e-mail: rowlandcotterill@gmail.com
https://orcid.org/0000-0002-6041-8578Independent scholar
e-mail: rowlandcotterill@gmail.com https://orcid.org/0000-0002-6041-8578
Keywords: William Shakespeare, early modern stage, doubling, performance, Brett Gamboa
This paper is prompted most immediately by the recent, fascinating and voluminous book by Brett Gamboa, Shakespeare’s Double Plays: Dramatic Economy on the Early Modern Stage (2018). Gamboa, through detailed arguments in the field of theatrical history and dramatic aesthetics, claims that the great majority of Shakespeare’s plays not only can be performed, but were intended for performance, by just twelve actors, these actors being professionals and including young men performing roles of female characters. Gamboa’s arguments are supplemented by charts setting out, on this basis, possible or probable allocations of roles to each actor; for some plays, he claims, fewer than twelve actors are sufficient (Hamlet, for example, requiring only nine). He suggests that such multiple and regular doubling, while serving (as others have suggested) to emphasize thematic motifs staged by regular groupings of individual characters, or by the recurrence of a single dramatic function through more characters than one, also offers, in itself, a major source of enjoyment for audiences and for performers – for audiences, an enhancement of the pleasures afforded by performers’ doubling skills and by their own access to thematic parallelism; for actors, a parading of the virtuosity involved in complex acts of self-presentation. Gamboa writes:
Throughout his career Shakespeare experimented with intensifying the energies inherent in live theatre by adding dimensions to the actor-character, and by imperilling his theatrical illusion by advertising their artifice. Doubling enabled him to achieve both ends simultaneously.… Doubling admits the fiction of the fiction while implicitly arguing for the primal ‘reality’ of the character being cancelled. (11)
Granting that his claims cannot be definitively established, Gamboa appeals, beyond these general and rather metaphysical principles of theatrical form, to specific traits of Shakespearean drama, “Regardless of intent or original practice, Shakespeare engineered plays particularly suited to small companies whose need to play multiple roles would inevitably enhance other paradoxes, contradictions, and replications found throughout the plays” (19). In what follows I shall develop responses to Gamboa’s claims, and to issues of doubling in general, in several ways. After a number of introductory remarks, I shall consider the undoubted necessity, in Shakespeare’s time and within the productions of his company, for regular and substantial doubling of roles. Next I shall discuss a few studies of the topic prior to Gamboa’s book, and shall present, in brief, Gamboa’s arguments, his stipulations, and his claims. After a presentation of my experience of doubling as an audience member, I attempt to negotiate some agreements and disagreements, vis-à-vis Gamboa’s work, at the level of general factors of Shakespearean dramaturgy. Finally I raise and discuss a few of my experiences of doubling as a director of student productions of four Shakespearean plays.
My responses, then, are partly shaped by the place of Gamboa’s work within the history of recent discussion of the general topic of Shakespearean “doubling” (the term is intelligible but far from exact). Gamboa’s suggestions supervene upon the comparable, though differently orientated, study by T.J. King, published in 1992, of Shakespearean actors, roles and castings. They also engage with a number of particular and general proposals about possible or desirable doublings, whether in original or contemporary stagings, within the secondary literature of the past forty or fifty years; perhaps the greatest number of such proposals is to be found in the charts for possible doublings regularly offered at the back of the ongoing third series of Arden editions.
It is worth reflecting at a general level on the resources available for debates about Shakespearean doubling, and on the extent to which these do or do not permit any degree of certainty, or even any sense of agreed hypotheses, in this field. On the one hand, no early text of any of Shakespeare’s plays carries any indication whatever of doubling between roles. On the other hand, as I shall spell out later in a little more detail, doubling was certainly a normal and indispensable feature of performances of these plays by the companies with which Shakespeare was himself involved, and it has been a common, though not universal, feature of performances, whatever their style, in subsequent generations. The need for doubling, in performances, is likely to be reduced in relation to the numbers of actors available within a production or an acting company: fewer actors means more doublings. Some doublings are precluded, one might think, by the simultaneous onstage presence of roles, which therefore require different actors (though, as I shall indicate, even this criterion has occasionally been breached). Other doublings, it has often been claimed, were unavailable to Shakespeare’s and to other early acting companies because they would have involved an actor’s alternation between a female role (such roles being, it is argued, performed only by boys) and the role of an adult male; Gamboa’s book seeks, among other things, to overthrow this claim.
Some doublings, feasible in terms of these familiar criteria, would also have been desirable, it has sometimes been claimed, on grounds of aesthetics – whether in theme (two roles, perhaps, of similar types) or action (comparable narrative functions). These latter issues rely inevitably on matters of critical interpretation as well as physical possibility. Critical interpretation, within the general debate, has been regularly supplemented, or inspired, by the personal experience of critics. Such experience may be, and has perhaps usually been, that of membership of an audience.
Sometimes the experiences of directors, and of actors, in matters of doubling have been broached within critical discussion – sometimes but, I shall suggest, not frequently enough; such experiences merit serious and sustained reflection, and they will form the concluding section of this paper, where I shall present arguments based on my own experiences not only as an audience member but as a director of Shakespearean productions. The actors in these productions were students, and I related to them not as their tutor but simply as their director for each production (across a period of sixteen years – thus very few performers took part in more than one show); neither they nor I received any payment in connection with rehearsal or performances, and to this extent conditions were significantly different from those obtaining in any professional production. Yet a number of issues around doubling, of which I became conscious from the outset of each rehearsal period, seem to me to be salient also for professional options, and for academic interpretation, in this area.
