In the introduction to her book on Shakespeare's sonnets, Helen Vendler indicates that her analyses will only occasionally involve metrical commentary. This is a striking demurral on the part of the most skilled modern interpreter of lyric poetry, whose readings involve highly focused scrutiny of every other dimension of Shakespeare's verbal artistry. Vendler herself seems aware of how strange is the omission of prosodic commentary from her interpretations, since her own word for it is “regret”: “I regret the absence, except in occasional cases below, of metrical commentary.” And yet, given her stated reason for largely forgoing prosodic considerations, one understands the demurral; she claims “not yet [to have] found an acceptably subtle and yet communicable theory of scansion” (11). I share Vendler's view that there are subtleties in Shakespeare's management of his verse rhythms that even the best prosodies have not yet prepared readers to perceive, appreciate, and discuss. Therefore, although my ultimate aim in this essay is to identify one expressive purpose to which Shakespeare puts his verse rhythms—to generate a particular kind of unrest—my approach to that goal is along the pathway suggested by Vendler: through a refinement of existing theories of prosody that might provide an acceptably subtle and yet communicable instrument for the analysis of Shakespeare's practice in composing rhythmic verses (and that of other skilled poets, I hope).

The distinctive prosodic effects in Shakespeare's sonnets can best be perceived, described, and appreciated if one first delineates six or seven forms of syllabic prominence that are involved when someone reads a Shakespearian sonnet aloud. My focus throughout the essay is on reading aloud, and I on occasion prefer the word voicer to reader as a way of keeping focus on the audible production of rhythmic verses with the lungs, vocal cords, palate, tongue, teeth, and lips. The sonnets are perhaps not intended for oral delivery in quite the way that speeches from the plays are, but the prosodic effects on which I focus are most perceptible to someone preparing to read a sonnet aloud, who must make decisions about the exact level of vocal emphasis to place on particular syllables.

The several forms of emphasis involved in reading verse aloud can in some cases be differentiated from one another only by the finest of distinctions. In fact, a number of them are not absolutely distinct from one another; they overlap and interoperate in various ways, or they are the same quality of prominence in a syllable as viewed from different vantage points. The exact number of them, therefore, is in the end immaterial; the terminological and conceptual taxonomy I offer is intended to be clarifying and useful rather than theoretically exhaustive (though I hope it will be found reasonably precise and rigorous). Most of the forms of emphasis on which I focus are mentioned in various treatises of prosody, but some of them tend to be noted only in passing. In any case, I have nowhere seen all of them coordinated in one cohesive consideration.Footnote ¹ Doing so requires what may seem an extensive, painstaking, and extremely minute consideration of what is involved in delivering metrical verse—a highly analytic account of operations that the voicer in fact conducts rapidly, naturally, and generally with little conscious effort. But I believe the payoff for such minute analysis is considerable, for when the various forms of emphasis are first distinguished and then coordinated, it provides a rich and precise vocabulary for describing what Shakespeare is about, prosodically, in many of the sonnets (and a potential resource for discussing the verse of any poet). I am not offering an altogether new method of scansion; the insights I offer can be overlain on either traditional foot scansion or Derek Attridge's beat scansion, and I make use of both systems here, as one or the other helps me communicate a given point more clearly or efficiently.

Three of the forms of emphasis—lexical stress, beats, and what I call “sonic emphasis”—must be considered in conjunction with one another, as the very foundation of prosody for English verse. I start with sonic emphasis. It is one that most prosodists mention only in passing, in preparation for a discussion of lexical stress. This first form of emphasis is simply the precise degree of vocal energy involved in producing, or registered in hearing, a particular syllable. As George Wright says, “the English syllables we speak can be spoken with many degrees or shades of emphasis” (2). Attridge similarly observes that “[w]e actually utter, and hear, not just stressed and unstressed syllables, but a complex hierarchy of stresses” (Poetic Rhythm 27). There is no acoustical device that can measure this form of emphasis, because the perception of stress is actually based on several qualities of a syllable's sound—pitch, duration, amplitude, and quality—with no one of those invariably predominating (Attridge, Rhythms 62–63; Chatman 46–49). But if there did exist some way of measuring this level of emphasis, the various measurements would fall along a finely graded continuum. Consider the schwa sound in the phrase “Tom's a-cold,” where it can nearly vanish into inaudibility, by contrast with that same sound in an irately spoken, “I told you to buy a gallon of milk, not three,” where it may be by far the most emphatic syllable in the entire sequence. And, of course, every level of emphasis between those two extremes is possible as well. The values of what I am calling sonic emphasis fall along a range or scale.

When they mention it at all, prosodists tend to pass over sonic emphasis quickly. Here, for example, is the continuation of the quotation from Attridge given above: “We actually utter, and hear, not just stressed and unstressed syllables, but a complex hierarchy of stresses; however, in dealing with the major types of meter, it is enough to think in terms of two categories of syllables” (Poetic Rhythm 27). Wright similarly goes immediately from observing that we deliver syllables “with many degrees or shades of emphasis” to saying, “it seems likely that in most English speech we perceive two major levels of stress” (2). Prosodists pass over sonic emphasis for good reason. Poetic rhythms arise from the alternation of different levels of prominence in syllables not as dispersed along a finely graded continuum, but as grouped into two broad values, often called “stressed” and “unstressed.” Syllables at the higher end of the continuum of sonic emphasis are perceived or treated as stressed, and syllables at the lower end are perceived as unstressed, and it is the sense of syllables being grouped in these two broad categories that provides the building blocks for poetic rhythms. In fact, one can go a little further than Attridge. It is not just “enough to think in terms of two categories of syllables”; rather, the beat-style rhythms characteristic of languages like English can in fact arise only from the contrast of precisely two values of syllable. The syllables in an utterance might exhibit a broad range of sonic emphasis, but rhythms will emerge from those syllables as a result of some kind of mental sorting by which they are experienced in a binary fashion, as stressed and unstressed syllables becoming beats and offbeats.

