Printer Friendly

Re-proofing the "zero part of speech" in Hamlet.

O mysterious apostrophe, teach us to understand your workings! Show us your varied talents here!

--Jonathan Culler (2)

Designated as "mystery particles" or "the zero parts of speech," interjections such as oh have fallen under the radar of linguists until relatively recently. Indeed, Laurel Brintons description of them reads like an exercise in definition-by-negation:
   They are normally marginal in word class, heterogeneous in form, of
   high frequency, phonetically short, outside the syntactic structure
   of the clause, sentence-initial, lacking in propositional content,
   optional, difficult to translate, and stylistically stigmatized.

Even native speakers who employ these zero parts of speech in a variety of contexts are often hard-pressed to explain the nuances they themselves negotiate in everyday conversations.

In Shakespeare's day, the mystery of the particle oh was compounded in that compositors had to differentiate between two forms of the interjection, O and oh. Little wonder, then, that some Shakespeare scholars have given interjections short shrift. Although in his "Think On My Words": Exploring Shakespeare's Language, David Crystal does briefly take up the interjection in terms of Shakespeare's use of exclamation marks, he quickly elides the difference, referring to oh and O as "either the emotional noise or the vocative ... (it is often unclear which is which)." (4) Assuming--wrongly--little difference between the two forms, the editors of the Folger Hamlet ignore discrepancies between the folio and quarto editions by simply defaulting to the O-form, even though they use "square, pointed and half brackets" to indicate other differences among editions. (5)

Other editors follow suit by simply defaulting to "O" in all instances where options between the two forms present themselves. In doing so, editors risk erasing an important discourse marker from the play, one whose "mysteries" have been unraveled and systematically defined and categorized by today's discourse analysts. Where the compositors of the quarto and folio versions of Hamlet diverge in their selection of one over the other marker, the work of discourse analysts can prove a godsend to the modern-day compositor laboring to select the most appropriate form. For readers who are not specialists in discourse analysis but who nonetheless distinguish between the literary O and the conversational oh, selecting the most appropriate form when two variants appear adds to the play's richly textured discursive contexts and tonal qualities, both of signal importance to how it is received by each new generation.

The findings of discourse analysts can complement the efforts of Shakespeare scholars like Lynne Magnusson, whose focus on social dialogue in Shakespeare concerns "first, verbal reproduction--that is, the reiteration of everyday speech forms; second, verbal maintenance." (6) She maintains that Shakespeare studies have "neglected the interactive features of Shakespeare's language," the very features that discourse analysts indicate are negotiated by discourse markers such as oh. ()7 On the other side of the ledger, Jonathan Culpeper and Merja Kyto note that despite the eminence of Shakespeare, "play texts have not received as much attention from scholars as other literary genres." (8) They observe that plays in particular belong to the class of "speech-purposed" language: that is, they "are designed to produce real-time spoken interaction (they are performed')." (9) In their survey concerning o-related forms in speech-related text types, they find these forms have the highest frequency, suggesting that such forms "have had a strong relationship with the spoken interaction of the time." (10) Following suit, Norman F. Blake demonstrates how the markers "why" and "what" function in Shakespeare's plays according to prescriptions laid down by modern discourse analysts. (11)

Focusing on discourse markers brings into relief how such markers often operate as a "bridge between the spoken and the unspoken" in signaling characters' thought processes. (12) In participation frameworks between speakers, such markers facilitate communication in a variety of ways. Finally, a consideration of the relationship between oh and O demonstrates how the former was associated with everyday speech while the latter constituted for Shakespeare's compositors a "memoire intertextuelle," an inheritance from the Greco-Latin tradition. (13) Choosing between the two thus required compositors to make a choice between a marker rooted in everyday speech and one with a more refined lineage. O and oh serve in this regard as generic markers indicating where and for which class of speakers its use was appropriate. Moreover, the compositors' struggle to differentiate between the two forms is an early instance of efforts to define a national character and literary tradition in contrast to the inherited ones of Greek and Latin. Thus, Culpeper and Kyto demonstrate that the increased use of oh+vocative over O+vocative from 1560 to 1760 suggests "the decreasing authority of classical rhetorical figures within certain text-types." (14)
   Measure for [Printer's] Measure

   O O O O that Shakespeherian Rag--
   It's so elegant
   So intelligent

      --T. S. Eliot (15)

Establishing a baseline for a text such as Hamlet is essential to the project of re-proofing the text: that is, choosing the variant--oh or O--that makes the most semantic sense. Re-proofing, thus, depends on an ability to discern from the noise of individual preferences a pattern of usage (as opposed to simply imposing a modern distinction on the text based on today's improved understanding of oh's functions). The most salient distinctions by which to establish such a baseline between oh and O is the latter's function in apostrophes (oaths such as "O Heavens" and simple addresses such as "O Titus") and optative constructions ("O that this too too solid flesh would melt"). In contrast, oh serves a wide range of discursive functions, from affording a space for a pause to indicating a shift in a speaker's subjective orientation. (16)

Genre played a decisive role in compositors' choices of one form over another, histories and tragedies tending toward the O-side of the ledger and comedies toward the oh-side. Thus, referring to generic conventions, Yana Grinshpun notes that both France and England had witnessed here "a stratification of styles,' from informal to highly formal" in which oh has functioned increasingly as a marker of the conversational, while in literature it has been associated with the "baser" genres (e.g., comedy and satires). (17) Reviewing Grinshpun's work, Michele Monte comments on how genre over time exerted a gradually increasing influence in the choice of forms, noting that "o [O] encroaches upon oh in poetry, while oh encroaches upon o in comedies and correspondence." (18) This encroachment occasionally worked in reverse. Thus, in Thomas Nashe's prose work Pierce Penniless (1592), where the ratio of oh to O is 7:4, we find this crossover much in evidence, so much so that in one instance oh is employed quite anachronistically in a quotation from Ovid: "Oh decus atque aevi gloria summa tui." (19) J. B. Steane, the modern editor, rightfully corrects the passage in his translator's gloss: "O ornament and great glory of your age." (20)

Just as oh has served as a preferred marker in prose and the "baser" genres, so has O served increasingly as a "literary marker" or, more elevated still, a sign of "poeticity." (21) As Mark Dunn informs us, O is always capitalized: "In the Land of Interjections, it would definitely be a member of the clergy, and a respected one at that." (22) Citing the work of Irma Taavitsainen, Ulrich Busse argues that "primary interjections ... are a reliable means to differentiate between different genres." (23) Thus, in the more "noble" genres of history and tragedy, where we find a more aristocratic cast of characters employing blank verse, we would expect to see a greater incidence of O. Correspondingly, in the less "noble" genre of comedy, where there are greater representations of the lower and middle layers of society as well as a higher incidence of prose, we should expect to find a greater incidence of the more colloquial oh. Of course, any examination of Shakespeare's corpus must deal with a considerable amount of noise representing individual compositors' preferences. Even so, the distribution of Os among the three genres confirms the notion that in Shakespeare's day O was beginning to emerge as a mark of poeticity and that oh was beginning to be associated more with the "baser" genre (see Table 1).

