From British and Irish Drama 1890 to 1950: A Critical History
by Richard Farr Dietrich
 
Link to Title Page & Table of Contents for Entire Book
End of Chapter 1
Link to Chapter 2

 

CHAPTER 1

 

INTRODUCTION:

A RENAISSANCE OF THE DRAMA

 

There is a deep-lying struggle in the whole fabric of society:

a bounding grinding collision of the New with the Old.

—Thomas Carlyle

Table of Contents

1.    The New Drama

2. Realism and the Reaction

3. The Well-Made Play and the Problem Play

4. Shavian New Drama

5. The Stage and the Age

6. Figures 1-7

 

 

 

 

THE NEW DRAMA

                    TOP

 

There have been two periods of great drama in British history, the first in “the Renaissance,” Shakespeare’s age, and the second, confusingly called “the Renaissance of the British Drama,” featuring George Bernard Shaw and the New Drama.1   It is this renaissance of the modern period, roughly occupying the years 1890 to 1950, that is the subject of this book.

 

         William Archer (1856-1924), the most influential drama critic of the New Drama movement and translator of Ibsen, thought of the ages between the Puritans’ closing of the theaters in 1642 and the creation of the New Drama in the 1890s as the dark ages of the drama, with only a few glimmerings of light along the way—Congreve, Wycherly, Gold­smith, Sheridan, Robertson—to give hope for the future. Throughout The Old Drama and the New (1923), Archer used metaphors of light and dark or wasteland metaphors to contrast the New Drama with the Old (“the whole century from about 1720 to 1820 was a dreary desert broken by a single oasis—the comedies of Goldsmith and Sheridan”), metaphors he applies to most of nineteenth-century drama for its pleasure-seeking addiction to melodrama, low comedy, and other escapist fare.2  

 

To read the diatribes against the nineteenth-century theater by certain critics and dramatists of the 1890s is to be reminded that nothing really changes in popular culture. The most debased of our own film and TV fare is a lineal descendant of nineteenth-century popular theater, except that the ante on thrills and “laffs” has been considerably upped, making the Victorian plays complained about by Archer seem tasteful and thoughtful. Yet there’s no doubt that this escapist, often simplistically moralizing drama of the nineteenth-century popular theater was of a decided mediocrity, not only because it catered to the low tastes of a very undereducated and rather uncouth audience but because it was virtually without literary value. This can partly be attributed to the fact that without copyright laws protecting them playwrights had gotten out of the habit of publishing their plays (except as prompt books) and thus of thinking of them as literature, subject to criticism. Archer’s insistence on literary quality had much to do with the return of substance to British drama, as well as a return of improved technique. Perhaps the most significant feature of this period is that in it the literary drama overtook the old Theatrical Theater, making necessary a critical approach fundamentally literary.  

 

         But Archer’s partisan condemnation of nineteenth-century drama must be qualified in several ways. First, though the nineteenth century was largely a desert for the drama, it was the scene of a theatrical harvest, during which the theater as an institution grew and flourished in the hands of great actors and actor-managers, and all the arts and crafts of the theater were refined. Second, though the drama of the times was mostly mediocre as literature, much of it was first-rate as theater, causing Britain’s growing middle class to flock to it for amusement, thus sparking its physical and institutional development—around fifty theaters were built in London alone between 18oo and 1890. It’s true that those seeking greatness in the nineteenth-century theater found it mostly in the acting and staging, and in revivals of Shakespeare and other classics, not in the contemporary drama; but at least they found it. And those primarily seeking entertainment were seldom disappointed. The thousand or so playwrights who wrote between Shakespeare and Shaw, though now mostly forgotten, could at least be generally counted on to amuse the populace according to the tastes of the times, and occasionally even to elevate those tastes slightly. Another consideration is that the theatricalism, abstraction, and musical nature of much nineteenth-century drama has been partially vindicated by the dramatic practices of twentieth-century drama, though of course the difference is that the best twentieth-century drama made these properties or qualities serve higher purposes. These qualifications aside, Archer’s characterization of over two centuries of theater as desert or dark age had much validity, considering the standard set by Shakespeare, and interested Victorians agreed that a dramatic revival was in order. The alternative was to follow Matthew Arnold’s example in abandoning the theater out of disgust.  

 

         But which exactly needed to be revived—the drama or society? The word renaissance connotes the rebirth of a people and thus might be thought too strong a term if applied only to the drama. Most Victorians did not think of their age as especially benighted, at least nothing a little reform and technological and business progress couldn’t take care of.  It may have been a dark age of the drama, but in the novel and poetry and the other arts, and certainly in the sciences and in business and industry, most Victorians considered theirs a progressive, enlightened age. So it was hard to convince an otherwise forward-looking people—industrializers of a world empire and avid users of railroads, telegraphy, electricity, photography, and telephones—that they were backward in much else besides this very specialized and seemingly unimportant area known as the drama.  

 

         But George Bernard Shaw, this era’s chief playwright, argued and demonstrated that, technological progress notwithstanding, backwardness was so deeply entrenched in the moral, religious, and governmental systems of the day that it was not too much to call the entire age a dark age and to play its “progressiveness” as an ironic joke. For Shaw, as well as Archer and many others, the word renaissance was not too ambitious for the extreme measures that were needed to breathe new life into a morally rotten society, and the drama, with its ancient roots in the life-worshipping Greek religion of Dionysius, was precisely the means needed. As an institution for the gathering of people together to commune on the issues of social health and spiritual well-being, and to plumb the mysteries of human identity in a riddling universe, employing thereby all the arts and crafts in a unifying effort, the drama was ideally designed to be the focus of culture, as it had been at its beginnings in ancient Greece. In fact, the health of a nation could be determined by how central it made its drama and how seriously it took it. For Shaw and Archer, a major diagnostic of Victorian society was its trivialization of the theater. Only in a dark age would the spiritual light that may illuminate the stage be allowed so nearly to go out.

 

         A renaissance is a period of enlightenment. The original Renaissance was awakened to the long-lost past of the Greeks and Romans—its chief reality was that of ancient truth rediscovered, as, for example, the way Aristotelian principles were henceforth applied to drama. In contrast, the modern age, guided by science to be irreverent toward the past and skeptical of received truth (as the maverick Galileo had been skeptical of Aristotle), thought of itself as more concerned with present reality, especially awakening to physical reality, since it could be empirically verified. It supposed that the physical world was scientifically knowable and controllable, and that therefore the future could be commanded through the invention of new technologies and new methods. From about the middle of the nineteenth century there was a gathering insistence that art follow science in a more “realistic” investigation of the physical world, thereby joining the March of Progress. And so the novel followed painting, to mention two of the arts, in becoming more “realistic” and thus supposedly less escapist. By the 1890s the New Drama as well was identified with “realism,” with British experiments in “realism,” however timid, as early as the 1860s (T. W. Robertson). On the continent, Émile Zola had argued in 1873 that playwrights should be scientists too, “realistically” and tough-mindedly examining in the laboratory of the stage the physical operation of human society and human consciousness. And, from the seventies on, Henrik Ibsen, with his microscopic dissection of modern Norwegian society and individual personality, had shown how best to do it in a dramatic form. For Archer, Ibsen was the model for the future.  

