|CHWP B.2||Steele, "TACT and the Explicit Structures of Shakespeare's Plays"|
I should like to turn now from the abstract to the concrete, and describe some of the ways in which I have found TACT useful in my own work with Shakespeare. As I said, I do not sit down with the intention of conducting a 'computer study' of a play, but rather I use the computer as a specialized tool to answer specific sorts of questions, and to defamiliarize the text in order to absorb it more fully. As a result, I cannot offer you an integrated set of connected examples, leading to a carefully orchestrated climax, but only individual examples which may suggest some of the uses which are possible.
For example, in re-reading Hamlet a couple of years ago (the horizontal way), I was struck by the fact that Laertes' incestuous concern with his sister Ophelia's sexuality is even greater than Hamlet's much-touted incestuous interest in his mother, Gertrude. Laertes' parting words to Ophelia, in particular, are uncomfortably explicit:
Now, it is pretty clear what Laertes means when he speaks of Ophelia's "chaste treasure", but his concern with protecting her ears seems just a little misdirected.
Any attentive reader of Hamlet can tell you that ears do crop up in a number of places. For example, Hamlet's father is murdered when Claudius pours poison in "the porches of [his] eares" -- something which the Ghost narrates to us early on, and which the players perform for us in dumbshow later. If you are a little more familiar with the play, you might also recall some notable lines, like Hamlet's "spleet the eares of the groundlings", or Gertrude's "these words like daggers enter in my eares". But I do not know anyone who could remember all 26 mentions of the word eare/eares in Hamlet. In fact, of the 418 uses of the word in Shakespeare, Hamlet contains almost twice as many as any other play.
Figure 6 reproduces the TACT Index [KWIC] Display for eare/eares in Hamlet Q2. In addition to the lines I have already mentioned, you might notice that at 2.2.389, the Player describes the amputation of Pyrrhus' ear -- and remember that Pyrrhus is an avenging son, much like Hamlet, Laertes, and Fortinbras. The ear, the passive organ symbolizing human receptivity and vulnerability -- to words and to poison -- has appeared in Hamlet from the opening scene to the final lines. The Ghost's description of his poisoning haunts the play, and, at 4.5.80, even Claudius inadvertently describes Laertes as not without "buzzers to infect his eare | With pestilent speeches". Modern critics would regard this as a Freudian slip, a trick of language which reveals Claudius' guilt as much as his many references to "the first corpse" or "the primal eldest curse".
The haunting motif of the poisoned ear persists in Shakespeare, well past Hamlet, where it first appears. For example, the image recurs as Iago encourages Othello's suspicion of Desdemona's fidelity:
In the late romance Cymbeline, too, when Posthumous suspects his equally chaste wife, Imogen, Pisanio the loyal servant demands,
In both cases, the poison poured into a man's ear is linked to suspicions of his wife's fidelity. And, to return to Hamlet, if we start to wonder just how Hamlet's father could have seen the poisoning occur, during his sleep, we may begin to see that his description of events really serves primarily to symbolize his own jealousy and resentment.
None of this will appear in the Oxford English Dictionary, nor does it appear in the footnotes of editions. But, using TACT to read Shakespeare vertically, we can get closer to the text and closer to the connotations that images and words might have held for its author. Because TACT makes it so simple to conduct such searches, too, it becomes a ready way to experiment, browse, and serendipitously stumble across intriguing patterns and evidence.
For another simple example, we could also try pursuing the image of the bed in Hamlet. Figure 7 includes a variety of interesting TACT displays for Hamlet Q2. Aside from Barnardo's reference to actually sleeping in a bed, however, most of the beds in the play are being used for more strenuous activities, and naturally it is Hamlet himself who speaks most of beds (as you will see from the left-hand Distribution display). Beds in Hamlet seem to be associated repeatedly with sexual violation, incest, and death, but not with rest or dreaming, as one might expect (remember, Hamlet Senior sleeps in his garden, not in the royal bed of Denmark). The right-hand Distribution display shows a gradual accumulation of references to beds from the first scene until 3.4, which is (of course) the scene in which Hamlet confronts his mother in her "closet". The collocations at the bottom of the page do not show much statistical significance, but it is striking that words like excrements, garbage, stewed, incestuous, graves, sweat, corruption, kisses, and wanton are all more closely related to beds than sleeping (which is the third-last collocate included in this excerpt from the list).
TACT can also help us study the distribution of partial words, like the prefix o're or over, which is used in more than twenty hyphenated compounds in this play. I suspect that this emphatic use of the prefix helps create the atmosphere of Hamlet, in which kings are overthrown, princes overruled, moral laws o're-stepped, and the spectator o'rewhelmed. The prefix itself suggests hubris, excess, and the failure of the infamous Elizabethan World Order.
It is also possible to compare the lexicons of each character in the play, using the various structures so laboriously tagged in the text. Figure 8 illustrates a simple but, I think, revealing example: the use of first person singular pronouns. It is hardly surprising that Hamlet speaks most of them -- after all, he speaks most of the words in the play, too. What is striking, is just how many more times Hamlet, Claudius, Polonius, Laertes, and Ophelia all use these pronouns. The colossal self-centredness of Polonius and Claudius (like most of Shakespeare's villains) was predictable, but the high scores for Ophelia and Leartes were perhaps a greater surprise. Ultimately, we would have to take into consideration the density of these pronouns in their speeches -- and ultimately TACT or a related utility may allow us to do this.
Figure 9 shows, similarly, the distribution of the exclamation oh/o among the characters in Hamlet. It is surprising to see how many people exclaim while singing, or dying, or crying, of course. It is also striking just how many of the central characters are portrayed in suffering in the play. If we chart the occurrences across the scenes of the play, we find rather spectacular concentrations of the exclamation in 1.5 (where the Ghost describes his murder to Hamlet), 3.2 (the play-within-the-play scene), 3.4 (Hamlet confronting Gertrude in her closet), 4.5 (Ophelia's madness), and somewhat less in 5.2 (the final bloody scene). TACT's collocation reveals some extremely strong connections: the number one collocate for oh/o is -- of course -- God, and surprisingly, number two is list (as in the Ghost's "list, list, o list!"). Of course, if you were interested in such things, you could also compare the use of the spelling oh with the spelling o between the various compositors of the quarto. (Shakespeare's preferred spelling, we are told, is the simple o, and 104 of the 107 occurrences of the word are spelled that way.)
So far I have described using TACT primarily with single words, or variant spellings of single words. It is, of course, also possible to create categories of words and to analyse the distribution of these categories between scenes, speakers, or stage directions, for example. Figure 10 displays the distribution of a category of theatrical terms, which serve to highlight metatheatrical moments in Hamlet. The words involved are listed in the footnote; high-frequency words like part/parts, act/acts and acting could not be included because of the software's current memory limitations. Also, it was necessary to explicitly exclude words spoken by the stage director, because directorial references to the players would have skewed the results.
The first distribution display shows a very clear concentration near the centre of Hamlet, with an earlier ripple and a final echo. When we look at the second display, across traditional scenes, we see that 2.2 and 3.2 have the greatest concentration of theatrical language, which is understandable, considering that Hamlet greets the players in 2.2 and the players perform 'The Murder of Gonzago' in 3.2. The striking surprise was 5.2, the final scene of the play, in which Hamlet and Laertes duel and most of the characters on-stage are left in puddles of blood. Theatrical language builds up in this final scene until Fortinbras concludes the play with his order to "Beare Hamlet like a souldier to the stage". Note also that Hamlet, Polonius, and Horatio get most of the dialogue involving theatrical terms -- and that Rosencrantz gets substantially more than one would expect for the amount he actually says.
Finally in figure 11, I have traced another verb, drinke, and derived forms throughout the play. (After all, the Denmark Shakespeare shows us in Hamlet is fixated on sex, sleeping, death, and drink.) The Index [KWIC] listing reveals many of the examples we probably remember from the play: Claudius promising to drink to Hamlet's health, Hamlet promising to teach Rosencrantz and Guildenstern to drink deep ere they depart, and Hamlet's bitter criticism of Claudius' late-night revelry. What we might not remember are Hamlet's vows to drink hot blood, to murder Claudius when drunk, or to drink up esill to prove his love for Ophelia. Most striking of all, fully half the occurrences of drinking occur in the final scene of the play, in which characters are urged to drink or not to drink from the poisoned cup.
I have tried to stick fairly closely to a single, familiar Shakespearean text, the 'good quarto' of Hamlet, but that has meant, unfortunately, missing some of the more interesting and complicated things I have been able to do with TACT and Shakespeare. For example, I recently charted variations in the printed speech prefixes throughout the first quarto of Love's Labour's Lost, and I think that I have been able to learn a good deal about Shakespeare's habits of thought while composing the play, as well as uncovering some additional evidence for revision and discontinuities in the writing process of that play. I have also examined a series of words in A Midsummer Night's Dream which constitute Petrarchan diction, and in so doing I have discovered strong connections between the language used by the Athenian lovers while under Puck's enchantment, and the language used by Bottom and his fellows in staging the lamentable tragedy of Pyramus and Thisby.
In general, though, I think that I have described the primary ways in which I use TACT to better understand a writer who would have been mystified at our interest, I am sure, not to mention mystified by the technology we are using. My title, "the whole wealth of thy wit in an instant", is drawn from The Merchant of Venice, 3.5.56. It suggests the way in which TACT allows us to draw upon the whole of Shakespeare simultaneously, reading vertically through one play or many plays, making connections which might never have occurred to us otherwise. But there is another side to the phrase: in context, the words are spoken by Lorenzo to the clown Launce. Lorenzo says,
It is not a compliment. But I think that the phrase neatly catches the double-edged nature of computer applications to literature. TACT gives us a wealth of evidence in an instant, more than we might otherwise accumulate in weeks, but like all software it is usually thinking literally, mechanically, and worst of all, it is always doing exactly what we tell it! The computer will help supply the wit, in an instant, but the user must supply the wisdom, and must rely on traditional scholarly methods to supplement a resource like TACT. Using this quotation without examining its context would be both foolish and potentially embarrassing. We must continue to read and re-read, horizontally and vertically, in print and on stage, combining methods born of modern technology with those of time-proven scholarship, if we ever hope to fully understand Shakespeare's plays, or ultimately learn something of significance from them.
 Hamlet Q2 C3v (1.3.29-33).
 Hamlet Q2 D3r (1.5.63).
 Hamlet Q2 G4r (3.2.10-11).
 Hamlet Q2 (3.4.95).
 The Riverside lineation for this line is 2.2.477; the lineation of the quarto differs (here and elsewhere) because of differing treatments of prose and verse.
 Hamlet Q2 L1r (4.5.90).
 Othello Q1 F3v (2.3.356).
 Cymbeline F1 aaa2v-aaa3r (3.2.3-6).
 Information for density, or 'relative frequency', is planned for an upcoming release of TACT.
 Hamlet Q2 D2v (1.5.22).
 Hamlet Q2 G2r (5.2.396).
 See my "'Leaden Contemplation': Ambiguous Evidence of Revision in Q1 Love's Labour's Lost", contributed to a seminar entitled "Shakespeare's Quartos: Text, Performance, Memory" at the 19th Annual Meeting of the Shakespeare Association of America, Vancouver, March 20-23, 1991.
 See my "Vowing, Swearing, and Superpraising of Parts: Petrarch and Pyramus in the Woods of Athens", delivered at the 14th International Patristic, Mediaeval, and Renaissance Conference, Villanova University (Pennsylvania), September 15-17 1989.
 The Merchant of Venice Q1 G2v (3.5.55-6).