This second form of syllabic emphasis, the broad grouping of syllables into one of two categories, is often connected to lexical stress. Sonic emphasis and lexical stress are of course not two alternative varieties of syllabic emphasis, but the same quality in syllables as viewed in either a gradated or a binary fashion. Two further clarifications regarding lexical stress are important. First, as prosodists are quick to note, verbal rhythms emerge from the alternation of syllables that are perceived in their immediate context as relatively more heavily or less heavily stressed. Lexical stress is not some absolute and fixed quality of each syllable in the language. Consider the syllable land in the word Iceland, where it is the relatively lightly stressed syllable, versus the word landed, where it is the relatively heavily stressed one. Then consider landed against land grant; in both cases, land is the heavily stressed syllable in the word, but in the second, it requires even more sonic emphasis as it jockeys for relative prominence with the also heavily stressed grant. “Stressed,” then, is always more accurately understood as “relatively more heavily stressed than neighboring syllables.”

Second, lexical stress works differently in polysyllables than in monosyllables. The usual way to address lexical stress is to note that polysyllabic words have built into them a specific pattern of relative emphasis, accentuation. This relative emphasis is what makes a given word the word that it is; it is how to tell CONtent the noun from conTENT the verb or adjective. Monosyllables, for their part, tend to register as more heavily or lightly stressed based on what part of speech they are: content words like nouns, main verbs, adjectives, and adverbs tend to be heavily stressed, whereas function words like articles, prepositions, pronouns, and auxiliary verbs tend to be lightly stressed. But note the slight difference in phrasing when one speaks of polysyllables and monosyllables: monosyllables “tend to be” regarded as stressed or unstressed based on their part of speech, whereas one syllable in a polysyllable word is more heavily stressed than the other, and failing to observe this results in a jarring distortion of the word (or, in some cases, a different word altogether).

When a series of words is arranged in such a way that the syllables making up those words manifest some relatively regular alternation of stressed and unstressed syllables, readers will begin to discern a rhythm in those words and, if reading the passage aloud, will find their voices falling in with the rhythm, cooperating to bring that rhythm out and sustain it.

This brings me to the third form of emphasis involved in the production of verbal rhythms, beats, which arise as voicers cooperate, physically and mentally, in the production of a rhythm they have begun to discern. Human beings are rhythmically predisposed and rhythmically cooperative. As Alfred Corn puts it, we are profoundly rhythmic creatures—“given room to walk without hindrance, we naturally fall into a precise rhythm as we move through space” (xix)—and, in his most lovely observation, “[b]efore an infant is born it develops a sense of hearing, and the first thing that it hears is the heartbeat of the mother, a heartbeat perceived in a regularly recurring sequence” (xviii–xix). Paul Fussell goes so far as to say that we have a “lust” for rhythm (19). He and many prosodists observe how strong is our tendency toward rhythm by noting that we will mentally superimpose on the ongoing tick tick tick of a clock a TICK-tock pattern instead (Fussell 18–19; Attridge, Rhythms 77; Chatman 25). This happens with no deliberate effort. If we will impose rhythm on a regular series of identical sounds, how much more will we cooperate to help bring out a rhythm in a series of sounds that is already (at least partly) in a TICK-tock arrangement: “From FAIRest CREATures we deSIRE inCREASE” (1.1).Footnote ² The heavily stressed syllables in the four disyllabic words fall so regularly in the TICK position, as it were, that with no deliberate effort whatsoever a reader experiences “from” as a tock and “we” as a TICK, so as to realize a very regular rhythm in this line. To be clear, the “we” need receive no enhanced sonic emphasis in order to function as a beat within this rhythm (in fact should not). The accented syllables in the disyllabic words do receive greater sonic emphasis—namely, the relative sonic emphasis provided by accentuation. And those lexical emphases fall into so regular a pattern as to draw “from” and “we” into their proper rhythmic role in that pattern: as an offbeat and a beat, respectively.

It is important to clarify the relation between lexical stresses and beats, for while they are to some degree independent, the latter do depend on the former. A rhythm, with the beats that make it up, arises in the first place only if some number of lexical stresses fall into a sufficiently regular pattern. Where that does not happen, there is prose, “words as they chanceably fall from the mouth” (Sidney 219). Once it does happen, though, those lexical stresses in which the rhythm is discerned become beats within that rhythm. Once the reader discerns such a rhythmic patterning, other syllables—so long as their lexical-stress level is not so discordant as to disrupt the rhythm—can serve as either beats or offbeats within the established rhythm. Seymour Chatman's distinction between “meter-fixing” and “meter-fixed” syllables is useful here (133). In the case of the line “From fairest creatures we desire increase,” the eight syllables in the four disyllabic words are meter-fixing. Because they must be pronounced with their conventionally established pattern of verbal accentuation, and because the relatively stressed syllables within them fall in a regular pattern of alternation, they establish the rhythm. The words “from” and “we” are meter-fixed and are drawn into the rhythm established by the other syllables. In an anapestic version of the same line, “From the FAIRest of CREAtures inCREASE we deSIRE,” it would still be the syllables in the disyllabic words that functioned as the meter-fixing syllables (and note that “we” would in this case be an offbeat). Although I have subjected the process to detailed analysis, readers actually perform this operation, in such passages as lend themselves to it, easily, naturally, and with little or no conscious effort.