 

         But the “realism” of an art based on illusion, as is drama, was immediately challenged, even by some of those playwrights labeled as “realists”—Ibsen himself fused “realism” with symbolism and flirted with expressionism. Many artists argued that the use of “nonrealistic” modes of expression did not necessarily mean that art was escapist; rather, art’s approach to reality could only be through illusion (i.e., spiritual reality)—a play, for example, was a “playing” with reality. And the substitution of electric lights for gas lights in the eighties and nineties, made possible by science’s regard for physical reality, did not necessarily add to the stage’s spiritual illumination but actually seemed at times to obscure its presentation of spiritual reality. A sign of the times, however, was that the aggressively positivistic science of the day made people feel apologetic about using a word like spiritual, though some playwrights were less intimidated than others. And thus began a very complicated debate on the nature of dramatic reality.  

 

 

REALISM AND THE REACTION  

 

 TOP / Table of Contents

 

         “Realism” as a dramatic style refers to the appearance of lifelikeness (verisimilitude) in setting, costume, dialogue, gesture, facial expression, and so on. A realistic play was to be a photographic copy of common, observable experience (in practice, usually middle-class domestic experience, to accord with the reality of the rise of the bourgeoisie). The stage was to appear, not as a stage, but as a room or any actual environment; props were to be seen, not as props, but as authentic parts of a particular everyday environment. All the developing technology of the modern theater—hydraulic machinery, cycloramas, lighting boards, etc.-—was brought to bear in creating the illusion of authentic environment. A proscenium arch separated the stage from the auditorium and framed the action taking place on the stage in a three-sided box set. Some theaters (such as London’s Haymarket under the Bancroftssee below) eliminated the apron stage in front of the proscenium altogether and bordered the proscenium so that the effect was that of looking at a framed picture. The proscenium’s “fourth wall,” through which the audience peered, as Peeping Toms might look into bedroom windows, was invisible by convention.  

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: http://chuma.cas.usf.edu/~dietrich/britishdrama1_image002.png

         As for the “realistic” play, no authorial intrusion was allowed, and neither audience nor actors were acknowledged for what they were. The idea was to achieve the illusion of re-created life, in its immediacy and dense actuality. At its best (Ibsen and Chekhov) it did indeed give the feeling that one was peering through an open window into someone’s house and, unobserved, overhearing private conversation. “Real­ism” at its best was very persuasive in making audiences believe that the illusion they were seeing was not an illusion.  But of course that simply made “realism” the most outrageous of all of the theater’s pretenses—one had to make-believe that one was not in a theater and not looking at actors acting on a stage. But it must have been a relief to those tired of plays that pointed the moral, and owing to its relative subtlety, riveting to those wishing to know what it all meant. The supposed neutrality or scientific objectivity of the author, the indirectness of the characterization, and the relative inconclusiveness of the action forced the viewer to pay close attention to the details in order to form judgments, with much of the meaning of such plays occurring in the subtext and accumulating gradually, almost imperceptibly, detail by detail.  

 

“Realism” at first was associated with “social drama,” for its immediate goal was to display accurately and authentically the social environment and behavior of the day.  But this association gave realism a reputation for being superficial, for getting lost in relatively unimportant surface detail at the expense of portraying the more important soul of things. That was why Ibsen resisted “realism” for so long, preferring to go on writing obsolete heroic drama, often in verse, rather than stoop to “mere photography.”  But then it dawned on this genius that the surface of life could be used in a poetic, symbolic way, just as great photographers were learning that the camera need not just copy life’s exterior but could interpret and poetically evoke the hidden depths as well.  Ibsen converted to “realism” when he found that he could use the surface to suggest the deeps and so invented what came to be called “psychological realism,” in which the picturing of society is employed to suggest the underlying soul or psyche. And insofar as his plays penetrated mundane appearances, reaching to the significance of things, they were examples of “philosophical realism” as well, and of “critical realism” insofar as they saw through the humbug of the day. It was a neat trick, this elevating of what seemed a trivial and mundane art into a high art, but so many missed the trick that Ibsen was often erroneously dismissed as a mere social realist, thus leading other dramatists to become overtly “nonrealistic” in their expression of the psychological deeps and intellectual heights in order to separate themselves from what was thought a second-rate art.

 

          The point to be underscored is that as human reality is multidimensional, the word realism should not have been limited to the imitation of our most superficial reality. This early mistake in terminology plagues us like an original sin, accounting for the quotation marks around “realism” and “nonrealism” to this point, to signify that the standard notions of these terms have created a false distinction, for “nonrealistic” plays are no less capable of showing us reality than are “realistic” plays, and in fact the reality conveyed by “nonrealistic” plays may be more significant. Theater departments often wisely use the alternate terms representational and presentational, but English departments, caught in the toils of literary history, seem to be stuck with the confusing “realism” and “nonrealism.” Having acknowledged the confusion, however, we may henceforth drop the annoying quotation marks if we keep constantly in mind that “realism” and “nonrealism” are misnomers.  

 

         Another, related confusion in terminology might just as well be mentioned here—that over the term naturalism, which is used in at least three different ways.   Naturalism may refer to nothing more than the natural-looking or natural-sounding quality of a play.  Chekhov’s plays are often cited as naturalistic in this sense, as his characters create the impression that they are as disorganized, spontaneous, and inarticulate as life outside of art frequently is.  Shaw’s plays are not naturalistic in this sense, for his characters are articulate well beyond what is considered natural.  In acting, Gerald du Maurier is particularly credited with developing the most naturalistic style, which consisted mainly of giving the appearance of not acting, a style Shaw had little use for. This sort of naturalism, as a kind of hyperrealism, is just as often referred to as realism pure and simple, the critics being hopelessly inconsistent.  

 

         Naturalism (sometimes capitalized in this sense) may also refer to a particular philosophy of life and/or to a particular literary-dramatic embodiment of that philosophy. Naturalism as a philosophy refers to the Social Darwinist idea that human beings are purely the product of heredity and environment, utterly determined in their behavior by these shaping factors of the natural world. This philosophy may be embodied in any kind of play, realistic or nonrealistic; and a play may have characters in it who express a naturalistic view without the play itself being totally, or at all, supportive of naturalism as a philosophy. Literary naturalism refers to works that attempt to embody a naturalistic philosophy in a very specific form, in which realistically portrayed characters are obviously and entirely at the mercy of environment and heredity. Typically, plays of this type (Gorky’s The Lower Depths, O’Neill’s The Iceman Cometh) focus on a lower-class or primitive environment, where the iron laws of heredity and environment are most nakedly exposed in an elemental struggle for survival.  