Central to such issues is what I refer to, in my title, as “stage presence”; I have in mind here the sheer fact of an actor, or actors, remaining visible and “present” (and no doubt, given appropriate ability, notably and memorably present) on stage, whether or not speaking. If an actor is not speaking, she or he is silent; she or he is, very probably, within the terms of a role, observing, and – once an audience becomes aware of it – visibly and significantly observing and hearing the speeches and actions of other actors. In this way actors successfully simulate the processes of individual and interpersonal interpretation, both emotional and conceptual, operative within and between the characters they embody. Thereby they give indispensable cues for an audience to follow them, and often to precede them, in developing an ongoing grasp of what, in a play and in the events it stages, comes to matter – of what, in fact, comes to happen. How does this line of reflection come to bear on issues of doubling? In several ways. Multiple doubling of smaller roles, in the repetitive patterns sometimes advocated by critics, can generate, or reinforce, a sense of a principal role beset by mechanically recurrent groups of enemies, or critics, or mockers, or (even) allies – with an effect (among many others) of reducing the focus of interpretation to a single major role. Conversely, an avoidance of doubling (or a play whose staging of roles makes doubling a negligible issue) can result in a sense of multiple agency, sometimes (in the world of the play) fruitfully cooperative, sometimes excitingly contingent and seemingly random. At another extreme, plays whose narrative and whose dramatic style requires constant doubling, without any obvious suggestion of recurrent groupings of roles, allow, given certain doubling options, the emergence of a sense of gradual process, at both individual and group levels, within the field of relationships and collective options. Moreover, advocates of doubling have often suggested, and Gamboa often suggests, that silent on-stage roles, in a particular scene or scenes, might be eliminated – that is, their actors might remain off-stage – with the effect, or even the purpose, of enabling doublings not otherwise possible; here I shall suggest, by contrast, that the silent presence of a role, and its actor, can contribute importantly, even decisively, to a proper understanding of the narrative and the emotional import of a play.
Nobody doubts that Shakespeare’s own theatre companies relied on the doubling of roles. If one considers merely the numbers of speaking roles in the plays, the limits are marked by The Two Gentlemen of Verona, with sixteen, and Richard III, with sixty-four. These are early plays, and subsequent conditions of company organisation may not be applicable to their original conception and staging. In Shakespeare’s mid-career 2 Henry IV contains fifty speaking roles, Twelfth Night only twenty, Othello twenty-four. Among later plays, Antony and Cleopatra and Coriolanus involve over fifty speaking roles, The Tempest just twenty. Such an extremely wide range may in itself put in question any theory positing a single (small) number of available and professionally qualified actors as the dramatic vehicle for each and every mature play. Gamboa excludes from consideration four early and two late plays – the three Henry VI dramas and Titus Andronicus, Henry VIII and The Two Noble Kinsmen – on the grounds, it seems, of different arrangements for company organisation likely to have been operative at these periods; he might have also mentioned, given these particular omissions, the likelihood that these plays were jointly authored – yet the same could be said of Timon of Athens, while Two Gentlemen is surely early enough to reflect, also, different company and casting conditions.
The Swiss traveller Thomas Platter, who saw Julius Caesar at the Globe in 1599, mentioned its “15 personen”; since that play has fifty-one spoken roles, Platter is likely to have been referring to the number of actors on stage (as, in the scene of Caesar’s assassination, they all are) (De Grazia and Wells 106). Some specific cases of doubling in early seventeenth-century plays are on record; this is not the case for any Shakespearean play, but the complexity of his story-lines (often involving “double plots”) and the large number of small roles, in every play – roles appearing in only one or two scenes – makes doubling an undoubted prerequisite for any conception of original and early professional performances. Different issues in turn are raised under theatrical and cultural conditions where a large pool of competent and experienced performers is available and money is no object. Sometimes, and in some places, this has been the case in the twentieth and twenty-first centuries (as it was occasionally in Britain in the nineteenth century); in these more recent times, also, many companies (though far from all) have been able to count upon a large and competent crew of supporting “backstage” staff – an issue which should not be neglected; more actors require more offstage resources, while a tightly-reduced cast needs tighter backstage organisation and stage managers with a good head for logistics (it’s not clear to me, from Gamboa’s suggestions, that his twelve-actor group would have the time and energy necessary for such responsibilities). Be this as it may, almost all the professional Shakespeare productions I have seen over the past sixty years have opted for some degree of doubling.
With amateur productions, different constraints apply. Sometimes few competent actors are available and doubling becomes desirable, even necessary. Sometimes there are good reasons for involving as many performers as possible; even here, financial limitations are likely to operate. It would be good if the increasing interest in such amateur and student Shakespeare productions were to focus on this issue.
Professional companies, obviously, have large financial interests at stake in Shakespearean performances, even where these (as in many European countries) receive state subsidy. Gamboa lays much emphasis on the economic advantages available for a company, of Shakespeare’s time, maintaining the limit of twelve actors – the fewer they were the larger the profits available to each individual. Ellen Summers disagrees, claiming that the company was successful enough to have no financially compelling reason for multiple doubling – which, in her view, was maintained essentially because it was an old performing tradition (65). There seems room here for a more chronologically nuanced approach than any scholar has offered; but the evidence is likely to be recalcitrant, or simply absent.