That the First Folio's history plays outnumber the tragedies in the percentage of Os may reflect a diachronic dimension here: almost all of the history plays were written in the first half of Shakespeare's career, prior to 1600. The tragedies, many written in the second half of his career, may thus reflect personal and cultural stylistic shifts in the acceptance of oh. (24) The increasing employment of oh in literary genres also may reflect a growing sense of national (and linguistic) identity (in England as well as France). Grinshpun, for example, notes that O is "not a form independent from a 'memoire intertextuelle" for both French and English speakers and writers, who "lived in a state of constant osmosis with the canonical texts attached to Greco-Latin Antiquity, in a colingual relationship." (25) Any change in the permeability of this relationship, towards the use of the vulgar tongue, would therefore of necessity favor oh and reflect a preference for the form of everyday speech.

The most clear-cut evidence of Os status as a literary marker comes from a chronological examination of the incidence of Os in the quartos, where we witness a sharp rise in poeticity (see Table 2).

The sharp uptick of Os in the plays from the latter 1590s on corroborates a pattern of literariness for the quartos noted by Peter Blayney, Lukas Erne, and Sonia Massai. Lynne B. Petersen cites Erne and Massai, who claim that "from the 1590s, and potentially as early as the Tudor period, stage plays were written for a literary market also, and that, moreover, the short quartos represent versions prepared for performance by literary-orientated revisers." (26) Erne, in particular, argues for a growing "legitimization" of playbooks. Shakespeare, he asserts, "is aware of, affected by, and an active participant in the theater's gradual emancipation from an existence that is confined to the stage." (27) In a related context, Erne finds the driving forces behind the rise in dramatic authorship were not the authors themselves but "the London printers, publishers, and booksellers eager to render respectable and commercially profitable what was initially an enterprise with little or no prestige." (28) At the generic level, this uptick demonstrates that early modern compositors were indeed aware of the distinction between the more literary O and the baser, conversational oh. When it came time to upgrade the status of dramatic productions, their preference for O in this regard signals that awareness, as demonstrated in Table 3.

Even where Table 2 marks a deviation from this pattern of increasing literariness, as in the ratios for Troilus and Cressida (73%) and Pericles (80%), it accurately reflects a subsequent shift in Shakespeare's career. In this respect, O tracks along the same lines as the commonplacing marker that Zachary Lessen and Peter Stallybrass investigate in "The First Literary Hamlet and Commonplacing of Professional Plays." Here, Lessen and Stallybrass argue that, unlike his contemporaries Marston and Jonson, Shakespeare "sometime around 1607 ... decided to become an apprentice again in the popular theater" and "seems to have turned his back on sententious plays like Hamlet, Sejanus, or The Malcontent." (29) The commonplacing found in earlier works recedes in this period. This trending toward the vernacular and away from "sententious plays" is confirmed by the chart, the dates lining up quite accurately. (30)

Even in cases where the individual's compositorial style favored O almost exclusively, we find indications that the particular compositor distinguished between the two forms. Thus, in Love's Labors Lost, one of only two ohs among its seventy-six Os occurs in the following passage:

Clow: O & the heauens were so pleased, that thou wert but my Bastard; What a ioyfull father wouldst thou make mee? Goe to, thou hast it ad dungil, at the fingers ends, as they say.

Peda: Oh I smell false Latine, dunghel for vnguem.

Brag: Arts-man preambulat, we will bee singled from the barbarous


Here, the Pedant uses the colloquial oh in what discourse analysts label a corrective or repair to the Clown's "false Latine." Given that Costard has employed the O+vocative a few lines earlier, Shakespeare's compositor clearly marks a functional distinction between the two forms.

As evidenced by variations between quarto and folio versions of the plays, not all of Shakespeare's collaborators were equally adept at sniffing out "false Latine." There is also a good deal of evidence that Renaissance compositors exercised great license in altering and "emending" the texts they set to type. Given this state of affairs, modern editors cannot help but operate, in Ernes formulation, as a "fourth group" of Shakespeare's collaborators, having to adjudicate between competing forms in the quarto and folio versions. (31) Fortunately, editors can rely in part on the findings of discourse analysts concerning generic conventions governing the use of O and oh, findings confirmed in a statistical analysis of the plays themselves. Further refinements can be made by examining what Deborah Schiffrin labels "the contextual coordinates" that markers like oh and O index and mediate in their integration of discourse coherence. (32) As these coordinates sometimes overlap, determining the preferred form is a matter of judging which functions are most prominently at work in a particular instance, not always an easy task. Often we will find it necessary to override one or the other of Shakespeare's compositors and, on some occasions, both. In the latter case, we enter uncharted territory, as it becomes a question of whether or not twenty-first century compositors should intervene when their fellow collaborators have erred in their selections. At the very least, by establishing statistical norms for the use of oh and O, we can identify plays that fall outside those norms and then employ techniques of discourse analysis to re-proof those texts. While such a method may or may not give us a picture of what the original text looked like, it will dampen somewhat the "noise" created by compositors working at times according to widely varying standards.

No Dead Poets in Our Society

In defaulting to O, editors risk offering us a "high-pitched" text in an unrelentingly elevated vocative "key," encouraging a tone-deaf reading of Hamlet (here, our test case, although this concern relates to all plays in the canon). This question of key is most poignantly shown in the selection from Dead Poets Society below. Displaying the two interjections in action, this exchange illustrates the difference in texture and tone between the colloquial and literary forms:

Keating: ... Today we're going to be talking about William Shakespeare.

The class lets out a collective sigh.

Boy: Oh, God!