 

         The reactions against realism and naturalism were various, some nonrealistic dramatic forms given names from the past, such as fantasia, burlesque, allegory, and extravaganza, some having names invented for them, such as symbolism and expressionism, and some seeming to fit no particular category (most of Shaw’s plays).  Of the new forms, symbolism and expressionism most typified the modernist reaction against realism and naturalism, having in common that they were evocations or assertions of a reality beyond the ken of positivistic science.  Symbolism pointed to a spiritual reality behind appearances, and expressionism projected outward an internal reality positivism overlooked.  

 

         Uncapitalized, symbolism merely refers to the use of some things to represent other things, as a single chair on a stage might represent all furniture, a tree might represent life, or a setting sun might represent the coming of death.  Symbolism capitalized refers to the specific use of symbolism, conceived by a late nineteenth- and early twentieth-century literary movement (beginning with such poets as Mallarmé, Verlaine, and Valéry, and finding its purest dramatic expression in the plays of Maeterlinck and Yeats), to evoke a spiritual world beyond the five senses through an associational technique that connects things in the material world with their correspondences in the spiritual world.  Symbolism in this sense employs symbols in the least definite of ways to suggest unseen powers and emotional realities, to evoke the “esoteric affinities” of the writer.  Symbolism was private and subjective in that the author’s system of association, the particular way he evoked the archetypes, was his own; but Symbolism overcame the implied chaos of subjectivism because the correspondences activated universal archetypes, buried in the psyche of everyone, that, when properly evoked, were capable of connecting individuals in a collective awareness.  

 

         Expressionism was usually a more extreme assertion of private, subjective reality, of the sort posited by psychoanalysis, though it too might appeal to universal archetypes. Prototypical were Strindberg’s A Dream Play and The Ghost Sonata, peopled by bizarre, rather abstract characters involved in dreamlike action, the logic of which was emotional and associational rather than rational.  Expressionism flowered in the Germany and Austria of the twenties, with the works of Oscar Kokoschka and Georg Kaiser, and spread to America (O’Neill’s The Hairy Ape) and elsewhere.  Dictionary definitions often say that expressionism distorted reality, but such distortion was only from the point of view of realism.  From its own point of view, expressionism did not distort reality; rather, it projected outward what was really there in the inner mind or soul of the writer or character. Its mimesis (imitation) was of psychological or spiritual reality.   Its aim was to assert the primacy of human will and imagination over external circumstance, in contradiction to deterministic naturalistic theory.  

 

         Realism of the extreme purity Archer wanted is an aberration in the theater, for the long tradition of the theater, before and after that brief period of the realistic movement, has been more nonrealistic than realistic, though many of the greatest dramatists seemed to derive strength from an alloy of the two.  From the Greeks to the New Drama, the stage traditionally presented reality through the device of acknowledged illusion; anything else seemed deceitful. And although the movement in nineteenth-century drama was generally from a non­realistic, or presentational, mode to a realistic, or representational, mode, the movement in twentieth-century drama to the present has been from a realistic mode not so much back to a nonrealistic mode as to a latitudinarian attitude that anything is possible in the theater and that the playwright is free to use realistic or nonrealistic modes, separately or in combination, as appropriate to the play.  But this has only become clear in the last fifty years, the postmodern era.  In the period of our study, 1890 to 1950, the last half was largely characterized by a reaction against realism, with the fifties and sixties capping it off with the aggressively antirealistic Theater of the Absurd. And so we come full circle.  

 

         Myron Matlaw, in his Modern World Drama: An Encyclopedia, defines realism thus:

 

REALISM is as loose a term in the drama as it is in the other arts.  It refers to any attempt at reproducing verisimilitude on the stage. Since this could mean the representation of external or internal, physical or psychological or philosophical—or even political, sociological, or economic—-”realities,” and since these may be perceived in many different ways, the term is almost meaningless.  Instead of being a description of anything, it is popularly used as an evaluation, usually an approving value judgment on the “truthfulness” of a work.3  

 

“The term is almost meaningless,” says Matlaw. How dismayed Archer and some of the New Dramatists would be to hear that this is the outcome of their struggle to force the drama to be realistic. Yet Matlaw, in throwing up the lexicographer’s arms at the futility of defining so slippery a word, is simply being true to the spirit of his own postmodern age, an age in which not only are the wisest scientists considerably less positive, not to mention less positivistic, about reality than they used to be, but also we have considerably less confidence about language’s relation to any reality outside itself.  From this skeptical postmodern perspective, then, latitudinarianism seems the most becoming position to take. In any age it is difficult to empathize with the heated arguments of the past if they are no longer of concern, but our postmodern perspective makes it all the more difficult to cast ourselves back to the 1890s and feel how burning the issue of realism was to the New Dramatists (even as we grow cooler to the passionate arguments of Ionesco et al. against realism). It seemed then a life-and-death issue, at least for the drama.  

 

The tone of William Archer’s plea for realism in The Old Drama and the New is very telling. He obviously felt beleaguered by those (such as Yeats) who would dismiss modern realistic drama as inferior or degenerate. Archer’s object was to discover in the history of drama a “guiding principle of evolution,” something that would help determine “the essence of drama,” and that would provide “the basis for a rational standard of values” by which the drama could be judged.4  

 

Archer begins with the assertion that the two sources from which drama arose were imitation (mimesis) and passion.  By “passion” he signifies “the exaggerated, intensified—in brief, the lyrical or rhetorical— expression of feeling.”5   He cites song, dance, and heightened speech as examples, but eventually includes almost any stage business that he considers not exactly imitative of reality.  Imitation is the essence of drama, and all the lyrical-rhetorical exaggerative elements, which he associates with “the primitive,” are impurities that need to be purged from drama and delivered to the music hail, the opera, and the ballet, the proper homes for the hysterical arts. The form that best accomplishes this purgation is modern realism, imitation triumphant. “Who can doubt that the future belongs to it?”6  

 

         With such notions, Archer’s history of English drama could only be mostly a catalog of failure, for that drama—with its speeches in verse, its asides and soliloquies, its direct address to the audience, its moralizings, its raked stages, its acting outside the stage-picture on apron or thrust, its indulgence in wit and rhetorical flourish, its formulaic characterizations, its boys disguised as females, its grand style of acting—was seldom realistic in the way he wanted.       

 

         Of course Archer has to qualify every condemnation of the passionate lyrical-rhetorical drama with the exception of Shakespeare. “Consummate genius can express itself in any form and can ennoble any form.”7   One might draw the opposite conclusion that Shakespeare succeeded, not in spite of his “passionate” form, but because of it, but Archer’s idealism is proof against any such argument.  Surprisingly, toward the end of his book Archer offers Ibsen, heretofore the model for the rule of realism, as another exception to the rule. “What he really did was, not to confine his genius within the limits of realism, but to show that realism of externals—of environment, costume, manners and speech—placed no limits upon the power of genius to search the depths of the human heart, and to extract from common life the poetry that lurks in it.”8  In other words, what made Ibsen great was not his realism but his ability to get at the poetry (i.e., the passion) beneath it. (Thomas Postlewait, in his Prophet of the New Drama: William Archer and the Ibsen Campaign, has explained that this seeming contradiction was based in Archer’s split personality.  Archer was often very appreciative of individual plays of the passionate-rhetorical type, including some of Shaw’s, but for the sake of the theoretical consistency of his Ibsen campaign he consigned such plays to a species more primitive than the New Drama, thus repressing a part of his own appreciative nature and creating the false impression that he was a fanatic.)  