In this whole area, one might usefully seek responses from actors, from audiences, and from directors who have enacted, watched and planned doublings. Such responses might touch upon sensitive points, since they involve decisions about, and the impact of, options which are not normally transparent to all the interested parties – above all, the perceptions by actors and directors of each other’s abilities. Nonetheless, this approach would surely be worthwhile. Yet, to my knowledge, it has not been adopted to any salient degree in academic studies (Gamboa does usefully refer to some of his experiences as a director.)
The critical debate about doubling – it has in fact been relatively sporadic, for understandable reasons – has focussed, in material terms, upon questions about the size of Shakespeare’s company; how many full-time members or “sharers”, how many non-full-time members (“hired men”, “apprentices”), how many “boys” and (importantly) of what ages and with what expectation of subsequent absorption into full company membership. Other material issues have also, rightly, been considered; how rapidly, upon what different types of stage, could an actor, leaving the stage, change costume (and/or make-up) and assume a different role (and, I would add, with what degree of help from off-stage non-actors)? Different issues in turn are raised in the area of what one might call performative psychology; how capacious was the memory of an actor, how limitless his physical stamina? More elusively: how far does an actor gain, how far does an actor confer upon audiences, positive enjoyment by doubling – and whose enjoyment (or otherwise) counts most? And then, at a more structural level – here discussion becomes less “material” and more ”thematic” – how might a playwright who was also a man of the theatre use doubling to imply or reinforce certain themes, certain patterns of action, certain piquancies of staged encounter? In turn – and here material and thematic considerations may be allied – can doubling between minor roles confer extra interest on “bit-parts”? among other possible advantages, in this area, for a company, one point should not be neglected; actors tend (in all ages) to read and study, first and perhaps only, their own scenes. Doubling, which gives an actor more scenes, offers greater chances for actors to grasp what, in a play overall, is actually going on.
One very readable and provocative study of doubling, by Stephen Booth, argued forty years ago that the practice reflected Shakespeare’s sense of “the unsettling but enriching effects to be had had from making an audience’s two incompatible consciousnesses indivisible” (104).
Booth’s preferred options were for doubling between roles in cases involving early deaths – Duncan doubling with Macduff, Julius Caesar with Octavius Caesar, Gaunt with Northumberland, Mamillius with Perdita. Each of these possibilities would, along with sheer feasibility and possible convenience, convey meaning – revenge, sustained alliance (Northumberland taking up Gaunt’s animus against King Richard), or restoration (a son lost but a daughter, against all the odds, surviving). Booth also desiderated more complex possibilities; Mercutio with Paris and Prince Escalus (all related, though an audience is likely to discover this, if at all, only retrospectively); Camillo with Antigonus (both marry Paulina); or, Archidamus-Antigonus-Autolycus (alliteration? – it’s often suggested that Autolycus might double with – no less than – Leontes); Antony and Dolabella (the issues here are interesting, but there are problems at the material level of timing); and, perhaps the critical favourite here, Posthumus and Cloten.
Another favourite with critics, Cordelia and Lear’s Fool, has been often advocated, sometimes by critics with no special interest in doubling as such; it has also been opposed – partly on the grounds that Robert Armin (if he played the Fool, as in other plays of the period he did) would have been a weird Cordelia, and, conversely, would have refused to relinquish to a “boy-actor”, playing Cordelia, his tailor-made Fool’s role. Ralph Berry, among other sharp points in this area, remarks that “One cannot erect a multi-storey edifice [of critical interpretation] on the bald fact that two characters never meet on stage” (17).
Similarly, Berry sees the Posthumus-Cloten doubling idea as “at best heavy and didactic” (18) (I have seen it done, and it wasn’t; but the production deployed just five actors for a textually complete performance of Cymbeline, not a circumstance likely to have been seen or envisaged by either Booth or Berry.) Of the much-cited doublings, in Peter Brook’s 1970 RSC Midsummer Night’s Dream production and subsequently, of Theseus with Oberon and Hippolyta with Titania, Berry observes that “What emerges … is a sense of the emotional tensions between the lead actors, in both roles: pre-marital, post-marital” (20). That is (I take it) the tensions are comparable – which doubling may emphasise – but distinct – which doubling may blur. In this same spirit Berry writes, “Doubling exploits likeness. What of unlikeness, that obstinate particularity which … is true of all Shakespearean roles?” (22). The thought seems congenial; yet “unlikeness” certainly, in Shakespeare’s dramatic practice, did and does coexist with considerable amounts of doubling of one kind and another.
Amongst other specific critical suggestions, contributions by Alan Armstrong and, again, Ellen Summers deserve comment here. Armstrong notes a possible metatheatrical reference, in Richard II, to characters who have recently died (Bushy and Green) addressed to a newly-present character (Scrope), and suggests, quite plausibly in my view, both that doubling is involved here, and that such cross-reference, by or concerning one character, to another, may generally mark cases of doubling (similarly, in his view, the dead Mowbray, or Norfolk, referred to, as dead, first by the Bishop of Carlisle, may have doubled roles with Carlisle) (152). The two cases are interestingly different, involving, respectively, small and medium-sized roles; and certainly the role of Mowbray requires a strong actor whose abilities could be used effectively after he leaves the play at the end of Act 1. Summers, for her part, notes the recurrent appearances, in Henry V, of groups of three characters – three English traitors, three royal English brothers at Agincourt, three English soldiers whom the King meets by night, and three “regional” Captains, Fluellen, Jamy and MacMorris (74). Doubling here, I would agree, seems likely; its effects would bear reflection. Does King Henry effectively work to unify such “threesomes” into a coherent “band of brothers”? Do they, rather, serve to highlight in Henry a dominant and even overbearing monopoly of effective agency? One would like to know more about doubling options amongst the play’s French roles, and between them and the English.