Keating: I know. A lot of you looked forward to this about as much as you look forward to root canal work. We're gonna talk about Shakespeare as someone who writes something very interesting. Now, many of you have seen Shakespeare done very much like this:

Keating holds out his right arm dramatically and begins to speak in an exaggerated British accent.

"O Titus, bring your friend hither." But if any of you have seen Mr. Marlon Brando, you know, Shakespeare can be different. "Friends, Romans, countrymen, lend me your ears." You can also imagine, maybe, John Wayne as Macbeth going, "Well, is this a dagger I see before me?" (33)

The boy's "Oh, God" is strongly interactional and affective, immediately eliciting Keating's "I know." To break through the students' resistance to Shakespeare, Keating first employs the Latinate apostrophe in emphasizing that he agrees with them how removed Shakespeare can seem from common speech. He then shifts tone altogether to drive his point home. Keating's Marlon Brando makes "ears" sound like "arse" and his vocalization of John Wayne copies his Western cowboy drawl. This passage demonstrates some of the distinctions between oh and O, the lowly and the elevated, the colloquial and the literary. Shakespeare himself plays upon these distinctions when he has Sir Hugh Evans bait the guileless William with the bawdy implications of the "Focative" case:

Eua: ... What is the Focatiue case (William)?

Will: O, Vocatiuo, O.

Eua: Remember, William; Focative is caret. (Merry 4.1.1867-70)

Defaulting to O in cases where oh constitutes an alternative and viable choice deprives modern readers of recognizable and familiar linguistic and tonal signposts that were also extant in Shakespeare's day.

In offering something of a guide for twenty-first century "compositors," discourse analysts bring to bear a highly nuanced understanding of the communicative process. For Schiffrin, oh's interactive functions help speakers negotiate the division of conversational knowledge in a variety of ways. Establishing a joint focus of attention, oh "not only allows transitions in information state, but it marks information as more salient with a possible increase in speaker/hearer certainty as to shared knowledge and meta-knowledge." Generally, it displays and aligns speakers and hearers as "mechanically defined nodes in a system of information transmission." (34) At a cognitive level, oh functions "pragmatically" in making accessible "speaker/hearer assumptions about each others subjective orientations toward information." (35)

In its strongly interactive functions, oh operates as a veritable switchboard in the communication process, facilitating verbal maintenance functions between speakers and hearers. These include requests for elaboration; establishing speaker alignments; registering rebuttals, repairs and completions (self- and other-); and communicating recognition displays. In "Oh as a Marker of Information Management," Schiffrin cites John Heritages observation that oh is a particle "used to propose that its producer has undergone some kind of change in his or her locally current state of knowledge, information, orientation or awareness." (36) She underscores the importance of oh in managing information state transitions and the participation frameworks that define the alignments between speakers and hearers. In demystifying "mystery particles" such as oh, Schiffrin explains how these markers "index the location of an utterance within its emerging local contexts." (37) They are, in effect, "indices of the underlying cognitive, expressive, textual, and social organization of a discourse." (38) Oh plays many multifaceted roles in mediating repairs, elaborations, clarifications, and alignments between speakers and hearers.

Although discourse analyst Raymond Person, Jr., believes--wrongly, I will argue--that oh and O were interchangeable in Shakespeare's era, he nonetheless offers considerable evidence that whether one or the other form had been favored to fill a particular slot, modern distinctions concerning oh's functions were extant in the Early Modern period; moreover, Person demonstrates that not only did Shakespeare observe the full range of such distinctions for oh, but he finds the "Shakespeare corpus includes some previously unidentified uses of 'oh,'" such as an indicator of epistemic authority. (39) In their own discussion, Early Modern English Dialogues, Culpeper and Kyto add more functions to the list. Collating these functions from the works of various discourse analysts, we can draw up a fairly exhaustive taxonomy of at least sixteen functions at work in Shakespeare's corpus: (40)

To prevent editors from becoming what Holofernes labels "rackers of ortagriphie" (LLL 5.1.1759), we can examine each of Folio Hamlet's 115 instances of oh and O to identify which marker is most appropriate given its context and the applicable rule involved. In three-fourths of the instances, simply identifying the slot as an apostrophe (nominative) or optative construction made it relatively easy to determine where O was appropriately employed. While it is not feasible to review each and every instance, the chart below cites the rule employed for choosing oh as the preferred form:

Reassessing the values oh and O (as catalogued in Table 6), we find that the plays ratio of 31% O adjusts now to 77% O, much more in line with the expected norm for a tragedy (68%). (48) The compositors overreliance on oh here is adjusted, restoring the tragedy's tonal balance while preserving the discourse functions of oh in the play.

While plays that existed only in the Folio format did not require compositors to make choices between the two forms, those occurring in both formats were not so quickly settled. When we turn our attention to the plays of the First Folio, we find that the movement toward "poeticity" is confirmed from quarto to folio versions in ten of the plays, with six plays moving marginally in the direction of oh. (49) Having established the baseline for tragedies at 68% O, we can readily see that two tragedies fall far outside the norm, reflecting the particular compositors preference for one form over the other. Thus, folios Hamlet and Othello reverse the quartos' trend toward poeticity quite markedly, with a preponderance of ohs. Comparing quarto 2 of Hamlet with its folio version, we find the percentage of Os dropping from 93% to 23%; similarly, from quarto to folio Othello, the percentage of Os drops from 96% to 23%. Given such divergent data sets, we can see why modern editors would find pushing the O-default button tempting. Thanks to the efforts of discourse analysts, however, we do have more precise tools at hand for selecting the appropriate form among contesting variables in most instances. While it is far beyond the scope of this article to re-proof every play in the canon, we can establish the rules for selection between these two forms and apply them to a total re-proofing of the First Folio and Quarto 2 versions of Hamlet.

"A rare command of modulation ..."

The nineteenth-century philologist John Earle wisely observed that "the interjection is of all that is printed the most difficult thing to read well aloud," requiring in his estimate "a rare command of modulation." (50) The "rare command of modulation" that Earle demands of the actor delivering the interjection is also required of the modern-day compositor who must choose between these competing forms. To some extent, at least, confusions can be clarified and resolved. For example, misunderstanding the distinction between the vocative and the clarification functions of O and oh, Alexander Schmidt provides in his Shakespeare-Lexicon an instance of the two figures being "confounded," when in actuality they are not. In Two Gentlemen of Verona, for example, Launce ascribes unflattering identities to various members of his family: his father, his right shoe; his mother, his left. Turning on/to himself, he reasons: "I am the dog: no, the dog is himself, and I am the dog--Oh! the dog is me, and I am myself" (2.3.22-24). A few lines later, he exclaims: "Now come I to my mother. O, that she could speak now like a wood woman!" In the first instance, framed in an aposiopeia of recognition, the Oh correctly reflects Launce's "clarification" of his identity. In the second, O functions as Schmidt claims it should, as an optative expression giving "the speech the character of earnestness." (51) Charting the coordinates the interjection mediates can prevent such confusions about the two forms.