 

         Archer also refused to acknowledge that that which excepts Shakespeare and Ibsen excepts Shaw.  The concluding chapters of Archer’s book are devoted mostly to encomiums of what we would now consider minor playwrights—Arthur Wing Pinero, Sydney Grundy, Henry Arthur Jones, Harley Granville Barker, James Galsworthy, etc.-—at the expense of a just estimate of the period’s one giant, G. B. Shaw.  In an irony fitting of the New Drama itself, Shaw as the playwright who best embodied Archer’s prophecy of a New Drama got little credit from the prophet himself because he embodied it in a way that Archer could not theoretically approve.  But, then, Archer began as Shaw’s friend, neighbor, and rival drama critic, and it must have been a struggle just holding his own against the irrepressible Shaw.  One wonders if Archer’s constant reference to realism as “sober,” with its implication that passionate drama was “drunken,” might not have been a sly joke at the expense of the teetotaling but rhetorically passionate Shaw.  

 

         At any rate, Archer ends by congratulating imitation and lyrical-rhetorical passion on at last consummating their divorce. “This divorce, so obviously inevitable, is a good and not a bad thing—a sign of health and not of degeneracy.”9  This could only have been written in the early twenties, for the triumph of realism that Archer here celebrates was soon to turn into the gradual rout of realism. Nothing could dramatize the issue more clearly than the fact that in the year The Old Drama and the New was published, 1923, Shaw produced his Nobel Prize-winning play, Saint ]oan, as passionate, lyrical-rhetorical, and nonrealistic a play as he had yet written. It was a harbinger of the turn he was to take in his last phase toward open “extravaganza,” one of Archer’s most despised forms; and of the turn other dramatists, following continental trends (Strindberg, Maeterlinck, Pirandello, Lorca, Cocteau, Anouilh, Brecht) and American trends (O’Neill, Rice, Wilder, Williams), were to take toward nonrealistic drama in general.  The final twenty-five years of our period saw realistic plays, looking increasingly stodgy, balanced off with a variety of fresh-looking nonrealistic plays, the most interesting perhaps being the Balinese-No inspired heroic drama of W. B. Yeats, the expressionistic experiments of Sean O’Casey, and the revival of verse drama in the hands of Christopher Fry, T. S. Eliot, W. H. Auden, and Christopher Isherwood. And then in the fifties the top blew off, with Samuel Beckett and Harold Pinter providing the antirealistic dynamite. Yet, though by the end of our period pure realism seemed an exhausted form, imitation was soon to refresh itself by connecting with a new sort of social protest (John Osborne’s Look Back in Anger) and by combining in hybrid forms (Peter Nichols’s Joe Egg), the moral seeming to be, the opposite of the one Archer drew, that imitation and passion are elements within drama seeking, not exclusion of the other, but a proper, dynamic relationship. The model for the future was Brechtian “epic theater” and Yeatsian “total theater,” in which audience involvement in realistic and semirealistic episodes, and audience “alienation” through nonrealistic dance, mime, ritual action, etc., alternated or combined in an all-inclusive art.  

 

         Archer’s polemic necessarily overstated the triumph of realism in his day; it was more a temporary redressing of an imbalance created by the nineteenth-century’s inept overindulgence in passionate, nonrealistic theater. Archer’s dedication to realism was partly due to his sitting through too much frivolous song and dance as a youth.  But musicality was a peculiarity of nineteenth-century drama caused by the historical accident that licensing laws dating from the seventeenth and eighteenth centuries prohibited straight drama from being performed in all but three major theaters (Covent Garden, Drury Lane, and, in summers, the Haymarket), making it necessary for the “minor” theaters to adorn their dramas with music during, before, and after the play. These laws were repealed in 1843, making straight drama possible for all theaters, but the habit of musical accompaniment to drama persisted because it suited popular tastes, and so melodrama still announced the arrival of the villain with a thrilling piano, driving the young Archer to long for someone to shoot the piano player. Or at least drive him out of the “legitimate” theater into the music hall and the opera hall.  If Archer had only noticed that his beloved Ibsen, in A Doll House, had merely put the piano player onstage, to accompany Nora’s tarantella, he might not have labored so diligently to separate imitation from passion.

 

         John Russell Taylor sums up our present attitude. “In drama what seems natural is natural—there are no legitimate or illegitimate illusions, only illusion achieved or not achieved.10 Ironically, our present liberation from the need to promote one kind of drama over another has actually been to the benefit of the realists. As Shaw had pointed out in the thirties, Ibsen’s psychological dramas would be served just as well by acting outside the proscenium, in styles emphasizing their symbolism and expressionism; nowadays one is just as likely to see Ibsen’s plays done in theater-in-the-round or on a thrust stage, which gives them a whole new life. Our freedom from dispute on these matters also allows us to appreciate old-fashioned realism, realism acted behind the proscenium in a box set, without any sense that that was ever an embattled form. Nowadays we care only that the theater artists do their jobs—that is, create effective theater—and how they do it is their business. Variety is the spice.   

 

 

 

THE WELL-MADE PLAY AND THE PROBLEM PLAY

 

 

 TOP / Table of Contents

 

         Realism was undoubtedly the dominant trait of Archer’s New Drama, but it had two other features as well, deriving from the development of two separate kinds of nineteenth-century drama—the “well-made play” and the “problem play.” Archer often spoke of realistic drama as though it automatically included the salient features of these two kinds of play, which were its “technique” and its “subject matter.”  Archer began by saying that “the advance of dramatic art has consisted, not merely in the negative process of casting out extraneous and illogical elements, but also in the positive process of acquiring a technique appropriate to the great end in view—that, namely, of interesting theatrical audiences by the sober and accurate imitation of life.”  He spoke of how French playwrights of the nineteenth century had “discovered the central secret of modern technique—the infinite ductility or malleability of dramatic material. In other words, systematic ingenuity had, almost for the first time, been applied to the ordering of plot. The art of keeping action always moving, and freeing it from the frippery of irrelevant wit and the adipose tissue of wordy rhetoric, had been invented and developed.”12

 