Gamboa’s own study, far more voluminous than anything I have cited, considers all these issues, material and psychological, theatrically collective and authorially distinctive. In his seven main chapters he offers detailed discussion of doublings that are possible, probable or desirable in five plays; three comedies, Midsummer Night’s Dream, Twelfth Night and The Winter’s Tale, and two tragedies, Romeo and Juliet and Othello. He offers, as I have said, doublings for twenty-seven other plays in an Appendix. As controlling principles for doubling he suggests two: “characters doubled by a single actor cannot meet onstage”, and “a performance will indicate changes of person by changes of costume”. He emphasises that “Though I expect that Shakespeare’s historical company regularly put on plays with nine to twelve actors speaking all or most of the lines, I cannot prove it” (19).
Gamboa’s modesty about proof is, as can be seen by now, par for the course, given a general lack of hard evidence. The challenges of his work are vested in the small number of actors for which he argues, in the small number of principles (positive or negative) on the basis of which he relies, and in the theatrical practices (above all concerning the ages and uses of “boy-actors”) which, given independent supporting arguments, he sees as consistent with his other claims. It is worth stressing what Gamboa (unlike some others) does not claim; he sets no lower time-limits for offstage costume changes; he does not seek to avoid any doublings, whatever the size of the roles they may involve, which may be, in his terms, practically possible; and he sets no limits on the memory of Shakespeare’s actors, nor on their stamina (one notes his claim that a performance of a “complete text” of Hamlet needs no more than nine actors).
Above all, he refuses to distinguish, within his 12-actor hypothesis, between the powers, and the doubling availabilities, of adult male actors and what it has been customary to regard as boys or apprentices habitually performing (as has usually though not always been understood) all and only female roles. He also overrides distinctions, traditional amongst theatre historians, between “sharers”, “hired men” and “apprentices”. For him one should conceive of Shakespeare’s company as including a number of “young adult male actors”, aged between about thirteen and about twenty-five, capable of playing, in any one play, both female and young male roles. This would enable a number of doublings not always considered feasible: Ophelia with Osric; Maria with Sebastian; in Cymbeline, the Queen with Iachimo. Such proposals resonate to some extent with recent advocacy of “gender-blind” casting, and they may divide responses along similar lines. I should perhaps say, while claiming no expertise in early English theatre history, that Gamboa’s proposals in this area seem to me fascinating and even plausible, whatever their bearing on doubling issues.
I have already indicated some of the positive advantages which Gamboa claims for a widespread and thorough-going practice of role-doubling in Shakespeare’s plays. He sees it as both in keeping with and offering enhancement for a “fundamental paradox” that actors are and are not the roles that they play. For him it allows “thematic patterning and resonance” “through the unifying agency of a single actor” – for example, one playing both Polonius and the First Gravedigger (this doubling was once more popular than it is nowadays, and the “Gravedigger” role may have been assigned to an actor regularly playing Fools or clowns). Doubling allows “heroes to re-emerge as villains, men to become women, and fools to become wise” (4). Would the Polonius – Gravedigger doubling be a case here? If so, which way round? – For myself I doubt whether the “hero-villain” option applies, either way, to many of the plays; men becoming women, and vice versa, on the other hand, is commonplace within single roles, and its effect might be enhanced, or might be undermined, by more widespread adoption across roles – again, one would like evidence from performers and audiences here.
I’ll move on, therefore, to present a few of my own experiences as an audience member, before closing with discussion of my experiences and reflections, around doubling, as a Shakespearean director. In 1973 I attended a late revival of Peter Brook’s Dream; I can vouch for the accuracy of Stephen Booth’s perception that the doublings of Theseus-Oberon and Hippolyta-Titania worked not only feasibly but superbly. Booth dwells on the moment – and it is no more than a moment, indeed in clock time much less – in Act 4 scene 1 when the exit of the fairy couple is followed, after a mere horn-call, by an entrance for the Athenian royal pair; as Booth says, the actors carried this off with exuberant and smiling self-confidence. Any change of costume could only be of the most minimal (head-gear?). In a sense this case supports Gamboa’s view that doubling hypotheses need allocate no particular length of time to changes of costume; on the other hand it perhaps qualifies a little his actual requirement for costume change between different roles sustained by a single actor, for the “new” identities of Theseus and Hippolyta were, in my viewing, established not by costume but by inescapable performative bravura – one knew that something wonderful had been carried off and then one realised what that wonderful thing was.
A few years later the John Barton RSC production of Richard II, most famous (in this area) not for doubling but for alternation of roles – Richard and Bolingbroke were played, alternatingly in different shows, by Richard Pasco and Ian Richardson – offered, also, a notable small role-large role doubling; the Bolingbroke actor (whichever) played, also, the Groom who, in Act 5 scene 5, visits the imprisoned King. I continue with the account given by Stanley Wells:
The groom threw back his hood and revealed himself as Bolingbroke in disguise …. The director seemed intent on suggesting a recognition on Bolingbroke’s part that both he and Richard have been the playthings of fortune … their shared experience of the hollowness of the kingly crown draws them together more powerfully than their former rivalry sets them apart …. It was a theatrically impressive moment … I confess all the same that I found it strained … in the theatre we could not help identifying the actor as Bolingbroke. The confrontation seemed to demand an explanation that was not provided by the dialogue. (79–80)
This case puts some pressure on Gamboa’s insistence that doubling be accompanied and to some extent indicated by costume-change. How would Gamboa handle the distinction between a Feste, or a Duke Vincentio, in a friar’s disguise, and a doubling, in Much Ado about Nothing, between, say, Antonio or Conrad and Friar Francis? Wells rightly stresses the audience’s awareness of the Groom as, simply, Bolingbroke; that is, we did not suppose ourselves to be watching a case of doubling. In Gamboa’s casting chart for the play the role of the Groom is doubled with that of Queen Isabel – a comparable case, perhaps, of ambiguous comfort offered to the King. Wells’s perception of “shared experience” and convergent relationship between former and present Kings may place, in fact, too much emphasis on comfort, as against irony or mockery. At all events, the effect of the moment, which was very powerful, could not be separated from the “linkage” between the two leading roles of the play already established, in this production, by the alternation of actors. Gamboa does not consider possibilities, in Shakespeare’s company, for such alternation – it has been practised several times in more recent theatrical history, most famously between Gielgud and Olivier, in a single production, as Romeo and Mercutio at the London New Theatre in 1935. Alternation, at least for actors, arguably offers many of the advantages claimed by Gamboa and others for doubling.
In 1979 I saw a touring production of Cymbeline staged by the company Shared Experience (one notes Wells’s use of the phrase, just quoted). Only five actors enacted a complete text. The doublings were as follows: Cymbeline-Iachimo; Imogen-Cornelius; Posthumus-Cloten-Belarius (a multiple role of well over one thousand lines); Queen-Philario-Cadwal-Philharmonus; Pisanio-Polydore-Lucius. Other roles were played by “all”. Moreover, while a clear playing area (three-sided) was marked out, all actors, when not enacting their several roles, remained visible, on benches placed to the sides, as watchers of the stage action which was also being watched by the rest of us in the audience. The result was, for me, an immensely fruitful sense of each role as absorbing, in some measure, the play’s events in their narrative sequence and through the relationships in which that sequence consists – in some measure but never adequately, for their receptive and reflective powers necessarily alternated with their commitment, while in their onstage roles, to agency and to its inevitable attendant levels of relative blindness to the agencies of others. In a play with a narrative as complex as that of Cymbeline, this effect was, for me, thought-provoking and revealing.
At a more performative level, while the specific doublings allowed an avoidance, to a remarkable extent, throughout most of the play, of “encounters” between roles played by a single actor, such encounters were, predictably, endemic during the play’s remarkable final scene of multiple mutual revelation. Here, then, Cymbeline (the actor John Dicks) addressed, in anger and scorn, (the actor John Dicks as) Iachimo; and Posthumus the King’s son-in-law shared the body of the actor Raad Rawi with Belarius, secret guardian of the Kings’ two sons. The effects were not in the least confusing; for the roles had been established, long since, in their several identities for an audience by linguistic style, bodily gesture, and above all by narrative coherence. No costume changes were needed, and very few were deployed. All this, certainly, might not work so well in other Shakespearean plays.
As an audience member, I was conscious that the Cymbeline production placed immense strain on its five actors’ powers of memory. No doubt this is just something that actors do; but one wonders how a slightly larger group of actors, on Gamboa’s hypotheses, would cope, if asked to sustain, not only doubled roles in a single play, but such doublings across perhaps ten plays (in the kind of repertory system the Shakespearean company probably sought to maintain) all at once. There will surely have been illnesses; there seem to have been no regular understudies. Not every actor will have been a competent singer or instrumentalist, and the incidence of roles requiring such skills is a constraint upon doubling which Gamboa does not negotiate in any detail. Moreover, twelve actors represent a small group on the basis of which to stage armies (as in most English and Roman Histories) or citizen bodies (as in Julius Caesar and Coriolanus). For these and other reasons – to cut to the chase – I would myself see Gamboa’s fascinating study as offering grounds, on the one hand, for a newly flexible approach to Shakespearean doublings, especially those involving gender-difference; while, on the other hand, underestimating the value, for many distinguishable purposes, in the staging of these plays, of a company larger than twelve performers.
Here I will briefly place some stress on a specific principle used by Gamboa, Booth and others: that involving “early deaths”. I have already noted Booth’s suggestions here, and considered their possible effects. But in King John three deaths, each before or soon after midway – Austria, Eleanor and Constance – seem to allow no specially significant subsequent doublings; the same seems to me to be true for the death of Cornwall in King Lear. In Richard III the role of Edward IV can usefully double with that of Richmond – but the actor will still be offstage for most of the play. In King John such “late-arriving” major characters as Pandulph and Hubert allow no significant doubling with earlier departing roles (the debate about whether “Hubert” is in fact also the “Citizen of Angers” rather underlines this point).
My reason for stressing such cases is this: there are important dramatic advantages in a character’s perceived absence from the stage. This is the obverse of my earlier point concerning the importance of an actor’s extended stage presence. Actors, no doubt, need breaks, especially in leading roles (and it’s often noted that tragic Fourth Acts tend to provide such breaks for their protagonists). But audiences also need breaks, if not from actors, then from their roles. I think it’s very helpful for Posthumus not to be seen for two whole Acts – and this effect is surely enhanced if the actor of the role, also, is not seen; similarly, I would judge, with Leontes. Even with smaller roles such a notion has some force; in Antony and Cleopatra the roles of Pompey and Lepidus, important for the first half of the play, thereafter vanish from the stage and will certainly be doubled – but, I suggest, not at once; the absence of the characters, as political forces, needs to be carefully registered by an audience, since it is the force driving Antony and Caesar into seemingly inescapable mutual confrontation. Similarly, “late arrivals” add more novelty in tone and energy if they have indeed not been seen before – one thinks of Holofernes and Nathaniel, and in 1 Henry IV of Glendower and Mortimer, or Douglas and Vernon (though perhaps not of all four of them).
A larger point, in connection with the use of “early deaths” as a criterion for hypotheses about doubling, is this: some of Shakespeare’s plays seem to have accepted, with regard to their narrative ordering, some constraints from “real history”, or of what was presented as such by Shakespeare’s sources. Such constraints, whether or not they enable doubling, clearly involve considerations of a quite different nature. Gamboa, who doesn’t pursue such considerations, offers, in my view, few new persuasive suggestions concerning English or Roman History plays (five of which are, by his self-imposed remit, excluded). Two cases suggest some of the factors at issue here. Julius Caesar must die at or near the mid-point of the play bearing his name, since Shakespeare’s play is concerned as much, or more, with the impact of his death as with the motives of his killers. Coriolanus, by contrast, must survive all his battles, and so must all his enemies and his tormentors (who include his supporters). Hence, in the earlier play, many doublings are possible – given the play’s bipartite structure – while none is, in isolation, necessary; in the later play, I would say, no doubling between any significant role is even possible, and the effect of the continuing presence of all characters is crucial and determinative for the play’s effect.
By contrast with these two plays, one finds, in the co-authored Titus Andronicus and 1 Henry VI, historical fantasies, in which the playwrights could and did arrange for multiple deaths with great freedom; still, while these plays certainly require much doubling, no particular options suggest themselves with any clarity or theatrical advantage – nor do the plays’ plots avoid a suggestion of randomness and inconsequence. Two different cases again appear in Macbeth and Antony and Cleopatra. Here Duncan’s early death allows a “significant” doubling of the role, while such a doubling of Antony’s role, enabled if not enforced by the extent of stage time (a quarter of the play) remaining after his death, nonetheless, in my view, lacks merit; in any case arguments, around these plays, in the area of doubling need to acknowledge the priority, in terms of dramatic structure and effect, of quite different factors. Most plays based on the subject-matter of Antony and Cleopatra employ smaller casts and start, in historical terms, much later; as Shakespeare’s choice here has vast effects on his presentation of theme and character, so it would be limiting to consider it chiefly in terms of the options for doubling which it offers. Similarly, in Macbeth, a Duncan surviving and reigning up to the midpoint of a hypothetical play would have effects, upon the presentation of Macbeth and of Scottish society, far outreaching the issue of doublings which might be, by such a different dramatic option, enabled or precluded.
The case of Hamlet, where the storyline had been often and variously dramatised before what we have as Shakespeare’s treatment, is also illuminating, particularly in view of Gamboa’s suggestion that as few as nine actors can manage the play (in any textual version). Certainly the play is full of pairs (as Henry V is full of trios) which can double with each other – Barnardo and Francisco, Voltemand and Cornelius, Rosencrantz and Guildenstern, two Gravediggers, Osric and the Lord. It also offers, among its larger roles, doubling options of some possible significance – Ghost with Player King (or Claudius?), Polonius with First Gravedigger, Rosencrantz (say) with Fortinbras. Only Hamlet, Gertrude and Horatio cannot double. Suppose all or most of these options were adopted; my sense is that the play would then be likely to communicate a sense of haunting, of oppression both political and emotional, and ultimately of determinism. It would become, in fact, more like Macbeth – and more like its acknowledged modern successors, Waiting for Godot and Rosencrantz and Guildenstern Are Dead; more like such plays than it in fact is.
In my view there are dangers in attempts to assimilate, to such minor modern masterpieces as these, the precursor which it is their enterprise and wager to transform and rewrite. Still, such conceptions of Shakespeare’s play may attract some more than others, or anyone at some times more than other times. Schools of thematic, of imagistic, and of structural criticism have tended to acclimatise us to the habit of construing sequences of dramatic events in terms of repeated topics, recurrent metaphors, and energies balanced around irresoluble ambiguities. Such construals may, by their nature, do less than justice to the commitment of Shakespearean drama to the presentation of events, of causal connections, of surprising transformations, and of one-sided but irreversible outcomes. Hamlet, more explicitly than most plays, invites attention to “purposes mistook/Fallen on the inventors’ heads”.
Now certainly, and by contrast, Macbeth, in his tragedy, ratifies a different experience, of temporal sequence as reduced to empty “tomorrows”. Antony, for his part, approaches suicide with the sense that, Cleopatra once (supposedly) dead, “All length is torture” and that “all labour/Mars what it does”. Shakespearean drama is amply capable of envisaging a condition in which causal logic and temporal sequence are obscured, abandoned or annihilated. Indeed the act, and the prevailing cultural circumstances, of theatrical performance can themselves seem to embody just such a condition. So much happens – people die – given enough doubling no death need be final – in the end everyone stays alive (and they all take a curtain-call); was anything ever done? These are major issues, in fact daunting issues, for critical interpretation. It seems to me unwise for them to be arbitrated, whether in intention or in practice, through hypotheses in a field as uncertain as that of Shakespearean doubling.
It was in connection with my production of Macbeth that I came to feel the force of such a possible theatrical condition – the condition of “nothing happening”; to feel it and, at least in subsequent productions, to resist it. I had opted, for Macbeth, to stage Charles Marowitz’s version; this contains not only three Witches but three Macbeths, and its Lady Macbeth is in league with the Witches to the point where she operates as one of them. The relatively independent identities of the remaining roles – Duncan and Banquo, Macduff and Malcolm – did not banish, in my production, a sense of immense over-determination; the three Macbeths argued each other to a standstill while the forces of darkness drove events towards a repetitive and nihilistic sense of evil. This, of course, is a sense often communicated to some extent by productions of Shakespeare’s full text with few or no doublings. It was a sense that, in choosing King Lear for a second Shakespearean production, I hoped to avoid.
Insofar as I succeeded, the outcome was, I take it, due to the nature of the play far more than to any features of my own production. As far as doubling issues are concerned, King Lear offers, I believe, little of interest. Gamboa’s chart, while accepting the Cordelia-Fool option, doubles only two named characters, Burgundy-Edmund and France-Edgar. (I think the reverse option might be less morally deterministic.) The point – at least one of the points – about this overwhelming tragedy is that a lot happens. For many critics, too much happens – over-complicated double-plotting, loose ends of causality, and certainly plenty of purposes mistook. At the end some things will never be the same. The play begins with three feisty young women very much alive and ends with three exhausted men on an otherwise bare stage looking, or trying not to look, at their corpses and at the corpse of their father. Nobody intended this. Doubling seems counter-productive, thematically supererogatory, and, in terms of performative virtuosity, emptily ostentatious and self-serving.
For Antony and Cleopatra I was able to find just sixteen student actors willing and competent (and much more) to take part. The leading roles, it’s often been noted, are both glamorous and in many ways, for actors, invidious; and doublings, completely necessary, are possible in an almost endless variety of ways. Few professional productions that I have seen have used more than twenty actors; but, for casting this play, the difference between twenty and sixteen is huge. In any case, given the number of extremely brief roles, it was important to keep everyone interested, and aware of the storyline, throughout the six-week rehearsal period. I began with an awareness that certain roles could not, or should not, be doubled; certainly Antony and Cleopatra, Caesar and Enobarbus, but also Agrippa and Maecenas, Charmian and Iras (Gamboa’s chart doubles all the roles in the play.) I also opted for doublings of Lepidus with Eros and of Octavia with Dolabella (thus a female with a male role). I should note that the cast included seven male and nine female actors.
One principle guiding me was this: the play is obsessed, even more obviously than Henry V, with groups of three characters – three leads (the titular lovers and Caesar), Cleopatra and two attendant women, Caesar and two aides. Antony by contrast has many followers – and loses almost all of them. An audience needs to see these things; to see triads, to see sudden solitude. Another principle was this: four roles, relatively or absolutely short in terms of spoken lines, require, on the other hand, extended times of presence on stage. Maecenas and Iras, above all, are silent (only slightly less so are Charmian and Agrippa) for long scenes, and are set on stage for short scenes in which they say little or nothing. Why, then, are they there? In order (I would say) to register, and to be seen to register, the rhetoric, the manoeuvres, and the ultimate purposes of their “principals”, Caesar and Cleopatra.
My point is not that Maecenas or Iras “undercuts” or effectively queries the policy of Caesar or the mood-swings of Cleopatra. Rather, by their presence they draw an audience’s attention to these things – policy and mood – for what they are; they expose them to reflection. Such reflection may in turn become critical, whether for them or for others. It may also become supportive; this is clearly the case with Iras, whose loyalty to Cleopatra reaches the point of sacrifice. Such largely-silent characters, if undoubled, can and do also carry the weight of an important feature of the play’s narrative – which I might call “contagion”, negative and positive; Antony gains and then loses friends, Cleopatra commands hearts, Caesar sways wills. Thus, while Antony’s followers mostly become Caesar’s subjects (an effect well noted, in connection with doubling of roles, by Summers), some characters become prominent by, as it were, going the other way. Eros, from mere interlocution with Enobarbus, emerges as Antony’s closest friend – himself contagious in that his suicide triggers Antony’s attempt. Dolabella, most notably, is led, from his initial tone with Cleopatra of casual flirtation, into positive alliance with her purposes, and treachery to those of Caesar. My chosen doublings, which claimed no visible merits beyond mechanical possibility, might allow a sense that “Lepidus”, rather brutally dismissed by Caesar, could appropriately re-emerge, as “Eros”, in lasting closeness to Antony; and that Octavia, Caesar’s sister and diplomatic tool, might, as “Dolabella”, become a tool turning in his hand.
Troilus and Cressida, among my four Shakespeare productions, involves, for casting and doubling, very different issues from Antony and Cleopatra. On Gamboa’s estimate the play deploys thirty speaking roles; my production, again with students, took place at the end of a summer term – that’s to say, after the end of examinations – and the auditions produced thirty castable actors (Again at least half were female, and many of these, in my production, played Trojans.) But, for Shakespeare’s company, using far fewer than thirty actors and possibly fewer than fifteen, the play will have caused unusual problems – I would say, problems unique in the whole oeuvre. Act 4 scene 5 – that is (since enumerations of scenes differ in different editions) the single long scene, as I take it to be, running from Ajax’s armed appearance for single combat up until the general clearing of the stage after Greek and Trojan leaders encounter each other in conversation – this scene most clearly focuses these problems.
Seven Greeks enter (with “a trumpeter”); next Diomedes arrives in the Greek camp, with Cressida; as they leave, five Trojans arrive (the Folio text names only four, but omits Troilus, who speaks later in the scene); soon Diomedes returns. This seems to amount to at least fourteen actors. Can Cressida double? Surely not – but technically she can do so, just about, with the small role of Helenus. Are all the Greeks needed on stage? Certainly; this is one of the play’s decisive moments, with Achilles and Patroclus leaving their tent behind and joining, at least numerically, the core group of five “loyal” Greek “leaders” (which includes both the young Diomedes and the mainly silent Menelaus) to confront a sizeable party of Trojans. Are all the Trojans needed? Helenus says nothing, which might support the idea (surely a weird idea) of the role’s doubling with that of Cressida; and Paris says nothing. But, without Paris and without Helenus, the Trojans, reduced to Hector and Troilus (Aeneas being active as an umpire of the duel), are seriously deficient in numerical terms as they confront the Greeks. Would this matter? Yes; the play has moved, from the staging of four small and largely distinct groups – Trojan lovers, Greek leaders, Greek absentees, and the Trojan royal family – towards this scene, where all these groups encounter, in formal duel, in free conversation, and eventually in verbal violence. This, if anything, is what happens in Troilus and Cressida – or rather, it is the beginning of many things that now and hereafter begin to happen at great speed.
Moreover, Act 4 scene 5, full of leading named and speaking characters, still lacks two of the play’s major roles, Pandarus and Thersites. Each of them can double – just about (significant here might be the relative silence of the role of Menelaus, though, as in Antony and Cleopatra, his stage presence as observer seems valuable, since after all the war is all about him and his former wife). Is it helpful, to the play, if either Pandarus or Thersites double? Gamboa assigns Thersites a doubling with Cassandra; this would leave the actor offstage for seven scenes, while still requiring a quick costume change after Cassandra’s second scene. He doubles Pandarus with Menelaus; this option would be much easier for the actor if there were an interval in performance after Act 3 scene 2. In modern productions this is often the case, but its various advantages fail to demonstrate that it would have been the case, or even conceivable, in Shakespearean productions.
An important rider here is this; Gamboa’s doubling chart takes Patroclus offstage early in Act 4 scene 5 (he suggests that the actor leaves along with Cressida and, presumably, Diomedes – the timing of Cressida’s exit has in itself been seen as controversial on both textual and theatrical grounds). To me it seems vital, on the other hand, for a sense of what, in the action of the play, is at stake for Patroclus and for Achilles and for their relationship, that Patroclus should be present on stage until the end of the scene; so that he may see, and consider, and remain movingly silent in the face of, Achilles’ homoerotic love-hatred, as it is eventually and explosively aroused towards and against Hector. Once again; silent presence is crucial, and doublings serving mere convenience, or even hypotheses of financial convenience and limitation, should not unthinkingly be given the preference against it.
Thus, finally, in my view, doubling is, among many factors determining or stimulating Shakespearean dramaturgy, one factor, with its own advantages, its own limitations, and its own claims, in the assessment of that dramaturgy, for consideration along with many other factors. I have tried to foreground, on the basis of my experience as audience and as director, the factor of what I have called silent stage presence. Concern for the visibility of such presence would, in my view, lead to an acceptance of the value, for a hypothetical reconstruction of Shakespeare’s options, of a group somewhat larger than twelve in number. (For Troilus and Cressida, leaving aside issues of individual performing skills, I would have been happy to work with the sixteen actors available to me for Antony and Cleopatra.) Such a larger group would, also, accommodate the important effects noted by me and by other critics and observers; individual uniqueness, cumulative action involving increasing numbers of participants, and, with both of these, the phenomenon, within a political and personal situation – the kind of situation central to all Shakespeare’s historical dramas – of, at both individual and group levels. a gradual change of allegiance.
Armstrong, A. “‘What is become of Bushy? Where is Green?’ Metadramatic Reference to Doubling Actors in Richard II.” Inside Shakespeare. Essays on the Blackfriars Stage. Ed. P. Menzer. Selinsgrove: Susquehanna University Press, 2006, pp. 149–155.
Berry, R. Shakespeare in Performance. Castings and Metamorphoses. Houndmills: Macmillan, 1993.
Booth, S., “Speculations on Doubling in Shakespeare’s Plays.” Shakespeare. The Theatrical Dimension. Eds., P.C. McGuire and D.A. Samuelson. New York 1979, pp. 103–131.
De Grazia, M., and Wells, S. The Cambridge Companion to Shakespeare. Cambridge: Cambridge University Press, 2001.
Gamboa, B. Shakespeare’s Double Plays. Dramatic Economy on the Early Modern Stage. Cambridge: Cambridge University Press, 2018.
King, T.J., Casting Shakespeare’s Plays: London Actors and their Roles 1590–1642. Cambridge: Cambridge University Press, 1992.
Marowitz, C., ed. and trans. The Marowitz Shakespeare: Adaptions and Collages of Hamlet, Macbeth, The Taming of the Shrew, Measure for Measure and the Merchant of Venice. London: Marion Boyars, 2000.
Summers, E., “A Double Heuristic for Shakespeare’s Doubling.” Staging Shakespeare. Essays in Honor of Alan C. Dessen. Eds. L.C. Orlin and M. Johnson-Haddad. Newark: University of Delaware Press, 2007, pp. 60–78.
Wells, S. Royal Shakespeare. Manchester: Manchester University Press, 1977.
Received: 20.02.2022. Revised: 30.05.2022. Accepted: 12.06.2022.
Received: 20.02.2022. Revised: 30.05.2022. Accepted: 12.06.2022.