More rarefied forms of modulation can tax the abilities of even the best discourse analysts. Schiffrin underscores the complexity of discourse markers, which "allow speakers to construct and integrate multiple planes and dimensions of an emergent reality: it is out of such processes that coherent discourse results." (52) At various points, these planes and dimensions intersect with those calling for O, making it difficult to fill in the slot only one marker can occupy. Separated by only two lines, the "Oh" and "O" featured below challenge the re-proofer's abilities:

Qu: Oh Hamlet, [Quarto 2: "O Hamlet"]

Thou hast cleft my heart in twaine.

Ham: O throw away the worser part of it,

And liue the purer with the other halfe. (3.4.2539-42)

The O and oh here both express the vocative function, a plane where the two intersect. Why then in the Folio version do we find a different form occupying each discourse slot? Here, we can profit from linguists' distinctions in verifying the Folio compositor's selections as just. Grinshpun defines oh as an interjection "strongly interactional" while O marks a break from the expression preceding it. (53) She employs a passage from Giraudoux' Ondine in illustrating the difference between the two forms:

The Chevalier: Where do you wish to go, dear Ondine?

Ondine: O Hans, listen to me. I know someone who could join us forever, someone very powerful, who could make it so we would be joined one to another like certain twins. Do you wish for me to call upon her?

The Chevalier: You're referring to the adoptive daughter of the king, Ondine!

Ondine: To the kings daughter! You want to know who she is, the daughter of the king! You want to know, you who tremble before her!

The Chevalier. Oh Ondine, howyou remind me of the commoners state. (54)

Syntactically, the two forms are "perfectly identical: O+proper name." (55) They both signal the beginnings of interventions by their respective respondents. Ondine's use of O signals "a strongly affective attitude towards Hans" by interrupting him. (56) She introduces a new phase in the exchange that excludes Hans as a participant. On the other hand, Grinshpun finds the Oh that follows "strongly interactive" and reactive on Hans's part. The choice of Oh has "a semantic value: it governs the interpretation." (57) In a similar fashion, Gertrude's "Oh" signals that she has been made receptive to Hamlet's effort to reform her by creating in her a crisis of conscience. Creating a joint focus of attention, oh displays speakers and hearers "in particular productive and receptive capacities." (58) Gertrude, in particular, signals her receptivity to Hamlet's efforts, a function of oh as a discourse marker. The Ghost's O represents an action in which he seeks to redirect and refocus Hamlet--"O step betweene her and her fighting Soule" (3.4.2493)--in light of the spectral figure's desire to turn her away from Claudius and back to honoring his memory. In Grinshpun's sense, it marks a "rupture" in the dialogue, representing "a new phase in the exchange." (59)

Other instances of hybridization of functions between the two forms also serve to tax the abilities of re-proofers. Schiffrin indicates the importance of oh "in action structures (since it marks certain actions, e.g. clarifications, which are designed to manage information state transitions)." (60) Below, Hamlet seeks clarification--"Oh say"--as to the reason for the Ghost's appearance, and oh thus prefaces his request:

Ham: Do you not come your tardy Sonne to chide,

That laps't in Time and Passion, lets go by

Th'important acting of your dread command? Oh say.

[Quarto 2: "6 say."]

Ghost: Do not forget: this Visitation

Is but to whet thy almost blunted purpose.

But looke, Amazement on thy Mother sits;

O step betweene her, and her fighting Soule,

Conceit in weakest bodies, strongest workes. (3.4.2487-94)

Hamlet's request for clarification occupies the same slot defined by the vocative; here the nominative address to the Ghost is implied. The re-proofer must decide between the two forms, a split decision on the part of the early compositors. Given the intensity of Hamlet's request for clarification, I would opt, with some reservation, for the Folio's "Oh." In a similar mode, Polonius's boast about knowing the cause of Hamlet's melancholy elicits an earnest elaboration request from the king: "I haue found / The very cause of Hamlets Lunacie. / King: Oh speake of that, that I do long to heare" (2.2.1072-74; Quarto 2: "O speake of that" [2.2.1074]). Given the intensity of the King's request for elaboration, Oh should be the preferred form here. Later in the above passage, the Ghost correctly, according to Grinshpun, employs O to mark a new phase in their exchange, redirecting Hamlet's attention to a Gertrude ready to be transformed. The interjections in this turbulent scene serve as contextual coordinates plotting the positions of the three characters as they gradually come into alignment with each other.

At times, the crafts of discourse analysis and prosodic analysis can work in tandem to help the modern compositor select the proper form for a slot. Thus, Earle provides prosodic criteria for differentiating between these two interjections. He opposes any blurring of distinctions between O and oh, arguing, "This distinction of spelling should by all means be kept up, as it is well founded." (61) He notes important prosodic and semantic distinctions between them:
   And as to the sound, the O is enclitic; that is to say, it has no
   accent of its own, but is pronounced with the word to which it is
   attached, as if it were its unaccented first syllable.... Oh! on
   the contrary, is one of the fullest of monosyllables, and it would
   be hard to place it in a verse except with the stress upon it. (62)

Earle's assertion that oh is always stressed should be tempered somewhat here: while it is true that oh tends to attract more accent than O, we should not view such attraction necessarily as a hard-and-fast prescription. The variability of such factors as intonation, rhythm, and stress should be taken into account in determining which form should fill a slot that might otherwise be problematic. Fortunately, we can defer to the actors "rare command of modulation" to take such factors into consideration and then make a final determination.