         Though Archer had gotten used to thinking of a certain plot construction as endemic to realism, realism and the plot construction of the well-made play actually came from different directions and were not mutually necessary. Given credit for fathering the well-made play (or pièce bien faite) was Eugène Scribe (1791-1861), fertile father of about five-hundred plays. Mass production of plays on that scale requires machinery, and it was Scribe who, so to speak, industrialized playwriting by creating a plot machine for keeping an audience constantly intrigued. As John Russell Taylor puts it, Scribe “saw that all drama, in performance, is an experience in time, and that therefore the first essential is to keep one’s audience attentive from one minute to the next. . . . His plays inculcated, not the overall construction of a drama . . . , but at least the spacing and preparation of effects so that an audience should be kept expectant from beginning to end.”13   Telling a story well, “so that there is not one moment in the whole evening when the audience is not in a state of eager expectation, waiting for something to happen, for some secret to be uncovered, some identity revealed, some inevitable confrontation actually to occur,” is Scribe’s simple secret (so ironically exploited by Beckett’s minimalist, plotless Waiting for Godot).14   It was left to Victorien Sardou (1831-1908) to convert Scribe’s machine for generating intrigue into a formulaic plot to give the well-made play its reputation for tight construction.  Sardou’s plot came in parts—first, a quick exposition, introducing characters and filling in their pasts, followed by an inciting event (a misunderstanding, a secret withheld, an intercepted message, a visit from a mysterious stranger, etc.) that causes the action to rise in tension, act by act in incremental steps, toward a scène à faire, a climactic confrontation that forces the action to a crisis and a denouement, a resolution of maximum sensation. And there was to be no wasted motion, no “fripperies” of poetry or rhetoric such as a Shakespeare or a Shaw would indulge in.   Eugène Labiche (1815-88) and Georges Feydeau (1862-1921) adapted the “well-made” formula to comedy, putting the emphasis on bigger and bigger laughs instead of on bigger and bigger thrills.  In the well-made play, pattern is all.  Character and theme are subordinated to plot intrigue.   Though “well-madeness” became identified with realism, it was originally intended to apply to any form—farce, melodrama, heroic tragedy, whatever. Chekhov’s plays, especially, illustrate that “well-madeness” is not only not necessary to realism but may actually be counter to the assumptions of realism, since life is seldom “well-made.”  

 

         Conjoining with the well-made play in the New Drama was the "problem play."  Shakespeare’s dark comedies, such as Measure for Measure, have been called problem plays, so the type has been around; but according to Archer, it was Sydney Grundy who first used the term, disparagingly, after he gave up writing such plays (Shaw would say it was because he wrote them poorly!).  

 

          However, it may have been the Danish critic, George Brandes, early celebrator of Ibsen’s genius, who in 1872 broached the idea of the problem play.  Wrote Brandes, “What is alive in modern literature shows in its capacity to submit problems to debate.”15  Supposedly, Ibsen later embodied this best in the discussion between Nora and Torvald at the end of A Doll House concerning modern middle-class marriage.  Such debate connected with realism in that it provided a way for drama to come to grips with “real life” by dealing straightforwardly with social problems. But Ibsen was misunderstood. To paraphrase a critic, to say that A Doll House is about women’s liberation or that Ghosts is about venereal disease or that An Enemy of the People is about political corruption is like saying that King Lear is about housing for the elderly.”16  Those who misunderstood Ibsen wrote plays about slum landlordism, prostitution, labor unrest, penal codes, class warfare, the double standard, divorce, business corruption, etc., in such a very limited and narrow way that they gave the problem play a bad name.  The reason even a Sydney Grundy might speak disparagingly about problem plays is that such mundane matters—when dealt with at only a literal, local, topical level rather than at a level that questions human identity and destiny, as in Ibsen and Shaw—could easily degenerate into political tracts or propaganda pieces.  Problem plays became identified with thesis plays, didactically presenting social problems for the sake of promoting a particular reform or upholding convention.  A favorite “problem” of New Dramatists such as Jones and Pinero was the issue of whether a “fallen woman” could be allowed back into respectable society, and the answer was always “no.” It’s ironic that the sort of play Archer championed for its subtle, objective presentation of reality was so often secretly moralizing, imposing ideals or ideology on reality rather than “telling it like it is.”

 

 

 SHAVIAN NEW DRAMA

    

 

TOP  /  Table of Contents

 

 

         The ending that allowed no happy solution to the fallen woman question was problematic when compared with the sudden and miraculous reconciliations, conversions, and other providential workings that marked the hasty denouements of melodrama, yet Shaw was quick to point out that problem plays of the Pinero and Jones variety were phony, for the conclusions were foregone, their unhappy endings not really following from character and event but merely mechanically imposed by moral conventions, ideology, external to the play.  In The Quintessence of lbsenism (1891), Shaw argued that Ibsen, on the other hand, in implying that the quintessence of morality was that there is no quintessence, that there are no easy or final solutions to the problems life poses, had meant that the conventional ending was part of the problematics—that is, if a play’s ending, simply as a matter of convention, automatically says no to the question of whether a fallen woman can get back into society, that ending should not resolve the issue, as Pinero and Jones would let it seem to do; rather, it should expose an irreconcilable conflict between the individual will and the conditions that seek to govern it.  Understanding Ibsen in this way, Shaw made “problem” the center of his own program for the New Drama. “Problem” simply needed to be understood correctly. “Drama,” said Shaw, “is the presentation in parable of the conflict between man’s will and his environment—in a word, of problem.”17   Shaw thought the dramatist should deal with social issues, but only as the context for a dramatization of the larger, universal conflict between private will and circumstance, showing the individual struggle to realize an identity and a purpose in a mysterious universe.  Such problems are timeless, however localized by their time and setting.

 

         With the well-made play, however, Shaw was not interested in saving a misunderstood form—he attacked it wholesale. Well-made plays were merely “mechanical rabbits” leading the audience like dogs on a merry chase, but to no lasting or significant purpose.  Shaw believed that plays should grow organically, from character and situation, rather than have a ready-made plot imposed on them.  It was greater realism, he argued, to let life go where it would rather than force it into an artificial, prejudging mold.  Further, in The Quintessence of Ibsenism he declared that the discussion in dramatic form of a “problem” was the technical novelty in Ibsen’s plays that should replace the old Scribean art of intrigue.  For intrigue “Ibsen substituted a terrible art of sharpshooting at the audience” through a discussion technique that implicates the audience.18  And if discussion is allowed to follow naturally from character, then the play, like life, will not be “well-made.”  The well-made play of “Sardoodledum” (referring to Sardou) was a falsifying apparatus that Shaw saw as contradicting all the assumptions of that realism to which it was allied.

 

         Yet Shaw was not interested in defending dramatic realism either. “Stage realism is a contradiction in terms,” he said.19   It was not Ibsen’s literary realism that made him great but his psychological, philosophical, and critical realism.  The Ibsen Shaw presents in The Quintessence of Ibsenism is a visionary, much akin to Ibsen’s own view of himself.  His surface realism was subterfuge, a cover-up and symbolic signpost for the poetic divination that was going on behind the scenes.  