The relation between lexical stress and rhythmic beat forms the foundation of most prosodies of English verse (whatever terms a particular prosodist may prefer for each of those phenomena: stress, accent, ictus, prominence), and systems of scansion are a set of signs for representing visually some combination of the lexical stresses that can give rise to rhythm and the beats and offbeats those syllables constitute once a rhythm has emerged. What is important to keep in mind is that sonic emphasis, lexical stress, and beat emphasis are in effect ways of describing the same phenomenon of syllabic prominence but from three different vantage points. Sonic emphasis and lexical stress are the same quality of vocal prominence of a syllable, just experienced either on a scale or as a binary, and lexical stresses (some of them anyway) become beats in passages that are rhythmic. In the passages where it happens, this shift from mere lexical stress to beat emphasis is the result of a complex psychological action, even if that action takes place rapidly and below the threshold of consciousness: hearing words pronounced with their standard pattern of lexical emphasis, the ear discerns the possibility of an overall rhythmicality, and the voice begins to make that rhythmicality more distinct. This can (and most agreeably does) occur without any major alteration in the pronunciation of words from how each of them would individually be pronounced in a prose context. And yet some syllables in the utterance now serve as beats in a rhythm; they are felt to have another kind of emphasis superimposed, as it were, on their lexical stress. And the whole passage is felt to be different in kind from passages where no such rhythmicality emerges: measured verse as opposed to prose.

Whether there is any alteration in the pronunciation of unstressed syllables that come to function as beats is an open question for prosodists, again given the fact that no acoustic device can measure what we experience as verbal stress, and that any alteration would be slight. Take the case of a “promoted” second syllable of three unstressed syllables, such as the “we” in “From fairest creatures we desire increase.” Is the beat that readers experience in the “we” just the equivalent of “TICK”: a merely mental superimposition on the second of three equivalent-level sounds to confer a greater degree of felt rhythmicality on the passage as a whole? Or does our cooperating with the rhythm elicit, as Chatman considers it might, a “slightly clearer and steadier pronunciation of the syllable” (124)? I cannot resolve this question, but I return below to the effect it has on the experience of verse, for it is at the heart of some of the distinctive prosodic effects of Shakespeare's sonnets.

Before I can take up such matters, there are several other forms of emphasis that require consideration. The first of these is rhetorical emphasis. All prosodies treat rhetorical emphasis, but there are important considerations regarding how this form of emphasis intersects with the other three I have discussed. Rhetorical emphasis is entirely dependent on the meaning of a particular passage. Normally unstressed words can carry rhetorical emphasis, as the article in “I told you to buy a gallon of milk.” Rhetorical emphasis is often contrastive, so corresponding vocal emphases will fall on a pair of words to point the contrast: “The pain be mine, but thine shall be the praise” (38.14). Sometimes the contrast is expressed by rhetorical emphasis that falls on a single word, because the contrasting idea is implied rather than stated outright. Consider “Oh, let my books be then the eloquence / And dumb presagers of my speaking breast” (23.9–10); the contrasting idea is “as opposed to my spoken words,” but that contrasting idea is implied. Certain passages can have pairs of contrasting pairs: “And see thy blood warm when thou feel'st it cold” or “Mine be thy love, and thy love's use their treasure” (2.14, 20.14). In such cases, one of the two pairs should probably be deemed primary and receive the vocal inflection signaling rhetorical contrast; a line with four separate words each receiving a special intonation risks seeing all meaningful emphasis and contrast muddied.

Several prosodists have observed that such rhetorical emphases are generally communicated through a change of pitch. As Attridge points out, this form of emphasis can fall on an offbeat syllable without meaning that syllable attracts the beat (Rhythms 227). Consider “O therefore, love, be of thy self so wary / As I not for myself but for thee will” (22.9–10). If the rhetorical emphasis here also made for a beat emphasis, there would be a trochee in the final foot of the line, which is extremely rare in the iambic pentameter line because extremely disruptive rhythmically. But one does not feel that degree of disruption in this case; “will” carries the beat. At most, the final four syllables constitute what Attridge would call a stress-final pairing (and traditional foot scansion a pyrrhic-spondee or ionic minor), “but for THEE WILL.” The reason that rhetorical emphasis can on occasion be independent of beat emphasis is probably that, of the four cues for stress mentioned earlier, rhetorical emphasis tends pretty consistently to use just one of them, pitch, and one may therefore be able to factor out that form of emphasis in the sense of the line's overall rhythmicality.

Quite often, of course, a poet will arrange for rhetorical emphasis to coincide with beat emphasis, for the word or words carrying rhetorical emphasis to fall in beat positions within the prosody of the line. “The pain be mine, but thine shall be the praise” (38.14) and “Oh, that you were yourself” (13.1) are examples. The beat, in such cases, simply helps underscore the rhetorical emphasis. The two uses of “him” in “[f]eatured like him, like him with friends possessed” (29.6) become deictics, rather than just standard pronouns, in part by virtue of falling on a beat; or, put another way, given that Shakespeare here wants “him” to serve as a deictic, it is appropriate that he arranged for each one to fall on rhythmic beats. But there is one important thing to observe about such cases that I have not seen mentioned in any prosody: namely, that this underscoring can occur either if the rhetorically emphasized word falls in the standard beat position or if the rhetorically emphasized word causes a metrical deviation—but with different degrees of underscoring in each of these two cases. In “The pain be mine, but thine shall be the praise,” “mine” and “thine” have their rhetorical contrast underscored by those words’ positions in a pentameter line where the beats are expected. The line “Mine be thy love, and thy love's use their treasure” operates differently. As readers work out the sense of the line, with its rhetorical contrast between “mine” and “their,” they realize that between the first two words here—each generally unstressed—in this case “mine” should take the beat rather than “be”; rhetorical emphasis in effect turns “mine” into a stressed syllable and a beat. What results is a permissible, in fact common, deviation in the iambic pentameter line: a trochee in the first foot. In this instance, then, it is a metrical deviation that underscores the rhetorical emphasis that falls on “mine.” One knows to put the beat on “mine” by virtue of the line's content: by noticing the rhetorical contrast between “mine” and “their.” Meaning and prosody, in such cases, are mutually constitutive. The semantically appropriate intonation of the line dictates where one sounds the beat, but that beat in turn assists the reader in delivering the vocal emphasis appropriate to the line's content. The reader works out both meaning and rhythm simultaneously and settles on a delivery of a line that maximizes their constructive interaction: “semantic and metrical demands are constantly in operation together” (Attridge, Rhythms 225).

In the hands of a skilled poet, these two ways of adding emphasis represent a powerful prosodic resource. Since deviations from an established rhythmic pattern call attention to themselves, a greater level of vocal prominence can be given to syllables involved in such deviations than to syllables that simply fall where the beat is expected. The poet in effect has two degrees of underscoring. Consider Sonnet 84, which makes the claim that if a poet were merely to convey “that you are you,” it would constitute sufficient praise of the beloved. Throughout the sonnet, Shakespeare arranges for the usually unstressed pronoun you to fall in beat position, and that nicely coincides with the meaning of the poem, since in this poem “you” is not just an address, but a theme. Line 13 opens with the word “you” carrying the beat in a trochaic reversal: “You to your beauteous blessings add a curse, / Being fond on praise which makes your praises worse.” This “you,” by virtue of being a metrical deviation, calls for a bit more vocal energy or punch in its delivery than the other instances of “you” throughout the sonnet, adding sharpness and force to the volta that occurs as one moves into the couplet. In line 13 of Sonnet 86, “when” falls in beat position, to begin the volta—“But when your countenance filled up his line”—and in line 14 the “Then” in “Then lacked I matter” receives an even stronger rhetorical emphasis, in effect supplying the word “then” with the enhanced meaning “in exactly those circumstances.” In the couplet to Sonnet 90, in line 13—“and other strains of woe, which now seem woe”—“seem” receives a rhetorical emphasis, the stronger in that it falls out of expected beat position, whereas in line 14 the rhetorical contrast word, “not,” gets its emphasis by falling in beat position: “Compared with loss of thee, will not seem so.” Notice how finely calibrated this all is. “Not” takes vocal emphasis off of the “seem” that follows it; it does that by being the second half of the rhetorical contrast with the “seem” in line 13. But relative to the thematically charged “seem” in line 13, it takes a lighter prosodic underscoring by virtue of falling in beat position, whereas the “seem” deviates from the expected meter.

Whatever terms a particular prosodist might employ for them, lexical stress, beats, and rhetorical emphasis represent the three main forms of syllabic prominence at the center of most treatments of English meter. I have begun to introduce two other kinds of emphasis involved in the delivery of rhythmic verses—two forms that are treated at best glancingly in other prosodies. First, I have used the term “vocal emphasis” (in places where the reader may have wondered why my earlier term “sonic emphasis” was not used). Second, one of the examples I used in my treatment of rhetorical emphases involved a form of emphasis that probably deserves a different descriptor. Let me begin with this latter form. Sonnet 84 is, as I mentioned, premised on the notion that if the poet could just say of the beloved “that you are you,” he would have said everything that needs to be said in the beloved's praise. I observed that, throughout the poem, Shakespeare consistently arranges for the word “you” to fall in beat position. This happens in line 7 and 9 and twice in lines 2 and 8. Most of these promoted instances of “you” do not carry full-fledged rhetorical emphasis, however. They are not contrastive and, with the possible exception of the one in line 9, they do not call for the change in intonation that generally marks rhetorical emphases. I would like to call this form of emphasis “thematic emphasis.” It need involve no special vocal realization, except perhaps something akin to Chatman's “slightly clearer and steadier pronunciation of the syllable.” And yet, there is something agreeable about a word, thematically important to a particular sonnet, carrying whatever form of extra prominence it is that beat syllables carry, even if it is only the superimposition of the TICK-tock pattern that the mind supplies.

Consider Sonnet 31, which tells the beloved that in him the speaker experiences all other people he has previously loved and lost. Since the relation between the addressee and these other people is the direct focus of the poem, it feels appropriate that the pronouns in the closing line—“And thou, all they, hast all the all of me”—fall in beat position. It would be perfectly fine for a reader to put some extra vocal emphasis on “thou” and “they,” to turn them into full-fledged rhetorical emphases. But even in a quieter rendition that did not mark the pronouns in any particularly striking vocal fashion, the thematic resonance they carry in this context would come through—in part as a result of whatever special enhancement inheres in beat.

This kind of thematic emphasis is not limited to promoted pronouns, but it can involve thematically pertinent words of any sort that fall on a beat (either where the beat is expected or as part of a metrical deviation). “To love that well which thou must leave ere long” (73.14). The thematic importance of the words “love” and “leave” is underscored by alliteration and near rhyme, but only further strengthened by virtue of their falling on rhythmic beats. In Sonnet 92, the speaker spends thirteen lines steeling himself against the possibility of the beloved's infidelity, but in line 14 another tormenting possibility belatedly occurs to him: “Thou mayst be false, and yet I know it not.” Someone delivering the sonnet might choose to put a rhetorical emphasis on “know,” a contrast with the implicitly stated known infidelity of the rest of the poem; falling on a beat might be a means of underscoring this rhetorical emphasis. But the fact that “false” falls on a beat feels appropriate as well. One does not give it any special vocal intonation, as one gives to rhetorical emphases. It has its lexical stress relative to its neighboring “be” and “and,” and it carries beat emphasis. Yet because infidelity is thematically the most important concept in the sonnet, and because this “false” represents the flattest and most direct articulation of that theme, it feels appropriate that this thematically crucial word gets whatever extra punch can ride along with a rhythmic beat.

Among the various forms of emphasis that this essay treats, thematic emphasis is the most subjective one, the least susceptible to rigorous identification. There is, moreover, a way in which this “emphasis” is not altogether distinct from what is necessarily involved in making verse be rhythmic or metrical in the first place. As I have observed, among monosyllables, it is content words that tend to receive relatively heavy lexical stress, so the very aspect of a word that makes it capable of carrying a beat is the same one that may make it capable of carrying thematic importance. In the line “Thy end is truth's and beauty's doom and date” (14.14), there is not some special artistry in getting thematically important words to fall in beat position separate from the artistry that was involved simply in composing a line of verse that would be experienced as rhythmical. Still, I would argue that there is an appreciable, if slight, rhythmic and thematic effect when words important to the meaning of a poem have their resonance enhanced by falling on a beat.

The key point for my taxonomy of emphases is that unlike rhetorical emphasis, this thematic emphasis does not involve any special intonation, any pronounced vocal marker of prominence. The words involved are understood, within the meaning context of the sonnet, to be significant, and they are experienced as having that significance confirmed by the rhythm of the sonnet. As Karl Shapiro puts it, “we feel the double weight of the right thing said with the right pressure” (32). We feel that.

At last I am ready to consider the most subtle and slippery form of emphasis, what I have been calling vocal emphasis—slippery because there is one key dimension of it that has not been brought into sharp focus by any prosody I have consulted. As a reader gets acquainted with a particular sonnet, that reader will develop a progressively clearer sense of the precise level of vocal emphasis to place on each syllable to bring out whatever the reader regards as the full meaning of the sonnet. The important thing to note is the mix of freedom and necessity in this process. On the one hand, settling on the exact level of vocal emphasis for certain syllables is an inherently (and even intensely) deliberative undertaking that may involve trying out alternatives and calibrating very precise degrees of relative emphasis to bring out every favored nuance of meaning. On the other hand, many of the emphases in a particular poem, at least their relative degrees, are established unalterably by the poem itself, and the reader discovers them in the sonnet instead of imposing them on it as performative choices. The lexical stress in two-syllable words, for example, is fixed. Similarly, once a rhythm is established, most of the syllables in a poem fall into a role of either beat or offbeat, and there is room for performative discretion in only a small percentage of cases. The same is true for many rhetorical emphases; any sensitive reading of a given poem will have to honor most if not all of them, as a dimension of the poem's basic meaning. This necessity for a large percentage of the syllables in a poem to receive a set basic level of vocal emphasis is something Attridge often adverts to, when he speaks of how easily readers can generally discern and cooperate in realizing a poem's rhythm on even a first, cold reading (Poetic Rhythm 43–44). In even a first reading of a poem in measured verse, while one may stumble on a metrical deviation here or there, by and large the emphases that make up a poem's rhythm are built in to the words and meanings of that poem. So when I call this form of emphasis “deliberate vocal emphasis,” I recognize that there are significant constraints on the performative license that can result from such deliberation.

At the outset of this taxonomy, I labeled the precise degree of emphasis on a syllable as “sonic emphasis,” at that stage of my analysis treating it as though it were simply a function of the language itself. “Sonic emphasis” and “vocal emphasis” are both ways of communicating that different syllables carry minutely different levels of prominence. Together the two concepts capture the mix of constraint and freedom involved in a voicer's assigning these degrees of emphasis: that it is the voicer whose breath and vocal musculature place, as an intentional act, a specific degree of stress on each syllable, but that in doing so, that voicer is responding to qualities built in to syllables and words as they appear in a particular meaning context.

Again, prosodists rarely concern themselves with the phenomena I am calling sonic and vocal emphases, for the reasons I have already mentioned: first, that there are no means of measuring the precise value of these emphases, and second, that rhythms are built not from these minutely differing values but from the binary of relative lexical stress. When a prosodist marks a line of verse, the scansion codes are in effect saying “all readers can reliably be expected to render this line with the following broad levels of relative syllabic emphasis, whatever precise weight a given reader might give to one syllable or another.” But although sonic or vocal emphases are not functional in the generation of verbal rhythms, they are functional in the voicer's experience of rendering—and especially of preparing to render—a line or passage or sonnet. It is not a phrase much in use anymore, but what “oral interpretation” used to capture is how, as one develops a richer and more precise understanding of the meaning of a poem, one develops a surer and more precise understanding of how to deliver the poem to bring out that meaning. As one prepares to deliver a poem, one is discerning the values of each syllable (given its function in the total rhythmic and meaning context of the poem), but then also resolving to impose those values on each syllable, as one recites, so that one's listener will draw the same meaning from the poem.

Returning to concrete examples, I will start with a very stark one. Sonnet 61 begins, “Is it thy will thy image should keep open / My heavy eyelids.” In a first, cold reading, one might assume that a sonic emphasis here falls on “will”: “Do you really intend to prevent me from getting sleep? Are you doing this on purpose?” As soon as one reaches lines 9 and 10, however, one comes to realize that the guiding concern of the poem is whether the speaker's sleeplessness is caused by something the beloved is doing, or whether he is responsible for it himself: “Oh no . . . / It is my love that keeps mine eye awake.” With that reversal in mind, the proper delivery of the first line should probably involve a contrastive rhetorical emphasis on “thy,” to prepare for the later balancing emphasis on “my.” “Will” can still carry a beat, but so does “thy,” with the first four syllables being a pyrrhic-spondee, ionic minor, or stress-final pairing, as one's favored prosody would have it. This kind of thing happens also on a smaller scale, with the correction often coming so quickly that readers may scarcely remember there even was a misstep. “Nature's bequest gives nothing,” one reads in line 3 of Sonnet 4, and on a first reading one might be inclined to put vocal emphasis on “noth-” (since it is lexically stressed and falls in the expected beat position) and demote “gives” as the second in a sequence of three stressed syllables. But when the ending of the line contrasts this “gives” with “lend,” one realizes that the line is structured to express a rhetorical contrast, and that one should put vocal emphasis, likely the strongest vocal emphasis in the line, on the word “gives,” treating the third foot as a trochee and the word “gives” as all the more heavily stressed by virtue of falling on a metrical deviation and by having to muscle “noth-” into the status of a nonstress: “NATure's beQUEST GIVES nothing BUT doth LEND.” It is Shakespeare who has arranged for the line to have the sonic emphases on which the reader eventually settles, with a feeling of having discerned the proper, intended rhythmic shaping of the line. And yet readers are agents too, cooperating with their voices, deliberately placing vocal emphasis in one spot rather than another to help a listener take in Shakespeare's intended contrast. Sonic and vocal emphases are never altogether distinct, but simply reflect the roles these two cooperating agents, verser and voicer, each has in realizing the precise sound contour of a verse.

Equipped with a taxonomy of the various emphases involved in a reader's voicing of metered poetry (lexical, beat, rhetorical, thematic, and then sonic or vocal), I can now with some efficiency consider the effects to which Shakespeare puts this prosodic resource. His fine calibration of these emphases simply allows the precision, nuance, and force of great poetry, Shapiro's “double weight of the right thing said with the right pressure.” But in addition to that, I find one effect, distinctive to the themes of Shakespeare's sonnet sequence, to which Shakespeare puts these resources: he sometimes places the various forms of emphasis in a particular kind of tension or strain to generate a prosodic unrest that correlates with the unsettled emotional state expressed or explored in the sonnet. In her analysis of Sonnet 137, Vendler says that “[t]o have invented a frantic discourse of unrest is one of Shakespeare's chief accomplishments in the Dark Lady sonnets” (582). Vendler is not alone in characterizing the sonnets in terms of “unrest.” Heather Dubrow finds an “unsettling coexistence of readings” in Sonnet 49 (157). Margreta de Grazia regards the concluding dark lady sequence as marked by “frenzied” repetitions (65). In a recent collection of essays on the sonnets, they are described as “tortuous” (Barret 139), “promot[ing] . . . a sense of unease” (Roe 83), and evoking “restlessness” on the part of the reader (Burrow 112). I argue that his management of verbal rhythms plays a large role in the “discourse of unrest” that Vendler and other critics find in Shakespeare's sonnets (and not only the “dark lady” sonnets, but the more agonized of the “fair youth” sonnets as well).

I begin with Sonnet 137, for it is this poem that prompted Vendler to make her more general claim about the dark lady sonnets:

In this sonnet, the speaker agonizes over why he looks past ugly truths despite, on some level, knowing full well what they are. Shakespeare expresses this complex psychological state in part by stating it early in the sonnet as a full-fledged paradox: that his eyes “see not what they see.” The expressions of that paradox throughout the poem involve key words that jockey for various kinds and degrees of emphasis. In this first formulation of the paradox, for example, note the words “see not.” There are in effect two rhythmic contours that can convey the line's paradox, depending on which word in “see not” a given reader chooses to grant contrastive rhetorical emphasis. One performer might choose to put particular vocal emphasis on the word “not,” which would have the effect of making “not what” a trochee (and the word “not” to carry all the stronger vocal emphasis by virtue of being involved in a metrical deviation): “That they beHOLD and SEE NOT what they SEE.” Another performer might feel the paradox is even more pointed if the voice signals contrastive rhetorical emphasis on the two instances of “see,” in which case one relies on “see” falling in the expected beat position to serve as underscoring for that rhetorical emphasis: “That they beHOLD and SEE not what they SEE.” Either one strikes me as a defensible performance option, and at this stage of the analysis the line might just be thought of as containing a prosodic ambiguity.

Now press on to the next two lines. The speaker says of his eyes, “They know what beauty is.” In the developing meaning context of the sonnet, “is” here is not a simple use of the verb to be (generally unstressed); rather, it operates as a thematically charged word: my eyes do, on some level, know what things are genuinely beautiful. In that regard, it is appropriate that “is” falls in beat position and before a caesura. These factors allow the voicer easily to afford it a degree of vocal emphasis appropriate to its thematic importance. The next line also contains an “is,” and the performer of this sonnet must consider how vocal emphases are to be calibrated. The main rhetorical contrast of this line involves the stark opposites “best” and “worst,” and the voicer will probably decide to afford those two words the primary contrastive emphasis. But this contrast involves another: between what the speaker knows to be the case and what he takes to be the case, and that in turn involves the contrast between what actually is and what he takes things to be. The focus of the line is on the reality-distorting action of the speaker's mind; “is” is a thematically important word in this line, then, too. And again it falls before a caesura, where the pause allows the listener's mind to register its thematic importance. So it carries what I have called thematic emphasis, but can it be afforded any special vocal emphasis? Not, I think, without disturbing the contrastive rhetorical emphases on “best” and “worst” that must anchor any rendering of the line.

This thematically important word “is” has now appeared in two consecutive lines, one where it can easily take vocal emphasis and one where it cannot. The different ways it fits within the rhythms of the lines make the word “is” prosodically uneasy, in a way analogous to how, at the level of content, what actually is is reported as being distorted by the speaker's processing of what is.

Now consider lines 9 and 10. In line 9, “that” functions not just as a demonstrative pronoun (generally unstressed), but as a substantive: that person. Its falling in beat position allows the voicer easily to place vocal emphasis on it to bring out its enhanced meaning in this particular case. Next consider the agent and actions of these two lines. The agent in both lines is the heart. In line 9 the heart thinks something; in line 10 it knows something; it knows something different to be actually the case from what the heart is reported in line 9 as thinking is the case. In line 9, “heart” falls in beat position and “think” falls between it and the word “that,” which calls for vocal emphasis not only because it falls in beat position but because the voicer has to convey to the listener that this “that” carries substantive force. As the second of three stressed syllables, “think” will be rhythmically demoted. However, it forms a rhetorical contrast with “knows” in the next line, and in fact that contrast is the main meaning of the two lines: what the heart allows itself to think versus what it actually knows full well to be the case. So “think” perhaps requires some light vocal marking to convey its contrastive status and prepare for its reversal in the next line. In the rhetorical contrast between “think” and “know,” one wants “know” to carry the heavier emphasis—whatever falsehoods I may allow myself to entertain, I actually know full well what is true—and “know” can easily here carry the heavier vocal emphasis by virtue of falling on the beat, and one other thing. Because it follows the lexically stressed noun “heart,” it would (even in a context where it was not carrying contrastive rhetorical emphasis) have to jockey for relative rhythmic emphasis, to outstress “heart” in order to carry the beat; here that jockeying helps it carry its rhetorical emphasis, or coincides with its rhetorical emphasis. The passage requires careful consideration of how much and what kind of emphasis should be accorded to “know,” “think,” “that,” and “heart” (in each of its appearances)—and that is setting aside the fact that “several” and “common,” and “plot” and “place,” are also contrasting terms. With “think” in particular, rhythmic position pulls one way while lexical stress and rhetorical emphasis pull another. The voicer can work out a delivery of the line, but in doing so will sense—almost physically feel—various words vying for different kinds and degrees of emphasis.

This, I think, is the distinctive effect that Shakespeare achieves with his prosody, particularly within the more angst-ridden of his sonnets: he sets different forms of syllabic emphasis in tension with one another to give a prosodic analogue to the emotional tensions he explores or expresses on the level of content. There is a little prosodic dis-ease (as I allow myself to say in anticipation of this sonnet's closing line) as the syllables assert their claims to various kinds of emphasis until a settlement is imposed on them. Observe the continuation in this case. In the next line, the first “this” probably receives heavier vocal emphasis than the second. Both are substantives, meaning “a particular thing.” Both, appropriate to that richer meaning, fall on beat syllables. But the syllables that take emphasis to express this line's rhetorical contrast are probably not the two instances of “this,” but the “not” and the first “this.” If there is in fact a slightly different level of vocal emphasis on the two instances of “this,” that only accords with the content here: that there exists an actual “this” and a different “this” that the partly deluded speaker substitutes for the real this.

Notice one more thing by beginning to reread the poem. In a poem where “is,” “that,” and “this” are made to mean “things as they really are and that I do in fact know for what they are,” one realizes that the three instances of “what” in lines 2, 3, and 4 could theoretically carry something of the same kind of meaning: “what really is.” No performer, I think, would put any vocal emphasis on these words to audibly point to that meaning; trying to do so would disrupt all the finely calibrated hierarchies of emphasis that the performer worked out for each of those lines so that they could carry the precise meaning that Shakespeare is trying to convey. Yet these uses of “what” have no less claim to thematic emphasis than the “that” and the “this” of lines 9 and 11. On a purely mental level, one might acknowledge their congruence with the overall theme of the poem, but one cannot on the vocal level honor that thematic function, and this kind of strain is just another example of the unrest or dis-ease that I find in Shakespeare's deployment of the various kinds of emphasis in his overall management of his sonnets’ verbal rhythms.

The “unrest” generated by the claims to various kinds of emphasis by some of the syllables in a sonnet extends to many of the more abject young-man sonnets as well, particularly when the speaker takes up injuries that one of them has suffered at the hands of the other. I can offer here two focused examples of this. I draw my first from one of the critics cited above who finds “unease” in the sonnet he examines. That this uneasiness often has a prosodic dimension can be illustrated by John Roe's treatment of a line from Sonnet 120, even though Roe is not particularly focused on prosodic matters. He renders lines 9 and 10 as follows: “O that our night of woe might have remembered / MY deepest sense, how hard true sorrow hits” (89–90; emphasis in original). In context, “my” may well deserve the contrastive rhetorical emphasis that he places here, since the poem contrasts this time that the speaker has hurt the addressee with an earlier time when the addressee had hurt him. “My” does not, however, get the help of falling on a beat; in fact, prosodically, it must be subordinated to the accented syllable in the two-syllable word “deepest.” To steal the beat from “deep” would require a very pronounced contrast, and here there is no single word, for instance your, with which “my” forms a pointed contrast. The voice uneasily seeks the right way to honor such contrastive emphasis as exists here without overplaying it, because an actual “MY deepest SENSE” would place more weight on the “my” than is warranted, but a simple “my DEEPest SENSE” does not do justice to the special force on the pronoun here. Any discomfort the reader feels trying properly to voice this “my” accords generally with the emotional strain of the sonnet's melancholy content, the awkwardness in any effort to gauge the relative pain suffered by two individuals.

Sonnet 49—“Against that time (if ever that time come)” (1)—anxiously anticipates injury, anticipates some future moment when the beloved might withdraw his love. In preparation for such an undesirable development, the speaker begins psychologically to brace himself. He “ensconce[s]” himself “within the knowledge of [his] own desert” (9, 10). But this does not turn out to mean “console myself with the thought that I am deserving, despite your possible future renunciation.” Rather, it means “acknowledge that I never was deserving of your love in the first place”: “Since why to love I can allege no cause” (14). Accordingly, the speaker will “uprear” his own hand “against [him]self” (11), in testimony on the beloved's behalf instead of his own.

The first quatrain has the speaker as the grammatical agent: “Against that time . . . / When I shall see . . .” (1, 2). The second quatrain has the beloved as the grammatical agent: “Against that time when thou shalt strangely pass” (5). The third quatrain starts out with the speaker as the grammatical agent, “Against that time do I ensconce me here” (9), but in the course of it he becomes a spokesman for the beloved—an agent (in a different sense of the word) acting on behalf of the beloved, rather than himself.

The “I” versus “thou” in the first two quatrains is not a rhetorical contrast. Instead, it is just a structuring device: viewing the same feared future development first with a focus on the speaker's experience, then with the focus on the beloved's actions. So the words “I” and “thou” in lines 2 and 5 should not in performance carry the pitch change that marks rhetorical emphasis—or any special vocal emphasis at all. Nevertheless, given the extra role they play in this sonnet, as structuring devices, it is wholly appropriate that they fall on beat syllables. Gradually throughout the poem, but especially with the reversal in the third quatrain, “I” and “thou” become thematically important words; it is fundamentally a poem about how I will respond to how you might someday treat me. And what I argue is that that I-thou theme creates pressure for the “thy” in line 12 to receive vocal emphasis, as it ordinarily would not, based on lexical stress and line position: “I consider my claims to deserving and actually find myself testifying to ‘the lawful reasons on thy part’ for not loving me.” The word “thy” in this case, would take the beat from “on” and the line would end in a pyrrhic-spondee or stress-final pairing. Within the total meaning context of the poem, “thy” gets metrically promoted, just as within the content of the poem the speaker is “promoting” the beloved's perspective even over his own interests. There is a rhythmic analogue to the speaker's abjection.

In writing for the stage, Shakespeare would have had near daily experience of hearing actors’ voices navigate thousands of lines of his and others’ dramatic verse. From this I believe he developed an intuitive sense of various forms of emphasis, something akin to the taxonomy I have laid out in this essay. The prosody of his dramatic verse often amounts to instructions to the actors on how to deliver their lines, the vocal emphasis they should place on particular syllables. Consider the commands uttered by the prince in Romeo and Juliet to the heads of the two warring houses: “You, Capulet, shall go along with me; / And Montague, come you this afternoon” (1.1.92–93). He is announcing an intention to reprove each of them in turn, and so the two instances of “you” carry contrastive emphasis. The second “you” carries its contrastive vocal emphasis by virtue of falling on a beat. It needs to be delivered with special vocal force, though, in order to prevent the “come” (usually more heavily stressed than a pronoun) from turning that foot into a trochee: “COME you.” For the first “you” also to carry contrastive emphasis, it requires even more vigorous voicing, in order to outmuscle the first syllable of Capulet's name, which is a stressed syllable and falls in beat position. But that level of vocal energy is wholly in keeping with the commands the prince is having to bark in order to impose order on the situation. Getting the prosody right will mean that the actor gets the prince's attitude right. Now consider a more complex case. Later in the scene, in his conversation with Benvolio, Romeo says that he has “lost himself,” then adds, “This is not Romeo” (192, 193). The actor can put vocal emphasis on “this” or on “is” or on “not.” The first, with a gesture, might communicate a disjunction between body and essential self; the second, a disjunction between appearances and being; the third, a profound self-alienation. Any of the deliveries (and consequent meanings) strikes me as defensible depending on the thematic emphasis of a particular production, but the point is that the actor must choose one or the other, one at the expense of the other. To give voice to one meaning is to suppress another. From such tensions between what can and cannot be given voice did Shakespeare fashion the prosodic component of the “frantic discourse of unrest” that characterizes many of his more anxious sonnets.