Given such variability, principles of discourse analysis can be employed to cross-check prosodic considerations when determining which form to employ. Thus, the oh employed in Claudius' soul-searching soliloquy can be justified both by its stressed position and by what discourse analysts would label its speaker intensification function, indicating a dramatic shift in subjective orientation (stressed syllables are indicated with italics):

King: Then lie looke vp,

My fault is past. But oh, what for me of Prayer

Can serue my turne? (3.3.2326-28)

Contextual cues such as genre can also aid the modern collaborator in choosing between forms. When a choice has to be made in a prose section of the play, more weight should be given to the oh-form. In many cases, this simply reaffirms the choice of oh: "Ham. O[h] the Recorder" (3.2.2216) or "Ham. Is't possible? / Guild. Oh there has beene much throwing about of Braines" (2.2.1404-6). When a grieving Laertes asks Claudius why he has not acted against Hamlet after the killing of Polonius, Claudius replies: "O for two speciall reasons" (4.7.3017). This is clearly not in the vocative mode; oh's function in introducing a clarification sequence is what is called for here. That Claudius' response is rendered in a prose section further validates the selection of oh.

In other prose contexts, where the vocative is in play with oh, for example, as an instance of speaker intensification, the choice is not so clear-cut: "Ham. O[h] it offends me to the Soule" (3.2.1856). We find similar difficult hybridizations in the metered context as well, such as a conflict occurring when the vocative expression occupies the same slot as a rebuttal:

Ham: Horatio, thou art eene as just a man

As ere my Conversation coped withall.

Hor: O[h] my deere Lord.

Ham: Nay, do not thinke I flatter ... (3.2.1904-7)

Here, oh should mark Horatio's efforts to rebut or deflect what he perceives as excessive praise on Hamlets part.

Beyond helping us adjudicate between the two forms, the findings of discourse analysts can demonstrate how Shakespeare took to heart his own character Tranio's prescription to study the "rhetoric in your common talk" (as Magnusson points out). Shakespeare confirms Brintons assertion that texts of earlier periods contain an "oral residue" preserved for linguists as freeze-framed transcripts of speech practices of the past. (63) That Shakespeare observed and understood distinctions intuitively practiced by native speakers and later codified by discourse analysts can be readily observed in comparing and analyzing passages from both. For example, in her analyses of interjections, Schiffrin demonstrates just how many levels an interjectival discourse marker such as "but" can mediate in the following interchange between her friends Jack and Freda:

(2) Jack: [The rabbis preach, ["Don't intermarry"

Freda: [But I did- [But I did say those intermarriages that we have in this country are healthy.

Schriffin underscores the complexities of this exchange by identifying the multiple discursive levels mediated by the interjection But:
   Freda's but prefaces an idea unit ("intermarriages are healthy"),
   displays a participation framework (nonaligned with Jack), realizes
   an action (a rebuttal during an argument), and seeks to establish
   Freda as a current speaker in an exchange (open a turn at talk).

In the above exchange, the four functions mediated by the interjection serve to "locate an utterance at the intersection of four planes of talk." (65) Here, the interjection operates like a communication switchboard connecting and modulating people's dialogues and interactions. Schiffrin notes that reorientations occur "when speakers discover a discrepancy between their own and another's evaluation of a piece of information, e.g. their level of commitment to a proposition." (66)

In this light, the following dialogue between the Player King and Player Queen demonstrates just how close to modern discourse the Shakespearean text can be. Providing an example of other-initiated repair, the Player Queen interjects--and objects--to her husband's insinuation that she might not be faithful to him after his death:

King: Faith I must leaue thee Loue, and shortly too:

My operant Powers my Functions leaue to do:

And thou shalt liue in this faire world behinde,

Honour'd, belou'd, and haply, one as kinde.

For Husband shalt thou--

Bap: Oh, confound the rest:

Such Loue, must needs be Treason in my brest. (3.2.2040-46)

Written some four-hundred years prior to Schiffrin's exchange, Shakespeare's text could very well supply her with an added case study, differing only by rhyme and meter from her more modern examples. Here again, we find Oh setting up an idea unit (widows remarry), a participation framework (nonaligned with the King), a rebuttal ("Oh, confound"), and an opening for a turn at talk (the Queens interruption). Not only does her Oh mark a discrepancy between her and the King's evaluation of information, but it also demonstrates cognitively her anticipation of what he is going to say, as she "confounds" or forestalls his completion of the idea unit (i.e., widows remarry). Schiffrin's observation that oh mediates "speaker/hearer assumptions about each other's subjective orientations toward information" is in full force here. (67)

A true compositor, the Player Queen must fill in the slot she herself opens up here. As the back-and-forth of dialogue is a collaborative enterprise, she must process the Player King's speech, anticipate that "kinde" is about to be followed and completed by the rhyming complement "finde," and edit out and replace the offending word by opening up a space for interjection in the dialogue. We might note as well that her process of anticipation demonstrates Shakespeare himself in his own process of anticipation in composing and editing this passage. Thus, the originally intended "finde" would have fallen short metrically of the rhyme scheme here, falling on the sixth rather than tenth syllable: "For Husband shalt thou [finde]." The Queen's repair and replacement of the King's anticipated offending remark are mirrored in Shakespeare's own repair and replacement of the potentially metrically offensive line. Better than a "rare command of modulation," such a rare triangulation is perhaps the closest we can get to a sense of authorial practice in the editing and selection process. Here, the modern compositor might exclaim with Hamlet: "o tis most sweete / When in one line two crafts directly meete" (Q2 3.4.2577.8-9).

While Shakespeare's modern collaborators cannot ensure that the playwright's and editors crafts "directly meete," they can bring the two within a reasonable hailing distance. To some extent, at least, confusions can be clarified and resolved. Sharpening the criteria employed by collaborators will not provide a definitive choice in every instance, but even the educated guess is preferable to simply ignoring differences that can potentially provide readers with important discursive cues. In some cases there will be irreconcilable differences that simply force our hand toward the lower or upper case. Having done our best to resolve most of them, however, we can exclaim along with the Player Queen: "Oh confound the rest!"

University of Detroit Mercy


(1) All citations are derived from Shakespeare's First Folio (1623), Old Spelling Transcription, Internet Shakespeare Editions. Michael Best, ed. University of Victoria, n.d. Web. 24 Jan. 2016. Citations reflect the fact that the ISE numbers lines continuously throughout a play.

(2) Jonathan Culler, The Pursuit of Signs (New York; Cornell University Press, 1981), 135.

(3) Laurel J. Brinton, "Historical Discourse Analysis," in The Handbook of Discourse Analysis, ed. Deborah Schiffrin, Deborah Tannen, and Heidi E. Hamilton (New York: Blackwell Publishers, 2003), 142.