 

         Believing that great artists express themselves authentically, not by fitting into a standard formula for art, but by having an individual style, Shaw saw the appropriateness of Ibsen’s style, expressive of his secretive, subversive character.  In devising his own (acquired) style, that of an open, flamboyant extrovert, devoted to frontal attack, Shaw felt he must acknowledge stage illusion for what it was.  “Neither have I ever been what you call a . . . realist. I was always in the classic tradition, recognizing that stage characters must be endowed by the author with a conscious self-knowledge and power of expression, and . . . a freedom from inhibitions, which in real life would make them monsters of genius. It is the power to do this that differentiates me (or Shakespeare) from a gramophone and a camera.”20  Furthermore, the object of drama for him was “the expression of feeling by the arts of the actor, the poet, the musician. Anything that makes this expression more vivid, whether it be versification, or an orchestra, or a deliberately artificial delivery of the lines, is so much to the good for me, even though it may destroy all the verisimilitude of the scene.21  

 

         Shaw’s strategy was to make the realistic well-made problem play ridiculous by showing its self-contradictions and its failures to live up to its own model.  The New Drama of this sort having been exposed as fraudulent, the stage would thus have room for his own New Drama, sometimes called by him the “Drama of Ideas” but a more complex thing than that label suggests, as we’ll see later. His strategy succeeded admirably to the extent that it certainly made room for his own plays and the plays of other dramatists uncomfortable with realism, but the record is otherwise mixed.  A look at the drama of the twentieth century shows that Archer’s ideals seem to be more in practice than Shaw’s until about 1930 and that from then on there is a swing in Shaw’s direction, although in any given London season one could find both sorts of plays.

 

         So which was the real New Drama—the realistic well-made problem play of Archer’s theory or Shaw’s fabulous, passionate Drama of Ideas?  Archer’s type no doubt was in the majority (though, curiously, the few plays Archer wrote himself do not fit his own theory).  But as for quality and lasting value, Shaw’s type clearly prevailed, if only because Shaw himself was the best playwright.  It also prevailed in the sense that there no longer seems any question of pure realism being a superior art form or of there being any constraint on dramatists wishing to write passionate drama.  It may also have prevailed because, in a contentious age, full of great battles between Victorian and modern ideas and between rival theories of modernism, a rhetorically passionate drama of ideas was a fitter vehicle for expressing the age.

 

 

THE STAGE AND THE AGE

 

  TOP /  Table of Contents

 

 

          Henrik Ibsen as a young man, looking at what seemed a hopelessly corrupt and wrong-headed society, declared that a total revolution was needed, one more thorough than the biblical flood, which left survivors. In the next revolution, he said, we must “torpedo the ark.”22   A quick glance at our period suggests that the 1890-1950 era came surprisingly close to effecting complete revolution, even to the sinking of more than one ark.  It’s amazing how much historical incident and social and technological change was crowded into this period.  It was a time of both gradual and cataclysmic transformations, unprecedented in amount.  So much so that even Shaw, one of the most ardent advocates of change, complained toward the end of his long life of the dizzying rapidity of change and in many of his late plays presented characters who were victims of what we would now call “future shock,” the disease that comes from having the future come at one too fast.23  

 

Shaw’s exaggerative persona, the clownish G. B. S., like Ibsen’s “torpedo” metaphor, was a sign of how desperate a measure was thought needed to overcome Victorian inertia in social-moral-religious concerns; but the “dynamite” personality Shaw devised to explode Victorian conventions resulted in more than he bargained for (though of course he was not the only “dynamitard”). For a period conceived of as glacial in its movement at its beginning found itself more and more resembling an avalanche. There is some moral here for those who would “start the ball rolling,” but of course the moral wouldn’t be necessary if societies didn’t try to keep balls from rolling altogether.  

 

         A simple contrast between beginning and end is eloquent enough. In 1890 the horse and buggy still ruled the road; by 1950 automobiles had chased the few horses left to the country, and airplanes made both look like they were standing still. In 1890 Britain had the world’s greatest empire; by 1950 the empire had almost vanished—including Ireland, John Bull’s other island—although a very diminished Commonwealth of Nations had risen to replace it. In 1890 Britain looked back on the relatively peaceful reign of Queen Victoria, saving a few colonial skirmishes and the Crimean War; by 1950 it had experienced the Boer War, the Irish “troubles,” two debilitating world wars, much internal violence, and a constant imperial bleeding.  With the decline of the empire went the further decline of the aristocracy.  In 1890 being a lord or a lady still meant being deferred to, but after the wars, which leveled people more than buildings, there was more pride in being a worker.  In 1890 Britain was a bastion of laissez-faire capitalism; by 1950 several Labour governments had well under way the construction of the welfare state, the West’s answer to Russian communism.  Many of the “wastrel” aristocrats, faced with expenses they couldn’t afford and untrained in the ungentlemanly pursuit of trade, sold country estates to rich Americans or went into the tourist business.  In general, the rule of Britannia “went west,” so to speak, as America reluctantly assumed leadership of both the English-speaking world and an invention of “the cold war” called “the free world.”  At the center of all the changes were the women—hampered by head-to-foot clothing in 1890 but stripped down to the flapper by the twenties and bathing in bikinis by the fifties. And voting, earning paychecks, going to college, smoking, wearing trousers, swearing in public, getting elected to office, writing novels and plays, and doing other “manly” things.  It was enough to give a die-hard Tory a deep pain.  In 1890 many Britons escaped from their troubles by going to the theater and the music hall; by 1950 they were more likely to go to the cinema, and television was just getting started.  

 

         The philosophical and scientific underpinning for this social-political-technological change was the theory of evolution. Charles Darwin in 1859 had given voice to the inklings of a half century or more of scientific speculation on the origin of species, and by the nineties his theory of evolution had won sufficient acceptance. An alternative theory, that of Lamarck, was preferred by some, but Lamarckian and Darwinian were united in their understanding of life as evolving. With change being the law of life, and a better or higher life apparently being the result of change, or at least desired and aimed at,  some reasoned that the more change the better.  Some thereby deified Change and pursued it, in the form of novelty, for its own sake.  Soon people found themselves coping with “the tradition of the new,” in which fad was required to replace fad at an ever-accelerating rate.24  

 

         Art forms tended not to last long, for the modernist avant-garde was always moving on. The theater, too, though as usual slower to change, eventually got caught up in the same frenzy, as experimental theaters of the off-off-Broadway or “fringe” sort, following the lead of Archer, Elizabeth Robins, and J. T. Grein in the nineties, began to attract a select patronage and steal prestige from the established theaters of London’s West End.  

 

         Perhaps the most remarkable of the experimenters in staging was Edward Gordon Craig (1872-1966), son of the actress Ellen Terry.  Reacting against the heavy sets of a too literal-minded and literary-minded theater, he attempted to create a sparer, more fluid, poetically suggestive, and psychologically attuned style, which served well the aggressively antirealistic and antiliterary theater of a later generation. Craig’s revolution can also be understood as part of a general trend to replace the old actor-manager—who along with being the star was responsible for all the details of production—with a nonacting manager or director and other specialists, such as the artistic designer.  

