This is the text of my paper for a seminar at the Shakespeare Association of America conference (April 2015), called Form, Complexity, and Computation.
This paper explores literary complexity as it manifests in rhetorical figures, or the patterns of repetition and variation that make language beautiful and memorable, and thus make it powerful. Figures have the advantage of being computationally tractable. My research team has a Python script that uses regular expressions to detect them — first in Shakespeare’s works, and then in a 400-play corpus (supplied by Martin Mueller) from 1576 to 1642. Below, I compare Shakespeare’s use of one figure to these broader habits of usage. I conclude that while Shakespeare’s use appears to be more nuanced, it is also more narrow in its ambitions.
First, let’s distinguish between figures (or schemes) and tropes. “A figure is a shape or form, the meaning of the Latin word figura as of the Greek term it translates, schema.” It is more linguistically artful and self-conscious than a trope, which operates more on the level of cognition than of language. A trope “is what we have when the thought itself is changed, and not only the pattern of its delivery.”[1] It “describes a pattern in which thoughts are developed, which can in turn, to good effect, be reflected in language,” Raphael Lyne writes. Among the more conceptual tropes he lists are metaphor, metonymy/synecdoche, catachresis/metalepsis, and onomatopoeia.[2] But even simple tropes like aposiopesis, or interruption, “could probe the sometimes necessary truncation of thought processes, or the value of leaving something unexplored” (69).
Broadly speaking, tropes are to thought what figures are to language. Yet figures can also be cognitively revealing:
Even something as mechanical as anaphora – [a figure] where the same words are repeated to begin successive clauses – might testify to some sort of exploratory or categorical structure, wherein thoughts are managed and perhaps also instigated.[3]
My goal is to turn that “might” into something more decisive. I will argue that a “mechanical” figure, complicated by repetition and variation, can testify to a speaker’s cognitive processes. It even tells us something about the transitions between shades of meaning on which arguments are built.
My chosen figure is not the list-like exploration that anaphora entails (“Some glory in the birth, some in their skill, | Some in their wealth, some in their body’s force”) — but instead the more complex sequence of ideas in anadiplosis, or the repetitions of words at the ends and beginnings of successive clauses. In isolation these can be quite simple (“Featured like him, like him with friends possessed”), but in sequence they can be more complex. They can be the linguistic structure that enables cognitive leaps.
Anadiplosis in sequence forms a figure called climax, also known as gradatio.[4] It is the figure that George Puttenham called “the marching figure.”[5] Consider this model example from As You Like It, Rosalind’s description of Celia and Oliver’s progression from acquaintance to love to marriage:
your brother and my sister no sooner met but they looked, no sooner looked but they loved, no sooner loved but they sighed, no sooner sighed but they asked one another the reason, no sooner knew the reason but they sought the remedy; and in these degrees have they made a pair of stairs to marriage.
Rosalind defines the progress of their love so as to make it seem inevitable. A striking feature of this example, and the reason I begin with it, is its self-description as a “pair of stairs”. It marches up the stairs from A (looked) to B (loved) to C (sighed) to D (the reason) to the remedy of marriage. “In these degrees” the lovers have moved incrementally from first impressions to the lasting impresa of marriage.
Speaking of degrees, let’s consider a second example from Ulysses’ speech on degree in Troilus and Cressida. When hierarchies break down, Ulysses argues,
Then everything includes itself in power,
Power into will, will into appetite;
And appetite, an universal wolf,
So doubly seconded with will and power,
Must make perforce an universal prey
And last eat up himself.[6]
(As I will do throughout, I’ve bolded the repetitions that make up Ulysses’ figure.) He extends the climax by reiterating its terms in his conclusion, to suggest that appetite is both provoked and reinforced (“doubly seconded”) by will and power.