(4) David Crystal, "Think On My Words": Exploring Shakespeare's Language (Cambridge: Cambridge University Press, 2008), 75.

(5) Gabriel Egan cites Margreta de Grazia's observation that such a practice is comparable to Derrida's notion, "derived from Martin Heidegger, of placing words sous rature (under erasure), marking the imperfection of signification by printing a cross over a word to reveal the inadequacy of the sign without obliterating it." Gabriel Egan, The Struggle for Shakespeare's Text: Twentieth-Century Editorial Theory and Practice (Cambridge: Cambridge University Press, 2010), 196. See also Margreta de Grazia, "The Question of the One and the Many: The Globe Shakespeare, The Complete King Lear, and The New Folger Library Shakespeare," Shakespeare Quarterly 46 (1995): 245-51 (249).

(6) Lynne Magnusson, Shakespeare and Social Dialogue: Dramatic Language and Elizabethan Letters (Cambridge: Cambridge University Press, 2008), 153.

(7) Magnusson, Shakespeare and Social Dialogue, 4.

(8) Jonathan Culpeper and Merja Kyto, Early Modern English Dialogues: Spoken Interaction as Writing (Cambridge: Cambridge University Press, 2010), 9.

(9) Ibid., 17.

(10) Ibid., 268.

(11) Norman F. Blake, " Why and What in Shakespeare," in Chaucer to Shakespeare: Essays in Honour of Shinsuke Ando, ed. Toshiyuki Takamiya and Richard Beadle (New York: D. S. Brewer, 1992), 179.

(12) John Earle, The Philology of the English Tongue (Oxford: Clarendon Press, 1923), 190,

(13) Yana Grinshpun, "O et Oh: une graphie entre langue et genres de discours," Marge linguistiques 9 (2005): 242-255 (242). All quotations that follow are the authors translation.

(14) Culpeper and Kyto, Early Modern English Dialogues, 281.

(15) T. S. Eliot, The Wasteland (New York: Norton Critical Editions, 2001), 9, lines 128-130.

(16) Given that individuals' identities and very consciousnesses evolve as social, linguistic constructions, our intra-mental operations often reflect features found in the discourse between speakers in larger social contexts. Thus, the interjection's role in mediating information state transitions and participation frameworks at the tagmemic and syntagmatic levels among speakers sheds light on the operations of a larger "interjection" in the play, the soliloquy itself. I explore its role here in "Interrogating the Soliloquist: Does It Really Go Without Saying?" Symploke 18 (2010): 131-54.

(17) Grinshpun, "O et Oh," 248. Culpeper and Kyto cite Irma Taavitasainen's finding in her study of the Helsinki Corpus that interjections occurred "with greatest frequency by far" in comedies compared to other speech-related text types (Early Modern English Dialogues, 287). See also Irma Taavitsainen, "Interjections in Early Modern English: from Imitation of Spoken to Conventions of Written Language" in Historical Pragmatics: Pragmatic Developments in the History of English, ed. Andreas H. Jucker (Philadelphia: John Benjamin, 1995).

(18) The "miniscule" O, as Grinshpun indicates, serves as an alternate form of O: "Today still, vocative sequences such as 'o+Proper name' are readily recognized as Hellenisms" (242). See Michele Monte, "Review of O entre langue(s), discours et graphie" Corpus 8 November 2009 (Paris: Ophrys, 2008),

(19) Thomas Nashe, "Pierce Penniless," in The Unfortunate Traveler and Other Works, ed. J. B. Steane (New York: Penguin, 1989), 143.

(20) Ibid.

(21) Grinshpun, "O et Oh," 248, 253.

(22) Mark Dunn, Zounds! A Browser's Dictionary of Interjections (New York: St. Martin's, 2005), 123.

(23) Ulrich Busse, Linguistic Variation in the Shakespeare Corpus: Morpho-Syntactic Variability (New York: John Benjamin, 2002), 199.

(24) As Taylor and Jowett explain, Shakespeare's younger contemporary, Thomas Middleton, was more in line with the times in his preference for oh. Gary Taylor and John Jowett, Shakespeare Re-Shaped: 1606-1623 (Oxford: Oxford University Press, 1993).

(25) Grinshpun, "O et Oh," 242.

(26) Lynne B. Petersen, Shakespeare's Errant Texts: Textual Form and Linguistic Style in Shakespearean 'Bad' Quartos and Co-authored Plays (Cambridge: Cambridge University Press, 2010), 5.

(27) Lukas Erne, Shakespeare as Literary Dramatist (Cambridge: Cambridge University Press, 2003), 58.

(28) Ibid., 57.

(29) Zachary Lessen and Peter Stallybrass, "The First Literary Hamlet and the Commonplacing of Professional Plays," Shakespeare Quarterly 59 (2008): 371-420 (417).

(30) While a full analysis Jonson's practice is outside the scope of this article, O and oh also prove useful markers for tracking Jonson's own movement towards a "sententious" style and an identity as a literary dramatist. Thus, while his early comedies are "hybrid" productions where oh and O are concerned, later comedies affirm Lessen and Stallybrass's contention that Jonson was headed in a decidedly different direction from Shakespeare around the period of 1607:
   Everyman in his Humor (1598) 30 ohs/119 Os
   Everyman Out of his Humor (1600) 12 ohs/231 Os
   Poetaster (1602) 22 ohs/229 Os
   Sejanus (1603) No ohs/60 Os
   Volpone (1605-6) 12 ofts/87 Os
   Epicoene (1609) No ohs/l 17 Os
   Catiline (1611) No ohs/99 Os
   The Alchemist (1612) No o/is/199 Os
   Bartholomew Faire (1614) No ohs/l 16 Os

(31) Erne, Shakespeare as Literary Dramatist, 3. Andrew Murphy informs us that, in his discussion of the shift from blackletter to the more prestigious Roman type in play-text printing, "[Peter] Blayney presents some striking figures indicating the rapid decline in the use ofblackletter in play-text printing, noting that between 1583 and 1592, 'nine out of twenty plays (45 percent) were printed in blackletter, but in 1593-1602 the proportion dropped to ten out of seventy-six (13 percent),' and thereafter the overwhelming majority of plays were printed in roman." Andrew Murphy, Shakespeare in Print: A History and Chronology of Shakespeare Publishing (Cambridge: Cambridge University Press, 2003), 30-31.