 

         The most radical experimentation in the drama was found mostly in the “little theaters” in the provinces (Liverpool, Manchester, Bristol, Cambridge, Birmingham, Glasgow, etc.); in suburban London (the Lyric Theatre in Hammersmith, the Everyman Theatre in Hampstead, the Royal Court Theatre in Sloane Square, the Old Vic below the South Bank, etc.); in theater groups that moved about and hired halls (the Independent Theatre of J. T. Grein, the New Century Theatre of Archer and Elizabeth Robins, the Stage Society, etc.); or in Dublin’s Abbey Theatre. Their continental progenitors were Ole Bull’s National Theatre in Norway in the 1850s, the Duke of Meiningen’s company in the Germany of the 1870s and 1880s, Antoine’s Théâtre Libre in Paris from 1887, Otto Brahm’s Freie Bühne theater in Berlin from 1889, and Stanislavsky’s Moscow Art Theater from 1898.

 

          Another consequence of evolutionary theory that seriously affected the theater was the rationalization of “social Darwinism,” used as a justification for a ruthless free-trade economy that, following the law of the jungle, saw the strong get stronger and the weak get eliminated. The doctrine of “survival of the fittest,” which in business often meant the survival of the unscrupulous, seemed to justify unfettering the competitive instincts from ethical constraints. When certain men prided themselves on “being realistic,” they were not referring to literary realism but to an acceptance of the desire for aggrandizement as an honest basis for human society. This had a very sorry effect on the theater, for, while the theater has usually been at the mercy of the box office, the actor-managers, who still controlled most of the West End theaters in 1890 and who, even at their worst, had some aesthetic sense and some fellow feeling for the actors and artisans under their command, were gradually replaced by businessmen, often in multiple-ownership syndicates, who, having no feeling for the theater as a cultural institution with a special heritage, treated the theaters they owned as they would any other piece of real estate—no sentiment was allowed to mix with the cash flow. And, treating the actors as the capitalist everywhere treated labor, they forced the Actors’ Association in 1919 to reorganize as a trade union and theatrical agents to appear as middlemen. Although the modern period saw as many theaters built as had the nineteenth century, many of the new theaters were merely replacements for older theaters that had been razed or converted to make way for more profitable ventures.  Of course, capitalism’s Great War, inviting bombardiers to compete for the most destructive “strikes,” made rubble of a few theaters as well.  

 

         Newly built West End theaters at first followed the trend toward realism by building the sort of smaller, more intimate, fan-shaped picture-frame theaters that T. W. Robertson and the Bancrofts had shown in the 18ó0s to be the most suitable for the subtler, understated style of acting required by realism. The older style of huge, deep, horseshoe-shaped theaters with large aprons set by the patent houses of Covent Garden and Drury Lane had made broad, declamatory, histrionic acting a necessity.  It’s no accident that Covent Garden switched to opera in 1847 and that Drury Lane, for two centuries the bastion of the classical repertory, became notorious in the nineties for its lurid melodrama. When melodrama faded it became the home of the big Broadway musical.  

 

         The new theaters may have been designed for a less showy, less histrionic sort of play, but at the beginning of our period, the increasing wealth and social prestige of their patrons made luxurious display, greater comfort, and ceremonial opportunity features of theater construction. The backless benches of the nineteenth-century pit were pushed farther and farther to the back and replaced by plush, expensive seats called “stalls,” with the pit seats eventually disappearing altogether. Proscenium arches were supported by groups of statuary or gilded pillars; boxes were draped with colorful plush, their fronts embellished by vases, medallions, frescoes, caryatids, and the like; and gorgeous crystal chandeliers hung from decorated ceilings, adorned perhaps with gold leaf.  Foyers, saloons, smoking rooms, and buffets were found at the front of the house, lavishly decorated in various historical styles.  No expense was spared backstage either in the use of stage machinery or any modern technology that would make the show more impressive.  As the period wore on, however, the increasing democratization of the populace and the hardships of wartime and economic depression toned things down considerably in the theater. When “the talkies” came in, in the late twenties, some theaters were built in cinema style, with straight lines replacing the curved auditorium, stage boxes eliminated, and decoration reduced.   Eventually virtue was found even in the relative plainness of the “little theaters” of the suburbs and provinces.  

 

         At first the provincial theater lost ground to the London theater when, with the increasing availability and affordability of the railroads and other transportation, Britain’s far-flung population was able and eager to get to London for a holiday.  London-based national newspapers carrying the drama reviews of Clement Scott, William Archer, A. B. Walkley, Max Beerbohm, et al. were persuasive of the attractions of London theater.  Further, the replacement of the repertory system by the long-run system, largely accomplished by 1890, made it possible for everyone to see “hits.” And the touring system, which subverted the need for a resident company, provided the provinces with just enough of a taste of the star London actors to make them want more. London’s dominance grew through the 1890s and early 1900s, until, following the lead of Dublin’s Abbey Theatre and Granville Barker’s Royal Court Theatre, a few “little theaters” in the provinces began to rebuild resident companies and encourage local dramatic talent. This decentralization of the theater was supported by the rise of country theater festivals. Usually brief summertime affairs, such as the Malvern Festival (started by Barry Jackson in 1929), in some cases they developed more substantial repertories, as at Stratford’s Shakespeare Festival (started in 1886 by Frank Benson).  

 

         While one could still find theaters in the 1890s that offered the multiple fare that had been standard throughout the century—consisting of opening play, main piece, and closing play, with variety acts before, after, and in between, lasting sometimes from 6:00 to after midnight—the majority of theaters, catering to the advance of the dinner hour to 7:00 in polite society, had gone to a single play offering, opening at 8:00 or thereabouts, and soon this was universal. A wider adoption of the matinee further served the schedule of a more leisured gentility and of a female populace more inclined to venture out.  

 

         Following the lead of T. W. Robertson and the Bancrofts, the theater of our period begins by being obsessed with respectability. Theater people wanted the theater to be thought a proper place for ladies and gentlemen, on both sides of the footlights. Fighting to free the theater of the ruffian clement that had lowered the tastes of its audience throughout the century and of the bohemian element that had caused its actors to be suspected of vagabondage, many theater people craved above all else social acceptance, some as much for the elevation of their art as for themselves. Success came when the leading actor of the day, Henry Irving, was in 1895 the first actor to be knighted, soon followed by Squire Bancroft in 1897. Then began the great push for a similar acceptance for dramatists, led by Henry Arthur Jones’s campaign of probity, but not consummated until W. S. Gilbert was knighted in 1907 and Arthur Wing Pinero in 1909.  But far more knighthoods went to actors than to dramatists over the years.   Shaw eventually turned one down. 

