There are approximately 35,000 paintings on display at the Louvre. Most tourists feel overwhelmed by the prospect of viewing them. Indeed, if one were to devote 30 seconds to each painting (not counting the time it takes to walk between them), it would take more than twelve twenty-four hour days to visit them all.
Yet if one were to design an algorithm capable of identifying all 19th century British oil paintings, a visit to the museum would feel quite different. Indeed it would be much more efficient, as one could effectively skip those works irrelevant to one’s research. Providing the algorithm was correctly designed, one could apply it to London’s Tate Britain: the trick is in knowing which works speak to you—in my case, J. M. W. Turner’s Decline of the Carthaginian Empire.
Thus the critic’s quantitative search is predicated on her qualitative experience of the work. Catalyzed by emotion, she begins to wonder what techniques Turner uses to create this experience. In this way, all criticism stems from a passion for the work; a desire to uncover what moves us about a painting, poem, or sculpture; and how that work’s formal features elicit our response.
Automating the detection of rhetorical figures in Shakespeare, however, removes wonder from the equation. It forces us to work backwards: from quantitative to qualitative; from form to content:
583 (‘is’, ‘her’, ‘tomb’, ”, ‘what’, ‘is’, ‘her’, ‘burying’)
The earth that’s nature’s mother is her tomb;
What is her burying grave that is her womb,
And from her womb children of divers kind
We sucking on her natural bosom find,
Many for many virtues excellent,
None but for some and yet all different.
Here, the Zeugmatic program has directed our reading to act 2, scene 3 in Romeo and Juliet, highlighting subtler anadiploses with indirect (as opposed to direct) repetitions. In the above passage, “tomb” is paired with “grave” and “mother” with “womb”: a synonymic chiasmus that formally illustrates the generative, as well as destructive, forces contained “in man as well as herbs” (II.iii.28).
It is this quantitative / qualitative reversal that necessitates a completion of what we call the ‘critical cycle.’ The goal is not to build another This is Just to Say bot, but rather to say something about human experience; to go beyond the output files and understand how Shakespeare’s poetic form produces meaning. A tool is only a tool so long as it enables the artist to build something: a better literary critical essay; an improved history of rhetoric; a book on Shakespeare’s syntax—without taking the human out of the humanities.
What programming has to offer is a new perspective on poetics. Teaching the computer teaches us everything we don’t know about Shakespeare’s poetry. Programming Shakespeare forces us to distinguish between ordinary and poetic language; to define the boundaries of a clause. Like a child building a sandcastle, the computer sifts through ordinary language for rhetorical gems.
❧ What We Should Say When
The question of ordinary language is particularly complex. Intuitively, one might conceive of language as a game of chess, with certain rules governing its players—or as J. L. Austin put it, what we should say when (qtd. in Loxley 181). Such rules are constitutive, in that one plays ‘chess’ by moving one’s bishop diagonally, just as one speaks ‘English’ by placing one’s adjective before one’s noun (a stormy night in England; un soir orageux en France). Yet just as we cannot point to an original constitution or book of laws governing everyday speech, we cannot use words to mean anything without appealing to some collective agreement within our linguistic communities; some conception of normativity. In this way, ordinary language is “not an aggregate of individual appellations; it is the very attunement in signification that gives us language in the first place, and that enables us to have a shared world in which to live” (Loxley 32). Comparing language to a codified game like chess is problematic, in that grammatical rules are not the equivalent of “basic algorithmic rules for generating a mathematical series,” and so literary works1 cannot “furnish us with an exhaustive calculus or set of algorithmic procedures” (Loxley 34; 33). To appropriate a line from Wallace Stevens, one speaks of the sea using ‘ordinary’ language as one improvises on a piano.
Thus the rules of language (unlike the rules of chess) are subject to rupture or change, and the ordinary / poetic binary becomes an oversimplification and, by and large, arbitrary distinction. The reason that Beowulf and Shakespeare still resonate is because metaphor is embedded in our collective unconscious. Poetic wisdom, as Giambattista Vico states, is the “first wisdom of the gentile world” (316).
For when we wish to give utterance to our understanding of spiritual things, we must seek aid from our imagination to explain them and, like painters, form human images of them. (Vico 319)
One still finds the vestigial “leg of a table” and other dead metaphors swept under the rugs of everyday conversation: it is possible, in English, to have ‘seeds of doubt’ while ‘fishing for compliments’ or listing the ‘branches of government’ in the ‘body of an essay’. A man might eavesdrop on a conversation in order to ‘catch her name’ as though it were a butterfly. As we row toward the horizon that separates the ordinary from the poetic, we realize that the sea of language goes on forever—the dark bar dividing the vernacular from the verse becoming clearer, “as if the sediment in an old wine-bottle had sunk and left the glass green” (Woolf 3). As Samuel Shaw writes:
…there is a certain Vein of Rhetorick running through the Humane Nature . . . which infects all their Sentiments, and modifies all their Actions . . . That the most illiterate people, in their most ordinary communication, do Rhetoricate by Instinct, as well as others do by Art. (qtd. in Adamson 11)
And so we leave questions of intention on the shore: for whether artists2 rhetoricate instinctively or laypeople do so artfully, the input / output for the Zeugmatic program is the same. Leaving the gallery, we should examine ourselves rather than our poets to determine what is ‘poetic’ and what is ‘ordinary’; which patterns are exceptional and which are residual and cease to give effect.
Thus the quantification of poetry is reductive only if we fail to augment it with our criticism. For what else is poetry but the condensation of imagery? A metaphor “is most praised when it gives sense and passion to insensate things,” writes Vico, and thus “every metaphor so formed is a fable in brief” (319).
O fan of white silk
Clear as frost on the grass-blade
You also are laid aside. (Pound 95)
Poetry is expansive, explosive—like mathematical formulae. As the Cambridge mathematician G. H. Hardy notes, numbers are as mellifluous as words:
The mathematician’s patterns, like the painter’s or the poet’s, must be beautiful; the ideas, like the colours or the words, must fit together in a harmonious way. (Hardy 84-5)
It is this “primacy of pattern”—as Stephen Ramsay calls it—which unifies all disciplines (xi). Numbers and atoms arrange themselves in series and chemical compounds, like the overlapping grammars of Indo-European languages. A man who studies the earth, seas, or stars studies the outer universe; the critic, a purveyor of words, studies the universe within. Both are questing for truth in life; both articulate the beauty and subjectivity of their human experience. Their goals are one and the same, whether they begin with quantitative observations
Or qualitative musings:
What might have been and what has been
Point to one end, which is always present. (Eliot 13)
To Boldly Go Where No DH Project Has Gone Before
It is a truth universally acknowledged that all scholars in possession of a good idea must find a means of executing it. Part of the challenge of digital humanities projects is that they employ traditional humanities methods to tackle large-scale science problems that require large-scale scientific methods: i.e., literary labs. The Zeugmatic’s objectives are long-term: the project could last anywhere from 5 to 30 years, with multiple research teams scattered across the country from Victoria to Waterloo. The Zeugmatic, like so many other DH projects, is part of a bigger conversation: one in which computer scientists, linguists, and literary critics meet in a room to dissect Shakespeare’s works, using tools from their respective disciplines. The Renaissance Rubik’s Cube is not a design fiction, but rather a metaphor for the multidisciplinary nature of this project: a multicoloured visualization that unites seemingly opposed arts and science methodologies.
❧ Works Cited
Adamson, Sylvia, Gavin Alexander and Katrin Ettenhuber. Renaissance Figures of Speech. Cambridge: Cambridge UP, 2007. 11. Print.
Eliot, T. S. Four Quartets. New York: Harcourt, 1971. Print.
Hardy, G. H., and C. P. Snow. A Mathematician’s Apology. Cambridge: Cambridge UP, 2012. Print.
Loxley, James. Performativity. London: Routledge, 2007. Print.
Pound, Ezra. “Fan-piece, for her Imperial Lord.” Imagist Poetry. Ed. Peter Jones. Oxford: Penguin, 1972. Print.
Ramsay, Stephen. Reading Machines: Toward an Algorithmic Criticism. Urbana: U of Illinois P, 2011. Print.
Shakespeare, William. Romeo and Juliet. Ed. Jill L. Levenson. Oxford: Oxford UP, 2000. Print.
Woolf, Virginia. The Waves. Ed. Gillian Beer. Oxford: Oxford UP, 2008. Print.
Vico, Giambattista. “The New Science.” Critical Theory Since Plato. 3rd ed. Hazard Adams and Leroy Searle, eds. Boston, Mass.: Thomson/Wadsworth, 2005. 313-21. Print.