(32) Deborah Schiffrin, Discourse Markers (Cambridge: Cambridge University Press, 1996), passim.

(33) Dead Poets Society: Final Script, original screenplay by Tom Schulman, director: Peter Weir,

(34) Schiffrin, Discourse Markers, 99.

(35) Ibid., 100.

(36) Ibid., 74.

(37) Ibid.

(38) Ibid., 66.

(39) Raymond Person, "'Oh' in Shakespeare: A Conversation Analytic Approach," Journal of Historical Pragmatics 10 (2009): 84-107 (84).

(40) See Andreas H. Jucker and Yael Ziv, eds., Discourse Markers: Descriptions and Theory (Philadelphia: John Benjamins, 1998).

(41) "Intensification occurs not only in argument: it occurs whenever speakers strengthen their reactions to what is being said. In fact, many cases of oh which might be traditionally classified as exclamatory can be seen as speaker intensification" (Schiffrin, Discourse Markers, 96). Culpeper and Kyto might include expressions of "amorous distress" in this category.

(42) Culpeper and Kyto offer as a subset to this category the vocative as a response to death.

(43) Oh functions in this example to challenge a presupposition (Culpeper and Kyto, Early Modern English Dialogues, 242).

(44) Ibid., 241. Evaluation can often take on a sarcastic dimension, as Culpeper and Kyto point out.

(45) Ibid. Horatio's deferral of Hamlet's praise constitutes, employing Culpeper and Kyto's concept, a politeness strategy.

(46) See Person, "'Oh' in Shakespeare," 86.

(47) The overriding factor for selecting oh above O in these examples is the discursive function. Three less critical but helpful factors strengthen the case for choosing oh over O:

1) When it occurs in a prose section of the play. [See examples 1, 5, 8, 12 and 16]

2) When there is separation between the marker and its nominative predicate: "Oh he is mad Laertes" (3469) versus "O Laertes, he is mad."

3) When the marker is separated from the sentence by a comma or, more rarely, an exclamation point. Grinshpun finds the same differentiation in the French tradition: "compositors tend to write OH when the form appears on its own, and O when it is followed by a proper noun" (244). [See examples 1, 3, 4, 8, 10 and 13]. One caveat, however. Early Modern punctuation is highly non-standard; thus, this determinant is less reliable than the two listed above.

(48) A similar re-proofing of Othello also markedly changes its ratio of 117/34 oh/O to 36/115 oh/O. The percentage of Os thus increases from 23% to 77%, far closer to the baseline for tragedy of 68% O.

(49) The fact that King Lear, Hamlet, Othello, and Richard II run counter to the predicted pattern of increased Os in the First Folio may be explained by Jowett's contention that they were among a group of plays printed in F "through a complex and dense intermixing of quarto copy and readings or annotations from manuscript." John Jowett, Shakespeare and Text (Oxford: Oxford University Press, 2007), 81. Thomas L. Berger maintains that the Othello quarto "differs from the Folio in its brevity and in many readings, indicating that the Folio text came from a different source." "Shakespeare Writ Small: Early Single Editions of Shakespeare's Plays," in A Concise Companion to Shakespeare and the Text, ed. Andrew Murphy (New York: Blackwell, 2007), 69. David Bevington follows suit, noting that the Folio version "seems to have been derived (via an intermediate transcript) from a revision of Shakespeare's original manuscript, copied over by himself." "Working with the Text: Editing in Practice" in Murphy, A Concise Companion to Shakespeare and the Text, 172.

(50) Earle, English Tongue, 188.

(51) Alexander Schmidt, Shakespeare-Lexicon: A Complete Dictionary of All the English Words, Phrases and Constructions in the Works of the Poet (London: Williams and Norgate, 1875), 2:785.

(52) Schiffrin, Discourse Markers, 330.

(53) Grinshpun, "O et Oh," 254.

(54) Quoted in Grinshpun, "O et Oh" 253.

(55) Grinshpun, "O et Oh," 254.

(56) Ibid.

(57) Ibid.

(58) Schiffrin, Discourse Markers, 316.

(59) Grinshpun, "O et Oh," 254.

(60) Schiffrin, Discourse Markers, 316.

(61) Earle, English Tongue, 190.

(62) Ibid., 190-91. Culpeper and Kyto follow suit in this regard, noting that O is "much less independent, typically functioning in conjunction with the following element" (Early Modern English Dialogues, 277).

(63) Brinton, "Historical Discourse Analysis," 142.

(64) Schiffrin, "Discourse Markers: Language, Meaning, and Context," in Schiffrin et al., Handbook of Discourse Markers, 57.

(65) Ibid.

(66) Schiffrin, Discourse Markers, 97.

(67) Ibid., 100.
Table 1. Percentage of O/oh in the First Folio, by Genre

Genres       Percentage of O/oh
             First Folio

Histories    76/24
Tragedies    68/32
Comedies     63/37

Table 2. Percentage of Os, by Quarto

Quartos                        Percentage of Os    Raw Scores: O/oh

2HVI (1594)                    29                  10/24
Titus (1594)                   31                  20/44
3HVI (1595)                    32 (Octavo)         12/26
Richard II (1597)              26                  10/28
Richard III (1597)             64                  55/30
R & J (1597)                   60                  43/29
IHIV (1598)                    92                  48/5
LLL (1598)                     97                  73/2
2HIV (1600)                    98                  40/1
HV (1600)                      96                  24/1
Merchant (1600)                100                 22/0
Much Ado (1600)                100                 46/0
Midsummer (1600)               94                  49/3
Merry Wives (1602)             100                 26/0
Hamlet Q1 (1603)               100                 76/0
Hamlet Q2 (1604/5)             95                  97/5
King Lear (1608)               96                  75/3
Troilus (1609)                 73                  64/24
Pericles (1609)                80                  37/9
Othello (1622)                 96                  128/5
Two Noble Kinsmen (1613-14)    85                  51/9

Table 3. Percentage of O/oh in the Quartos, by Genre

Quartos      Percentage of O/oh

Histories    62/38
Tragedies    80/20
Comedies     98/2

Table 4. Functions of Oh in Shakespeare

Functions of Oh         Explanation             Examples

1. Pause                Oh allows speaker       Qu: How say you: oh,
                        time to search for an   I should remember
                        information unit.       him: do's he not hold
                                                vp his head (as it
                                                were?) and strut in
                                                his gate?
                                                Si: Yes indeede do's
                                                he. (Merry 1.4.425)