 

         English-style realism was timely to the actor’s quest for respectability. Though continental and American realism often descended into the lower depths of the factory workers and peasantry with unpleasant depictions, English realism, at first anyway, tended to focus on the upper middle class and aristocracy. To look at the stage of a realistic English “society drama” was to see the latest smart fashions in interior decorating, clothing, hair styling, manners, and small talk. And as the acting of such a piece called for the sort of behavior one would encounter in the day’s drawing rooms, the actors, cup-and-saucer in hand, could, with properly clipped accents, downplay and understate in the best tradition of the reserved British gentleman. Such understatement can be made theatrical, but often it wasn’t. To those brought up on the grand style of acting so prevalent during the century, as Archer had been, however, the realistic cup-and-saucer school of acting seemed refreshing.  

 

         But all things pale, and among the first to feel the paling of this relatively untheatrical theater was Shaw, who as drama critic for the Saturday Review from 1895 to 1898 soon grew weary of staring into drawing rooms watching insignificant people do and say insignificant things in an insignificant way, compounding their insignificance by pretending they weren’t actors on a stage engaged in an important ritual.  For Shaw, the fashion show the actors provided was no compensation (though the pretty actresses almost sufficed).  For him such plays were basically “a tailor’s advertisement making sentimental remarks to a milliner’s advertisement in the middle of an upholsterer’s and decorator’s advertisement.”25  The fashionable, polite, well-bred, well-made, realistic drawing-room drama of the day, though winning respectability for actors and dramatists, Shaw perceived as simply watered-down melodrama, and he preferred his melodrama straight.  

 

         And so, with Shaw’s opposition, which was buttressed by the staging experiments of Gordon Craig, by the heroic acting style required by William Butler Yeats’s mythic drama and Gilbert Murray’s Greek revivals, by the Irish “soul music” of Synge and O’Casey, and of course by continental influences, no sooner was the realistic school of acting established than reaction against it set in. The history of the struggle follows that of the drama, first a seeming triumph for realistic acting, then a gradual retreat from it, moving toward a gradual acceptance of the view that a well-trained actor had better be schooled in both methods if he is to be fit to play the entire repertoire.  Not the least of the successes of this period was the establishment of several schools of acting where initiates could learn their trade, with Herbert Beerbohm Tree’s Royal Academy of Dramatic Art (1904) the most important (which Shaw acknowledged by helping to fund it). Yet what they learned at such schools, such as the team­work of ensemble acting, was not always germane to the way things were.

 

         The theater seems always plagued and blessed by the need for star actors.  Blessed because the star system gives the most opportunity to the best actors to display their art (the age would have been much the poorer if such names as Henry Irving, Ellen Terry, Florence Farr, Herbert Beerbohm Tree, George Alexander, Charles Wyndham, Mrs. Pat Campbell, Johnston Forbes-Robertson, Gerald Du Maurier, Gertrude Lawrence, Sybil Thorndike, Ralph Richardson, Cedric Hardwicke, Beatrice Lillie, Paul Scofield, John Gielgud, Laurence Olivier, and Edith Evans, had not graced its playbills, billboards, and marquees), plagued because the system exaggerates the importance of the star actor at the expense of the play and of the acting ensemble. Our period is no different in this tension. The dramatist’s drive throughout this period was to restore literary worth to playwriting, and his enemy was always the star actor who would sacrifice the written play to the display of a personal style. That had precisely been the bane of nineteenth-century theater, from the writer’s point of view.  In modern times only Noel Coward, star of his own plays, had it both ways.  Perhaps the happiest of the nonacting playwrights were ones, like Shaw, who wrote in plenty of personality and theatricality and who had made peace with the popular theater by employing all its “biz” and “shtick” in the interests of a higher drama.  Yet even Shaw was shut out of the commercial theaters for many years.

 

         The struggle between the requirements of the higher drama for disciplined ensemble acting and the commercial theater’s need for stars was only one aspect of the war between quality and quantity that marked the age. The only resolution to the dilemma of the higher drama’s inability to attract audiences in sufficient quantities was an endowed national theater. The need for such a theater had been noted for a long time, but as far as the New Drama was concerned, an 1879 visit to London by France’s subsidized Comédie Française can perhaps be pointed to as the inciting event.  It inspired Matthew Arnold to write, “The theatre is irresistible; organize the theatre!”28   And it led the young William Archer—from his early English Dramatists of To-Day (1882) to his practical promotion of Ibsen and other New Drama in both criticism and stage work—to launch a very serious and determined effort to fulfill Arnold’s command. He drew in many others along the way, so identifying the New Drama with the unifying idea of a national theater that the two became almost inseparable. As the corrupting influence of the commercial theater on the New Drama was noted, this call for a national theater grew. And as the higher drama accumulated, adding fantastic wealth to the national treasure of Shakespeare, the case for the establishment of a national theater became overwhelming, until not even the disasters of the planet’s bloodiest century could keep it from happening.  (See photo of London’s National Theatre below).

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: http://chuma.cas.usf.edu/~dietrich/britishdrama1_image003.png

 

 

 

 

 

 

 

 

TOP   /   Table of Contents

 

FIGURES
 

 

 

 

Figure 1 (below) – The Haymarket Theatre, 1880; Bancroft’s picture-frame stage.

Courtesy of the Raymond Mander and Joe Mitchenson Theatre Collection Ltd.

 

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: http://chuma.cas.usf.edu/~dietrich/britishdrama1_image002.png

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 2 (below)---The National Theatre on London's South Bank, opened in 1967.

 

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: http://chuma.cas.usf.edu/~dietrich/britishdrama1_image003.png

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 3 (below)---Oscar Wilde, a less than ideal New Dramatist.  

Photo: Eliis and Walery, LondonCourtesy of H. Montgomery Hyde.

 

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: http://chuma.cas.usf.edu/~dietrich/britishdrama1_image004.gif

 

Figure 4 (below)---Aubrey Beardsley's drawing

of a climactic scene from Wilde's Salomé

 

 

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: http://chuma.cas.usf.edu/~dietrich/britishdrama1_image005.gif

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 5 (below)---The diabolical Shaw.

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: http://chuma.cas.usf.edu/~dietrich/britishdrama1_image006.png

 

Figure 6 (below)---Punch cartoon portraying Shaw as Pan

(like Dionysius, a goat-footed deity representing the primacy of Nature),

which could stand for Shaw's attempt to lure the drama back to

its Greek origins as life-worship.  Courtesy of Punch

 

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: http://chuma.cas.usf.edu/~dietrich/britishdrama1_image008.gif

 

 

 

 

 

 

 

 

 

 

Figure 7(below)--Design for a statue of "John Bull's Other Playwright:

After Certain Hints by 'G.B.S.'"         Punch cartoon by E. T. Reed

 

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: http://chuma.cas.usf.edu/~dietrich/britishdrama1_image009.gif

Description: Description: Description: Description: Description: Description: Description: Description: Description: Description: http://chuma.cas.usf.edu/~dietrich/britishdrama1_image002.png

 

 

TOP

Link to Chapter 2: “‘Our Theatres in the Nineties’: Haunted by Ghosts”

Link to Title Page & Table of Contents

END