Ulysses’s principled argument has the unsettling implication of stifling debate; he suggests that questioning authority (“the specialty of rule”) will lead to universal chaos. “Degree being vizarded, | Th’unworthiest shows as fairly in the mask,” he laments.[7] It’s a legitimate worry among both Trojans and Greeks, whose armoured or even unarmed men are frequently unrecognized or confused for others. Shakespeare reinforces this later in the same scene, when Aeneas cannot recognize Agamemnon by his appearance. Despite his obsequious descriptions by Nestor (“thy godly seat”) and Ulysses (“thy place and sway”), there’s evidently nothing about Agamemnon’s face that suggests his majesty.[8] Shakespeare’s warriors repeatedly cite the difference between seeing each other’s faces and seeing each other in armour, on the field of battle. “[T]his thy countenance, still locked in steel, I never saw till now,” says Nestor admiringly to Hector.[9]
Troilus and Cressida is a play about swapping and confusing characters, about pride and public perceptions, and about exemplarity. More immediately, it’s about people doing things for bad reasons, about making bad arguments and suffering their bad consequences. Helen herself, stolen from Menelaus, is a bad cause for international war. “All the argument is a whore and a cuckold,” as Thersites puts it, “a good quarrel to draw emulous factions and bleed to death upon.”[10] “I cannot fight upon this argument;” says Troilus, exasperated, “It is too starved a subject for my sword.”[11] Yet he reverses this position a few scenes later, when in the Trojan debate that parallels the Greek council with Ulysses’ famous speech, Troilus leads the arguments in favour of keeping Helen. “She is a theme of honour and renown, | A spur to valiant and magnanimous deeds” — even if not, consistently, Troilus’s own.[12] Those are his triumphant words after Hector abruptly concedes to this view, having made reasoned but futile arguments for releasing her.
Shakespeare’s play exhibits the bad arguments that lead to bad consequences, to Hector’s fall and Troy’s defeat at the hands of its fractious and diseased enemies. His use of gradatio in Troilus is part of this wider project. Paris’s description of Pandarus uses the figure to conflate love and lust, another recurring theme: “He eats nothing but doves, love, and that breeds hot blood, and hot blood begets hot thoughts, and hot thoughts beget hot deeds, and hot deeds is love.”[13] Troilus, after witnessing Cressida’s betrayal of her vows to be true to him, accuses her with a gradatio of his own: “If beauty have a soul, this is not she; | If souls guide vows, if vows be sanctimonies, | If sanctimony be the gods’ delight, | If there be rule in unity itself, | This is not she.”[14] Troilus forgets (as Cressida does in her farewell speech) that without Diomedes’s protection she will be “kissed in general,” like a common whore.[15] In both instances of gradatio, love is debased by association with the deeds that Cressida or Pandarus undertake.
Shakespeare often uses gradatio for specious arguments, to signal his suspicion with them. Consider Touchstone’s own use in As You Like It, to overpower the bumpkin William, his competitor for Audrey’s affections: “if thou never wast at court, thou never sawest good manners; if thou never sawest good manners, then thy manners must be wicked; and wickedness is sin, and sin is damnation.” Or Dromio of Syracuse in The Comedy of Errors, recoiling from the Courtesan: “It is written, they appear to men like angels of light: light is an effect of fire, and fire will burn; ergo, light wenches will burn.” Or Tarquin in The Rape of Lucrece, making his assault on the sleeping Lucrece: “His drumming heart cheers up his burning eye, | His eye commends the leading to his hand; | His hand, as proud of such a dignity, | Smoking with pride, march’d on to make his stand | On her bare breast, the heart of all her land.”
I could go on. Our Python script uses punctuation to split the sentences into clauses, and looks for token & lemma matches within four words of either side of those clause divisions. It found 18 instance of gradatio in Shakespeare’s texts. Suffice it to say that a character uses gradatio for good ends in only one of those 18 instances, namely Rosalind’s use that I cited at the start. Otherwise they are used for political expediency (e.g. in King John), for accusations (e.g. in both Richard II and Richard III), and for the self-justification and self-aggrandizement of fools or villains. (My moral judgement depends, admittedly, on whose ‘good ends’ we’re talking about.)
The point isn’t just that Shakespeare often uses gradatio for expedient arguments by characters of questionable moral status. Rather, it is that gradatio uses repetitions (A to B, B to C) to make a kind of argument in miniature (from premise A to conclusion C). And when you make an argument, you often have someone to convince. Even Rosalind is trying to make her beloved Orlando see that love at first sight can and should lead to marriage. Gradatio repeats the terms of your argument to make their progression seem natural and convincing.
Shakespeare is just more suspicious of its inevitability than others are. His characters use repetitions to reinforce something that should be strong enough without a figure to reinforce it. When Touchstone or Paris uses a figure to underscore his claims, it’s because those claims are specious to begin with. They use gradatio to draw unnatural conclusions from natural premises — until everyone in Arden frets about their damnation, and everyone in Ilium eats hot doves.
But is Shakespeare’s problematizing typical of gradatio’s usage writ large? The caveat is that while 400 plays are more ‘large’ than Shakespeare’s 38, they are just a fraction of the texts printed from 1576 to 1642, which are themselves a fraction of the unprinted texts written in those decades. But we can still ask how Shakespeare’s characters’ usage compares to broader patterns of published dramatic usage, immediately before and long after his contributions.
The short answer is that Shakespeare’s gradatios seem more nuanced, but this may be due more to my familiarity with the contexts in which he uses it in plays like Troilus or As You Like It. The Python script recognizes word-patterns in texts that I have not read, and (moreover) do not intend to read — not in full, certainly, or with the same attention to their characters’ motives that I give to Ulysses or Rosalind. My goal is to describe the kinds of cognitive processes gradatio can reveal, not to delineate every variety of those processes.
As I say, the script found 114 instances of gradatio in these 400 plays. They range from the mundane (“What I have said, I have said, and what they have | Done, they have done, by this time”) to the seemingly bizarre (“I can lie close and see this, but not see, | I am hoary, but not hoary as some be”). There is an endearingly accidental quality in examples like these.
But the transition from one anadiplosis to the next usually requires a more artful, deliberate shift of meaning. Some have an overt sense of growth: “Minutes are hours there, and the hours are days, | Each day‘s a year, and every year an age.” Others expand from a literal to a metaphorical or even a metaphysical meaning, like when Touchstone’s “good manners” takes first a courtly and then a moral meaning. Consider this example from a 1578 royal entertainment for Elizabeth: “From slumber soft I fell a sleep, | From sleep to dream, from dream to deep delight.” Or this, from Fulke Greville’s 1633 Senecan tragedy Alaham: “Time fashions minds, minds manners, manners fate.” Both examples expand from an initial, literal state to one that requires more imagination to conceive or express. The latter two are mental states (delight) or conceits (fate) imposed on something physiological (slumber) or external (time), thus expanding their meaning.
Sometimes this expansion strains credulity. According to this gradato in Nathaniel Woodes’ Conflict of Conscience (1581), trouble can lead to good health: “for trouble bring forth patience, from patience doth ensue | Experience, from experience hope, of health the anchor true.” Deriving this consolation feels tortuous, but it is one of those paradoxes — like the bracing effects of a good tragedy — that gradatio can reinforce.
Deception is another recurring theme, and in this sense Shakespeare feels typical. It is a short leap from metaphor to deception, because “the truest poetry is the most feigning” — to quote Touchstone one last time. But a number of the 114 instances have overt or indirect suggestions of deliberate deceit. Four come from Thomas Kyd alone. In The First Part Jeronimo, his lesser-known prequel to The Spanish Tragedy, Kyd puts gradatio in the mouths of his conspirators: “Be secret then, trust not the open air, | For air is breath, and breath blown words raise care.” Then comes this accusation in the sequel: “Now in his mouth he carries pleasing words, | Which pleasing words do harbour sweet conceits, | Which sweet conceits are limned with sly deceits.” Kyd also uses gradatio for self-deceit, as in the Portuguese Viceroy’s lament:
My late ambition hath distained my faith,
My breach of faith occasioned bloody wars,
Those bloody wars have spent my treasure,
And with my treasure my people’s blood,
And with their blood, my joy and best beloved,
My best beloved, my sweet and only son.
Deceptions of oneself or others suggest that either Kyd or his characters use gradatio to transition from their own actions and intentions to those who misconstrue or misapply them.
The viceroy’s redounding punishment has an air of circularity, a quality that also recurs in gradatio. Kyd’s Cornelia (another Senecan tragedy) uses this circular image of waters returning to their source: “Those floods to waves, those waves to seas, | That oft exceed their wonted bounds: | And yet those seas (as heavens please) | Return to springs by under-grounds.” A song in Jonson, Fletcher, and Middleton’s The Widow, or The Double Disappointment explicitly addresses this circularity:
How round the world goes, and every thing that’s in it,
The tides of gold and silver, ebb and flow in a minute:
From the usurer to his sons, there a current swiftly runs,
From the sons to queans in chief, from the gallant to the thief,
From the thief unto his host, from the host to husbandmen;
From the country to the court, and so it comes to us again.
These instances suggest that gradatio can sometimes signal circular or recursive thinking, not just linear progressions from thoughts to their consequences or implications.
Gradatio is also used to describe multiple possibilities arising from a single source, or different responses to external prompts. Consider this example of audience reactions to different dramatic entertainments, from Thomas Middleton’s Hengist, King of Kent (1619): “We are sir, comedians, tragedians, | Tragi-comedians, comi-tragedians, pastorists, | Humorists, clownists, satirists; we have them sir, | From the hug to the smile, from the smile to the laugh, | From the laugh to the handkerchief.” In this case the players are delineating the range of audience responses, presumably to satire (the smile) or comedy (the laugh) or tragedy (the handkerchief). The figure includes a sequence (smile to laugh), but includes other reactions to other genres.
So what do these cited examples from the 114 instances of gradatio reveal? That when gradatio is deliberate it can be used for a variety of purposes: deception and misdirection; despair and consolation; circularity and variety. The cognitive processes unfolding through these uses are more complex and multivarious than Raphael Lyne suggests. Many (not all) uses of gradatio are the intellectual habits of an age, captured by these strings of language use.
Comparing these habits and possibilities to Shakespeare’s suspicions toward bad arguments and specious reasoning naturally makes his usage — and the cognitive processes it reveals — look narrower, more focused. But if I had begun this essay with an extended reading of gradatio in The Spanish Tragedy, I would now be saying something similar about Kyd’s gloomy vision of poetic justice, and relegating Shakespeare to the 400. Such is the fallacy of context: it is merely more text.[16]
WORKS CITED
Culler, Jonathan. Framing the Sign: Criticism and its Institutions. Oxford: Basil Blackwell, 1988.
Davis, Alex. “Revolution By Degrees: Philip Sidney and Gradatio.” Modern Philology 108, no. 4 (2011): 488-506.
Lyne, Raphael. Shakespeare, Rhetoric and Cognition. Cambridge: Cambridge UP, 2011.
“Introduction: The Figures in Renaissance Theory and Practice.” In Renaissance Figures of Speech, edited by Sylvia Adamson, Gavin Alexander, and Katrin Ettenhuber, 1-11. Cambridge: Cambridge University Press, 2007.
FOOTNOTES
[1] “Introduction: The Figures in Renaissance Theory and Practice,” in Renaissance Figures of Speech, ed. Sylvia Adamson, Gavin Alexander, and Katrin Ettenhuber (Cambridge: Cambridge University Press, 2007), 6.
[2] Raphael Lyne, Shakespeare, Rhetoric and Cognition (Cambridge: Cambridge UP, 2011), 29.
[3] Ibid., 69.
[4] Gradatio has a rising movement like another figure, incrementum (A > B > C), but explicitly uses repetitions in this rise (A, A > B, B > C).
[5] For a survey of climax (or gradatio or incrementum) in contemporary rhetorical manuals (Wilson, Puttenham, Fraunce), see Alex Davis, “Revolution By Degrees: Philip Sidney and Gradatio,” Modern Philology 108, no. 4 (2011), 491-92.
[6] 1.3.119-24. My citations of Troilus are from the Arden 3rd series, ed. David Bevington.
[7] 1.3.78; 83-84.
[8] 1.3.31; 60.
[9] 4.5.197.
[10] 2.3.69-71.
[11] 1.1.88-89.
[12] 2.2.199-200.
[13] 3.1.123-25.
[14] 5.2.145-49.
[15] 4.5.22.
[16] Jonathan Culler, Framing the Sign: Criticism and its Institutions, (Oxford: Basil Blackwell, 1988), 93f.