2. Context Change       Oh indicates the        Ham: Oh, Rosincrane;
                        entrance of a new       good Lads:
                        person or an address    {Ham. 2.2.2634)
                        to a new person

3. Change in Topic      Oh marks a shift in     Lys: Where is
                        the speakers focus      Demetrius? oh how fit
                                                a word
                                                Is that vile name, to
                                                perish on my sword!
                                                (Midsummer Night

4. Beginning of         Oh initiates a          King: Thankes deere
Soliloquy               soliloquy               my Lord. Oh my
                                                offence is ranke, it
                                                smels to heaven ...
                                                (Hamlet 3.3.2311-12)

5.Speaker               Oh indicates            She would mocke me
Intensification         speakers'               into ayre,
                        strengthening their     O[h] she would laugh
                        reaction to what has    me
                        been said. (41)         Out of my selfe.
                                                (Much Ado 3.1.1165-66)

6. Assessment of        Oh marks recipient's    Othello: ... his
Informings (42)         evaluation of           [Cassio's] mouth is
                        information from        stopp'd.
                        another participant     Honest lago hath
                                                tane order for't.
                                                Desdemona: Oh, my
                                                feare interprets.
                                                What is he dead?
                                                (Othello 5.2.3328-30)

7. Request for          Oh is employed by the   Ham: My father,
Elaboration/            hearer to gain          me thinkes I see
Clarification           further knowledge       my father.
                        about the               Hor: Oh where my
                        information unit.       Lord?
                                                (Ham. 1.2.372-73)

8. Repair               Oh indicates where an   Ham: Is't possible?
                        interlocutor locates    Guild: Oh, there has
                        and replaces a prior    been much throwing
                        information unit.       about of brains.
                                                (Ham. 2.2.1404-5)

9. Self-Repair/         Oh indicates where a    Ham: Oh this is hyre
Self-Completion         speaker self-           and Sallery, not
                        initiates a repair to   Revenge.
                        or completion of an     (Ham. 3.3.2355)
                        earlier assertion.

10. Other-Repair,       Oh indicates where      Ham: Then saw you not
                                                his face?
Other-Completion        a speaker repairs or    Hor: O[h] yes, my
(Rebuttal)              completes the other     lord, he wore his
                        speaker's statement.    beaver up.
                                                (1.2.228-29) (43)

11. Evaluation (44)     Oh is employed in       Bull: Goe some of
                        the evaluation of       you, conuey him to
                        information.            the Tower.
                                                Rich: Oh good:
                                                conuey: Conueyers are
                                                you all,
                                                That rise thus nimbly
                                                by a true Kings fall.
                                                (RII 4.1.2241-43)

12. Receipt of          Oh marks a              Isab: To morrow? oh,
Unanticipated Answers   reorientation toward    that's sodaine, Spare
                        a proposition whose     him, spare him.
                        completion had been     (Measure 2.2.836-37)

13. Response to an      Oh marks the refusal    Ham: Horatio, thou
Assessment              of a compliment,        art eene as just a
                        offer, or request       man As ere my
                                                conversation coped
                                                Hor: O[h] my deere
                                                Ham: Nay, doe not
                                                thinke 1 flatter ...
                                                (Ham. 3.2.1904-7) (45)

14. Shift in            Oh is employed to       Pol: We are oft too
Subjective              evaluate the content    blame in this, Tis
Orientation             of one's own talk       too much prou'd, that
                        and signal a change     with Deuotions
                        in knowledge or         visage, And pious
                        awareness.              Action, we do surge
                                                ore The diuell
                                                King: Oh 'tis true:
                                                How smart a lash that
                                                speech doth giue my
                                                Conscience? (Ham.

15. Epistemic           Oh indicates a          Ham: Madam, how like
Authority               discrepancy between     you this Play?
                        the speaker and         Qu: The Lady protests
                        addressees evaluation   to much me thinkes.
                        of information,         Ham: Oh but shee'l
                        implying the speakers   keepe her word.
                        superior                (Ham. 3.2.2097-99)
                        knowledge. (46)

16. Recognition         Oh marks the            Ham: Then you live
Display                 acknowledgment of       about her waste, or
                        new information.        in the middle of her
                                                Guil: Faith, her
                                                privates, we.
                                                Ham: In the secret
                                                parts of Fortune? Oh,
                                                most true: She is a
                                                (Ham. 2.2.1277-81) (47)

Table 5. Functions of Oh and Associated Editorial Rules

Folio                             Quarto   Preferred   Rule

1. Laer: Oh, feare me not.        O        Oh          Rebuttal

2. Claud: Oh speake of that       O        Oh          Elaboration
... (2.2.1074)                                         Request

3. Ham: Welcome good friends.     oh       Oh          Recognition
O my olde Friende? (2.2.1468)                          Display

4. Claud: Oh 'tis true:           O        Oh          Level of
(2.3.1701)                                             commitment

5. Ham: Oh, there be players      O        Oh          Speaker
... (3.2.1876)                                         Intensification

6. Ham: O reforme it              O        Oh          Rebuttal
altogether ... (3.2.1886)

7. Bap: "Oh confound the rest"    O        Oh          Rebuttal

11. Laer: It well appeares. But   O        Oh          Elaboration/
tell me, Why you proceeded not                         Pause
against these feates, / So
crimefull, and so Capitall in
Nature, / As by your Safety,
Wisedome, all things else,

You mainly were stirrd vp?

King: O for two speciall
Reasons ... (4.7.3012-17)

12. Claud: Oh he is mad           O        Oh          Clarification
Laertes, (5.1.3469)

13. Oph: Oh you must weare your   n/a      Oh          Clarification
Rew with a difference

Table 6. Results of Emending oh to O

                    Percentage of O    Ohio

Folio Hamlet        29%                80/33
Q1                  100%               0/76
Q2                  95%                5/97
Re-proofed Folio    75%                27/86
COPYRIGHT 2015 Comparative Drama
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2015 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Freeman, John
Publication:Comparative Drama
Article Type:Critical essay
Date:Sep 22, 2015
Previous Article:Langston Hughes and performing transnational presence: Scottsboro Limited and Harvest.
Next Article:Food production in the Croxton Play of the Sacrament.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters