senior University site. The texts in LION, on the other hand, are quite deliberately generated by a largely non-scholarly process. LION texts are "double-keyed" -- the works are typed in by two different typists, and the resulting texts collated by someone skilled enough to choose which is correct. In most cases, the typists did not understand the texts they were working with.

In practice this means that Chadwyck-Healey employs below-minimum-wage typists from third world countries to create the texts we search. Now, we may or may not find this practice ideologically repellent, but whatever our reaction, there is surely something disconcerting about the fact that the basis for our sophisticated computerized and critical work is a text most probably created by someone who did not understand it. There is, indeed, evidence that ignorance of the language can be an advantage in the accurate transcription of texts. [Note 6] Non-English typists will not see words as units of meaning, and will focus on the individual letter as an object of limited signification, with the theoretical likelihood that two typists will be unlikely to make the same error.

Reviews of the LION database, however, make it clear that the method is fallible: in a typical early printed work, there will be a number of type forms that are ambiguous, especially through accidents of inking, foxing of the paper, the inadequacy of the facsimile used, and so on, and there are many occasions when both typists will mistranscribe the same letter; the raw data of the LION texts cry out for the intelligent agency of a scholarly editor.

In an ideal world, we would have access to accurate and scholarly machine-readable versions of all the printed and manuscript works that Shakespeare might have read, and a substantial body of additional materials as well. But if we are to achieve this end, we will have to work to change attitudes in the discipline. Those of us on each end of the administrative hierarchy in our institutions will realize that decisions about tenure, promotion, and salary, are made on the basis of largely unarticualted assumptions. A CV is a text, and like any text the way it is read is historically constructed by a cluster of assumptions; we are all aware that publications are ranked according to a system of what less tactful disciplines call "value added": today, at the top of the scale might be a critical or theoretical book published by a major university press; editions rank somewhat lower, and have a sub-hierarchy of their own.

It was not ever thus. In his introductory remarks to a conference on computing in the Humanities (1989), Northrop Frye commented:

Just as societies have to go through a food-gathering stage before they enter a food-cultivating one, so there had to be a stage of gathering information about literature that might be relevant to it, even when there was still no clear idea of what literature was. . . The period of literary scholarship, which was dominant until about 1935, is sometimes called the Wissenschaft period, and its great scholars amassed an awesome amount of information. [Note 7] image

Frye may have been more confident than many of us of what literature really is, but his analysis is helpful in pointing not only to the renewed necessity for food-gathering -- the generation of high quality machine readable texts -- but to the clear implication that food-cultivation -- the performance of more abstract discussion of the "heart" of literature -- is to be considered an endeavour of higher civilization, and thus more deserving of tenure.

I do not advocate anything so simple or so doomed to failure as a return to Wissenschaft scholarship for us or our students. But I do think that there is room in the multi-textured, almost chaotic variety of Shakespeare studies to create a culture in which the process of creating good texts for the computer is recognised as something of real value. From my own experience, I can say that the exercise of creating diplomatic transcriptions of the Folio and early quartos of Shakespeare has made me vividly aware both of the limitations of the transcribed text, and of the value that is added to it by transcription. It is limited because it is necessarily selective in what it records; no matter how thorough the transcription, the original will always contain more information in analog form. On the other hand, the transcribed text can be indexed and searched in the way I have begun to explore in the earlier part of this paper. Even more interestingly, the transcribed text can be made available for other scholars to manipulate, and this modified text itself can then be made available to other scholars, in a collaborative spirit of cumulative research.

My argument is this: if we are to begin the long process of creating the large body of high quality electronic texts we would all benefit from in the discipline, we must revalue the activity of creating texts of high quality, recognising their value added whether they are diplomatic transcriptions, old spelling editions, or modern editions with the full panoply of apparatus. One way to do this might be to integrate more fully the creation of texts into the graduate curriculum; why not give our students the opportunity to learn about issues in textual bibiograpy "hands on," by transcribing, tagging, and editing a text from facsimile or microfilm, and making it available on the Internet? So much the better if the text is non-canonical literature, or a document of historical rather than literary interest. Once published, as our appointments, promotion, and tenure committees begin to follow the recommendations of the MLA and learn to discriminate in the electronic medium as they already do in the medium of the printed book, our students will enrich those vital CVs in the process.

The metaphor Frye used to distinguish between scholarly food-gatherers and food-cultivators does not fit this model very well. A more persuasive metaphor might be to distinguish between the farmers who provide us with raw texts, and the cooks who use them to serve up their critical dishes; I might go further and posit a more theoretical level of metacritics, those who write about cooking. I find this analogy more pleasing, in part because, while it acknowledges difference and interdependence, there is less sense of a class structure. After all, a good writer about cookery should be able to make an omlette, and it does not require a cultural revolution to suggest that even a senior chef might benefit from a spell in the garden.

And when we return from the metaphorical garden or kitchen to our offices and computers, where our "fingers walk with gentle gait" over the keyboard, the computer chips dance as they give us the opportunity to read and explore the text in new ways, to undertake a further quest in the search of meanings in Shakespeare and other writers.

The complete sonnet

Sonnet 128

How oft, when thou, my music, music play'st,
Upon that blessed wood whose motion sounds
With thy sweet fingers, when thou gently sway'st
The wiry concord that mine ear confounds,
Do I envy those jacks that nimble leap
To kiss the tender inward of thy hand,
Whilst my poor lips, which should that harvest reap,
At the wood's boldness by thee blushing stand!
To be so tickled, they would change their state
And situation with those dancing chips,
O'er whom thy fingers walk with gentle gait,
Making dead wood more blest than living lips.
Since saucy jacks so happy are in this,
Give them thy fingers, me thy lips to kiss.


  1. Available at [Back]
  2. Home page at [Back]
  3. Naylor, Edward W. Shakespeare and Music: With Illustrations from the Music of the 16th and 17th Centuries. London: J.M. Dent, 1931. [Back]
  4. Shakespeare, William. The Sonnets, and A Lover's Complaint. Ed. J. Kerrigan. Harmondsworth: Penguin, 1986, 355, 442, 446. [Back]
  5. "One Law to Rule Them All," New Scientist 156:2107 8 November, 1997, 30-35 (Cited from the cover page summary of the article). See also my article in EMLS 3.3 / Special Issue 2 The Internet Shakespeare: Opportunities in a New Medium, "Afterword: Dressing Old Words New." [Back]
  6. Charles Faulhaber, Evaluating Computer-Related Work in the Modern Languages: Draft Guidelines Prepared by the MLA Committee on Computers and Emerging Technologies in Teaching and Research (1993): [Back]
  7. "Literary and Mechanical Models." Keynote address presented at the at the Joint International Conference of the Association for Computers and the Humanities and the Association for Literary & Linguistic Computing, University of Toronto, Toronto, Ontario (June, 1989). 3-13 in Ian Lancashire, ed. Research in Humanities Computing 1: Select Papers from the ALLC/ACH Conference. Oxford: Clarendon Press, 1991. [Back]