In the theatre, music performed as part of the performance of a spoken drama. See also Film music, Radio and Television.
4. Full sets and changing circumstances.
ROGER SAVAGE
Music has been closely linked with theatre since theatre began. Dance music and song have played important roles in much folk drama. The classic forms of Asian theatre from India to Japan rely heavily on music, as do the dramatic rituals of sub-Saharan Africa and of the indigenous peoples of the Americas. The major Western theatrical forms include several in which music is all but continuous (medieval liturgical drama, ballet de cour, ballet d’action and classical ballet, 18th-century pantomime, some types of opera, much ‘modern dance’), as well as several in which extended musical sections alternate on at least equal footing with passages of spoken dialogue: zarzuela, masque in most of its varieties, comédie-ballet, semi-opera, 18th-century vaudeville, ballad opera, Singspiel, opéra comique, operetta, musical comedy, the ‘musical’ and some music theatre. That only leaves the kinds of Western drama which put especially strong emphasis on spoken dialogue (‘plays’); and these too have very often availed themselves of music between their dialogue scenes and/or at points during them. There is no one term that groups together all the types of predominantly spoken drama using music in this latter way; but the music they call on is known in several European languages as ‘stage music’ (Fr. musique de scène; Ger. Bühnenmusik; It. musica di scena) and in English as ‘incidental music’.
The English term seems not to have been current before the middle of the 19th century and may have been a borrowing at that time from the collateral German term Inzidenzmusik, a category of ‘theatre music’. In English at that time it designated only those musical numbers within dialogue-scenes – occasional songs, dances, marches etc. – which were specifically called for by the script of the play in hand. These were thought of as ‘incidental’ in the lexicon sense of ‘following on from, or incurred in the execution of, some plan or purpose’ – the purpose in this case being getting a play on to the stage. (In such a context the adjective did not then, and arguably should not now, carry with it the trivializing connotations of other senses of ‘incidental’ such as ‘fortuitous’, ‘casual’, ‘not strictly relevant’.) Usage by the end of the 19th century, as the second edition of Grove’s Dictionary had rather grudgingly to concede, stretched the ‘incidental’ category to cover orchestral mood music, and later musical lexicographers (Apel, HDM, 1944; Illing, A Dictionary of Music, 1950) raised no objection to this inclusion. However, they did feel that a line should be drawn at calling theatrical overtures and entr’actes ‘incidental’ too. Yet within a few years their successors (Grove5, 1954; Cooper, The Concise Encyclopedia of Music and Musicians, 1958) were allowing that such preluding and interluding could reasonably be included in the ‘incidental’ bracket; and this capacious definition – all the music performed as part of the performance of a play – has been fairly standard since.
The widening of the definition between the 1860s and the 1950s can be linked with the increasing expectation in British theatre over the intervening decades that all the music used in connection with a particular staging would be written by one composer, and hence perceived as a kind of unity meriting a single term to embrace it; and it is certainly useful to have such a term to cover the totality of music used in any play performance even if its origins are more diverse than that (as they often have been). However, the term is not watertight. For one thing, all the constituent elements of incidental music so defined – overture, entr’acte, dance, song, chorus, mélodrame etc. – can also be found in other musical contexts; and for another, spoken plays that have a very considerable quantity of incidental music can approach, straddle and sometimes cross a boundary beyond which some other term might more usefully be applied to them. Thus, for example, Adam de la Halle’s Jeu de Robin et de Marion (c1283), which might quite reasonably be thought of as a pastoral play with a host of incidental songs, is perhaps better seen as a Singspiel or ‘musical’ avant la lettre; a self-styled ‘tragedy’ which adds several masque-like scenes of spectacle and music to its spoken text, such as Charles Davenant’s Circe (1677), is clearly on the threshold of semi-opera (and indeed in the case of Circe was remembered as an ‘opera’ by some of Davenant’s contemporaries); and a remark of Federico García Lorca’s about his Amor de don Perlimplín con Belisa en su jardín (1933) – that in the performances he directed himself its action and dialogue were ‘embedded in music like a chamber opera’ – might suggest we would be better off thinking of the piece as music theatre than as a spoken comedy with an incidental score.
Incidental music in the Western tradition effectively begins with the choric odes that were an important feature of Athenian drama in the 5th century bce. In Greek tragedy, and quite often in comedy too, these odes (sung and danced by a male chorus supported by an aulos) were sited between the several extended episodes of linear, argumentative spoken dialogue in a play, and supplied a perspective on its events and ideas different from that of the principal characters, both ideologically and aesthetically. Apart from a few phrases in a fragmentary state, none of the music for these odes, which was composed by the dramatists themselves in the main, has survived; but the phenomenon does have three significant links with later incidental music.
First, ode texts from the Greek have been reset by composers since the Renaissance for such restagings of the original plays as have chosen to use song for the choric episodes rather than the common alternative of heightened speech: for instance, the epoch-making revival of Sophocles’ Oedipus rex in Giustiniani’s Italian version which inaugurated the Teatro Olimpico at Vicenza in 1585, its choruses set by Andrea Gabrieli; the performances in German of Sophocles’ other ‘Theban’ tragedies at the Prussian court theatre in the 1840s, with music for the choruses by Mendelssohn; and the triennial productions of tragedies and comedies in the original language given at Cambridge from 1882 onwards, which have had scores (inclusive of chorus settings) by several British composers, notably Vaughan Williams’s for the Wasps of Aristophanes in 1909.
Secondly, in emulation of these Greek models, classicizing European dramatists have sometimes incorporated choric odes intended for singing into their own play texts: notably Racine in his two biblical plays for the college at St Cyr, Esther and Athalie (1689, 1691), with scores initially by J.-B. Moreau (the Athalie choruses being set again several times later, by Gossec and Mendelssohn among others). Some more recent experiments with the articulation of dramatic form have incorporated a singing chorus, though sometimes adding elements to it from beyond the Greek tradition: witness Brecht’s Caucasian Chalk Circle (1945), with its linking songs for a balladeer-narrator (set first by Paul Dessau), and several of Yeats’s partly Japanese-inspired dramas from the 1910s and 20s, with their three-musician choric lyrics. (Among his composers, Yeats especially admired Edmund Dulac for his music to At the Hawk’s Well and George Antheil for Fighting the Waves.)
Thirdly, a further and more widely felt influence of the Greek chorus derives from the mutation of Aristophanic ‘old’ comedy into the ‘new’ comedy of Menander in the 4th century bce, as a result of which the comic chorus lost its ideologically significant, role-playing function but kept its structural, episode-separating one. (Surviving Menandrian texts simply stipulate at certain punctuation points in the plot that there should be choric performances, linked sometimes to the arrival of a band of revellers; but no words are provided for them.) So comes into being the idea that a comedy should comprise several distinct dialogue-sections separated by something in a different performance medium, allowed for but not pre-scripted by the playwright. This sectional concept made a vital contribution to the development of Roman, and later of Renaissance Italian, ‘act’ theory; and from around 1500 it became theatrical practice in Italy when staging ancient or modern five-act ‘new comedies’ to place articulating intermedi between the acts: instrumental and/or vocal pieces (performed sometimes on stage, sometimes behind the scenes), or dances or brief musico-dramatic divertimentos. This quasi-choric practice – in conjunction with the use in 15th- and 16th-century spoken drama north of the Alps of the instrumental silete and pause (designed there to attract an open-air audience’s attention at the start of a show, hold it during changes of location in the action, or mark scene breaks of other sorts) – opened a rich seam of incidental ‘entr’acte’ music which was to be very productive for the next four centuries.
The musical intermedi for an Italian Renaissance drama – six of them, often: four between the play’s five acts and two more as quasi-prologue and epilogue – had three possible functions beyond the purely structural ones of punctuating the spoken play and covering time lapses in its plot. They could dazzle and divert by their novelty and inventiveness (e.g. the sudden explosion of rustic hoeing, sowing, mowing and threshing danced to a moresca after Act 1 of Terence’s Eunuchus as given at Ferrara in 1499); they could establish links with the ideas and/or situations of their ‘host’ play (e.g. the madrigals of Verdelot to texts by Machiavelli designed to go between the acts at a mooted revival of Machiavelli’s own Mandragola in 1526); and on special occasions they could reach outwards to honoured guests in the audience, as in the sets of very grand and costly mythological-allegorical intermedi which paid tribute to noble brides and grooms attending plays in the course of dynastic wedding festivities, as at Medicean Florence in 1586 (L’amico fido) and 1589 (La pellegrina).
Following this latter tradition in their scale and texture, lengthy and elaborate balletic or operatic intermèdes became a feature of the spoken Latin tragedies performed at the Jesuit colleges in France in the 17th century (the intermède-set constituting M.-A. Charpentier’s David et Jonathas of 1688 is an example); but entr’acte practice closer to that of the earliest Italian intermedi in scale, strategy and kind developed in English court theatricals and in productions by children’s companies in London in the decades around 1600 – see, for instance, Marston’s Sophonisba (1606) and Beaumont’s Knight of the Burning Pestle (c1607) – spreading eventually to London’s adult commercial companies, where it was common by the 1620s. The practice survived the Puritan closure of the London theatres (1642–60) to return with new vigour in the Restoration period, a vigour maintained until the Italian operatic invasion of London moved the focus of interest in musical novelty there from playhouse to opera house.
Dispensing largely with song and dance, Restoration and early 18th-century entr’acte music, as composed by Locke, John Banister (i), the Purcell brothers, Paisible and Croft among others, was generally designed for string bands which would perform in ‘music rooms’ above the stage and later in the newly fashionable orchestra pits which were to become a feature of European playhouses for the next 250 years. The entr’actes made up part of a show’s standard set of seven to nine pieces: four ‘act tunes’ between the acts, preceded by an overture or ‘curtain tune’, itself preceded by two ‘musics’ (each comprising one or two numbers) which were played while the audience was assembling and which music lovers might make a special point of arriving early to hear. As with the intermedi of the 16th century, the act tunes could simply function as ingenious or airy diversions from the plot of the spoken play, or they might occasionally allude to events outside the theatre (as with the ‘Lillibulero’ Henry Purcell smuggled into the Jig for his music for The Gordian Knot Unty'd) or mirror the moods of the scenes they followed or preceded. Some members of the audience doubtless chattered vigorously through all such pieces: a response to entr’acte music still being recorded painfully, by Bizet and Sullivan among others, two centuries later. But others listened carefully; and John Dennis, in line with some 16th-century theorists, drew neat parallels between Greek chorus and modern entr’acte in his Impartial Critick (1693), and in 1699 gave John Eccles careful instructions as to how the act music in their Rinaldo and Armida was to meld with the action, since ‘all the Musick in this Play, even the Musick between the Acts, is part of the Tragedy’.
This is an idea that recurs in Germany a few decades later. By then there were German troupes taking over from France a tradition of public theatre entr’acte-playing parallel in some respects to the English. (In the French line, for instance, between 1672 and 1686 M.-A. Charpentier had written not only danced intermèdes for spoken plays given by the company that became the Comédie-Française, but also several sets of purely instrumental ones.) It is a German essay by J.A. Scheibe on such orchestral stage music in his periodical Critischer Musicus (no.67, 8 Dec 1739: commended and summarized in no.26 of Lessing’s Hamburgische Dramaturgie, 1767) that definitively articulates the concept of carefully integrated, ingeniously mood-specific orchestral overtures, interludes and postludes. It was a concept called on increasingly throughout the Classical and Romantic periods, notably at first in German-speaking areas. Thus a reviewer in 1774 can say of Haydn’s entr’actes for a revival at Eszterháza of Regnard’s 80-year-old comedy Le distrait that ‘Haydn and Regnard vie with each other in producing distrait caprice’ so that ‘the play’s value is much increased’. Leopold Mozart could add detailed plot and character allusions to the manuscript of the near-symphonic entr’actes his son composed for a Salzburg staging in the same decade of Gebler’s Thamos, König in Ägypten; and Beethoven could divide a couple of the entr’actes he wrote for a Viennese revival of Goethe’s Egmont in 1810 into two sections each, the first reflecting the mood of the scene just past, the second adumbrating that of the scene to come: a practice recommended decades before by Dennis and Scheibe. Later and further north, narrative relevance remains important to composers, playwrights and actors. So Bizet in 1872 was praised by Alphonse Daudet for enhancing his peasant tragedy L’Arlésienne with, inter alia, ‘the lovely entr’acte of the family council … wonderfully beautiful, elegant, heart-warming’; and Stanford, writer of the incidental music for the actor-manager Henry Irving’s production of Tennyson’s Beckett in 1893, proudly remembered that Irving ‘always came down to listen behind the curtain to the last entr’acte (The Martyrdom) in order to get into the right mood for the final scene’.
However, by no means all plays in the 18th and 19th centuries had complete or even partial sets of specially composed preludes and interludes. Where some kind of integration of music with spoken script was at a premium, the best hope often was that music might be found from somewhere that would fit the occasion well enough and contribute the desired sense of consequentiality and/or fizz. The need for such may well have been at the back of Locatelli’s mind when he published six Introduttioni teatrali as part of his op.4 in 1735: pieces which by then may well already have opened some plays satisfactorily (in Amsterdam quite possibly) and could well do the same for others in the future; and it was certainly at the back of Schubert’s when, short of time to write an overture to complete his set of incidental pieces for Helmina von Chezy’s Rosamunde, Fürstin von Cypern in 1823, he first called on the one that he had written for Alfonso und Estrella to fill the gap, and later perhaps approved the plan of replacing it with that to Die Zauberharfe (the piece which as a result came to be known as ‘the Overture to Rosamunde’).
Levels of engagement with the drama could be much lower than that. In London in the mid-18th century, for example, inter-act music, like the three ‘musics’ before the show, was generally much more in the audience’s domain, so to speak, than in the play’s, comprising a kind of popular concert of songs, dances and orchestral numbers in rather unsteady harness with the tragedy or comedy in hand: as when a playbill for Shakespeare’s Merry Wives of Windsor at Covent Garden in 1762 promised ‘Between the Acts some Favorite Songs from the English Opera Artaxerxes’. Things were not always much steadier in the 19th century. In 1892, for instance, the acts of Wilde’s Lady Windermere’s Fan at its première were preceded and separated by orchestral selections from Mozart, Schubert, Mendelssohn, Auber, Suppé, Delibes and Godard among others; and a few weeks before, Bernard Shaw, in a piece on incidental music for The World (Music in London, 27 Jan 1892), roguishly affected to hold that the proper music to Hamlet comprised two grand marches by Handel during the action, ‘the entr’actes being selected from no longer popular overtures such as La Sirène, etc.’. Such things make it clear why there was space for Norman O’Neill, musical director at the Haymarket Theatre in London near-continuously from 1909 to 1932, to gain a reputation for increasing the aptness and subtlety of entr’actes for plays lacking purpose-made scores by introducing appropriate novelties and making adaptations from unhackneyed earlier music (by Purcell, Corelli and Vivaldi among others). O’Neill on occasion also continued a late 19th-century practice of writing short, atmospheric preludes to individual acts, played under lowered auditorium lights, so as to ease the audience back into the mood of the play itself after hearing a not over-apt entr’acte or hurrying back from the theatre bar. However, by O’Neill’s time some theatre and music critics were questioning the relevance of orchestral overture, entr’acte and/or act prelude in the spoken theatre. As early as the mid-19th century, Ferdinand Hiller had attacked the whole entr’acte phenomenon (Kölnische Zeitung, 20 Aug 1855) and some German theatres had stopped playing them altogether; such views gained wider currency especially in connection with performances of the novel ‘naturalistic’ drama of modern life and the intellectually demanding ‘drama of ideas’. Theatre managers too were beginning to jib at maintaining expensive pit bands (even ones that composers tended to find constrictingly small) outside opera, operetta, ballet and pantomime. And developments in electronics were poised to bring about a radical change in how things musical were managed between, and during, the acts.
Incidental music, §3: Within dialogue scenes
The drama of the ancient Mediterranean saw the beginnings not only of inter-act music but also of dancing and singing set into scenes of spoken dialogue. Aulos music, and sometimes choral song too, are called for in Greek ‘old comedy’ to accompany episodes of ritual dancing – Aristophanes’ Frogs and The Poet and the Women are examples – and also dances marking the resolution of a play’s action, as in Lysistrata and The Assembly Women. This starts a long line of festive and/or climactic dance in European spoken theatre. It is later manifest in, for example, medieval French farce and Shakespearean romantic drama (the Capulets’ ball in Romeo and Juliet; the ‘measures’ to celebrate the happy endings of As you Like it and Much Ado about Nothing), in the comedias of Lope de Vega and the Italian commedia dell’arte, in Enlightenment comedy (such as the wedding fandango in Act 4 of Beaumarchais’ La folle journée ou Le mariage de Figaro of 1784 and the ‘ballet général’ that concludes the play) and the picturesque, sometimes exotic dance-fêtes and dance-pantomimes ingeniously set into the early 19th-century melodramas of Pixérécourt and his school. A darker line runs through these centuries too: for example, the dances of medieval ‘vice’-figures, sour antimasque-type numbers such as the Dance of Cuckolds that ends Wycherley’s comedy The Country Wife (1675) or the sinister Walpurgis dances in the first part of Goethe’s Faust (1801). In later naturalistic, verismo and avant-garde drama, dance that is ostensibly celebratory is often used with tragic irony: the farandole in the Daudet-Bizet L’Arlésienne, the wild tarantella danced by the trapped Nora in Ibsen’s A Doll’s House; the Dance of the Seven Veils which brings about the climax of Wilde’s Salome (and which combines with Japanese nō-dance precedents to influence Yeats’s ‘Plays for Dancers’). Beckett’s post-World War II drama bids an ‘absurdist’ farewell to the tradition, presenting dance music that cannot be danced to and dancing without dance music: as when the terminally immobilized heroine of Happy Days can only sway to the Merry Widow waltz played by her musical box and the slave Lucky in Waiting for Godot (who ‘used to dance the farandole, the fling, the brawl, the jig, the fandango, and even the hornpipe’) moves through his one sad surviving routine in silence.
As for song in spoken drama, the monody and kommos of Greek tragedy (i.e. the ‘aria’ occasionally given to a principal character, and the sung lament shared between a principal and the chorus) plus the cantica of Latin ‘new comedy’ (passages of verse dialogue, especially frequent in Plautus, where a change of metre signifies a change from speech to some kind of singing): these are early instances of the Western ‘inset lyric’, providing precedents for later play songs, should precedents be necessary when – before the late 19th century, at least – song’s links with the spoken theatre are manifestly so widespread.
Solo song given to flesh-and-blood dramatis personae over the last 500 years is broadly divisible into two categories. First there are the songs sung by characters – principal characters sometimes – who make no claim to any ‘professional’ musical skill. These may find themselves singing in a métier-revealing and/or self-characterizing way as they go about their daily lives: the Grave-Maker in Act 5 of Hamlet, for instance, or Gretchen singing her Ballad of the King of Thule in the Evening scene of Faust. They may be adept at making conversational points through snatches of ballads, chansons, folksongs and other pop tunes, as are Calbain and his wife in the late 15th-century farce Savetier nommé Calbain, Old Merrythought in Beaumont’s The Knight of the Burning Pestle and several of the peasants in Büchner’s Woyzeck (1836). Their wits may have gone lyrically astray, like Ophelia’s in Hamlet or Lyonel’s in Thomas D'Urfey’s A Fool’s Perferment (1688), his six mad songs set for the première by Purcell. Or they may simply be so full of unbuttoned, companionable bonhomie that they burst out with what Feste in Twelfth Night calls a ‘song of good life’, or so full of parental, filial or romantic feeling that they have to express it in a lullaby, a lament or (most often) a love song. (Clärchen in Goethe’s Egmont and Solveig in Ibsen’s Peer Gynt are exemplars of the latter, their songs memorably set by Beethoven and Grieg.) Some lovers – Pantalone on occasion in the commedia dell’arte, for instance, Almaviva in Beaumarchais’ Le barbier de Séville, Peer Gynt himself – even risk personally serenading their mistresses. Others, like the boorish Cloten in Shakespeare’s Cymbeline (with ‘Hark, hark, the lark’), choose more prudently to have their aubading and serenading done for them by paid musicians: and this introduces the other category of solo singers in spoken drama: characters whose singing is a quasi-professional skill, and who can therefore be relied on to deliver more complex songs in a more polished way – and songs on any subject germane to the play. These may be foreground characters like Shakespeare’s fools and clowns and ballad seller Autolycus, or like the guitar-carrying servants (Scapino, Mezzetino) common in commedia dell’arte; or they may be the often anonymous attendant gentlemen, waiting women, music masters, servant musicians and (by definition songful) shepherds or shepherdesses so often called on, in 17th-century Spanish and English theatre especially, to put a musical brake on the action with a song whose sentiments reflect or ironically invert those of the principal characters at that point in the plot. In certain circumstances the ‘professionals’ may be virtually the whole cast of a play, as in such Brecht pieces as Happy End, Die Mutter and Mutter Courage und ihre Kinder (their lyrics set by Weill, Eisler and Dessau respectively): plays embodying Brecht’s belief that characterization should incorporate, not obliterate, the presence of the performer as such, a performer whose skills should include the ability to sing ballads, cabaret-type numbers etc. when such things are ideologically apt and effectively ‘alienating’.
An extension of the solo song is the group song set similarly into the spoken dialogue. Medieval French farceurs had a habit of interrupting the action of their plays with a surprise group-chanson and rounding off the show with another; and (under the influence of 18th-century vaudeville) group singing as a means of unanimous self-expression can break out anywhere in sophisticated mid-19th-century French farce: witness the vocal behaviour of the massed villagers visiting Paris in Labiche’s otherwise almost wholly spoken La cagnotte (1864). But most choral singing in dialogue scenes tends to be more event-specific than that. At appropriate moments there are simple overflows of shared euphoria (songs for cheerful sailors, soldiers, drinkers, harvesters, hunters, gypsies, patriots, wedding guests, nymphs and swains and such) or of shared distress (dirges for mourners, howls for madfolk), and more lengthy and elaborate ‘production numbers’: sung for example by the incense-burning Ancient British priests in Fletcher’s Bonduca (as revised in the 1690s with a Purcell score); the yet more ancient Egyptian priests worshipping the sun to Mozart’s music in Thamos, König in Ägypten; the villagers processing with their Shrovetide effigy in Ostrovsky’s The Snow Maiden (1873), their sung processional set by Tchaikovsky; or the decadent emperor’s retinue in the 1888 revival of Dumas père’s Caligula making a sybaritic anthem out of the Pervigilium Veneris to music by Gabriel Fauré.
Incidental music, §3: Within dialogue scenes
If the beginnings of song and dance for earthbound characters can be traced to Greco-Roman antiquity, the Middle Ages made a major contribution to incidental music through the convention developed in the spoken vernacular religious drama of the time that scenes associated with heaven should in some way be musical. This was a scriptural matter in part: angels are described as ‘cantantes’ in the Vulgate, for example at Revelation xv.3. But it was also partly a philosophic one, since Pythagorean-Platonic-Boethian doctrine held that heaven in a sense was identified with music, and partly an associational one too, since the liturgy in medieval cathedrals and abbeys was wholly sung. Hence in open-air civic theatricals (French miracle-plays and mystères, English Corpus Christi cycles etc.) music accompanies angelic visitations; characters touched by God such as Simeon and the Nativity shepherds may sing; and there is instrumental minstrelsy around God’s throne. The angelic songs were often performed by choristers from religious foundations nearby, hired specially, dressed for the part and bringing with them appropriate chanted (or on occasion polyphonic) Latin psalms, hymns, antiphons etc. (in France sometimes vernacular chansons spirituelles as well).
Similarly in the professional theatre of Golden Age Spain, the theatre of Lope de Vega and Calderón de la Barca, traffic with the Christian heaven (angels descending, saints ascending, the Voice of Conscience, the Word of God) was regularly presented with music. In post-Reformation Protestant countries, on the other hand, there was much less call for such things. However, stage music’s links with the supernatural continued even there, though now largely under a pagan sign. Shakespeare, the crucial (because eventually a hugely influential) case in point, presents the ceremonies of fairies and elves, the doings of a magus and his familiar spirit, and (in dream-visions or inset entertainments) the epiphanies of classical deities. In almost every case some kind of music is involved: partly because such things are direct emanations of a musically ordered cosmos, partly because folklore at the time associated faery with song and dance, and partly because the entries of pagan deities in Elizabethan and Jacobean drama derive from similar entries in courtly masquing, which had music wherever it went. In the long run, Shakespeare’s international fame and his pandemic restaging from the late 18th century onwards were to ensure that supernaturally connected incidental music for his works would be written in many lands and idioms – by Mendelssohn, Sullivan, Chausson and Sibelius, among others. In the shorter run, his musical scenes of the pagan supernatural take their place within the wider 17th-century English convention that music should figure in scenes of spoken drama which involved the operation of magic (white or black) and the presentation of mythological or allegorical ‘visions’ and ‘shows’ of gods, demigods and spirits: as with the moonlit rites of Hecate in Middleton’s The Witch (incorporated in part into the First Folio printing of Shakespeare’s Macbeth) and the Orpheus masque in Settle’s The Empress of Morocco (1673), set for the première by Matthew Locke. (A preponderance of such invocations, epiphanies, visions and shows in the later 17th century – all of them full of song, spectacle and often dance as well – turned a ‘play’ into a ‘semi-opera’.)
A similar convention operated at the time on the continental mainland: in the elaborate mythological plays that Calderón wrote for the court of Philip IV of Spain, for instance. During his prologue to Fortunas de Andrómeda y Perseo (1653), he has the personified Música declare that ‘it is better that the gods do not speak as the mortals do’ but rather ‘have a different harmony in their voice’; and accordingly (as the anonymous surviving score for Fortunas shows) Calderón’s Olympians discourse in song while his earthlings rely mainly on speech. North of the Pyrenees, plays with a high level of Olympian spectacle, like the pièces à machines of the Corneille brothers, called for music at moments of epiphany, heavenly flight, mythic combat, etc. This music, provided by Dassoucy, M.-A. Charpentier and others, added pomp, circumstance and perhaps some echo of the high-Renaissance idea of the harmony of the superlunar cosmos; but it had its pragmatic dimension as well. It ensured sonic continuity through the show. As Pierre Corneille explained in the ‘Argument’ to his Andromède (1650), the concerts de musique in those plays were for ‘the satisfaction of the ears of the spectators while their eyes were taken with the descent or ascent of a machine, or fixed on something which stopped them attending to what the actors might have to say’.
Baroque theatre does not see the end of the tradition that in spoken drama ‘the grateful and needful role of representing what is supernatural and out-of-the-ordinary falls to the lot of music’, as Busoni put it in 1911 re his music for Gozzi’s Turandot (1765). The tradition extends to the less precise otherworldlinesses of the 19th and 20th centuries. These have been on the biggest, most elaborate scale, as in the visions of the Mountain King’s court, the Great Boyg and the singing Leaves, Dewdrops and Broken Straws in Grieg’s music for Ibsen’s Peer Gynt and the mingled mysticisms of D’Annunzio’s neo-medieval Martyre de St Sébastien (where ‘the cult of Adonis joins that of Jesus’, as Debussy, D’Annunzio’s composer, put it); or on a scale more easily manageable in a fairly well resourced early 20th-century theatre, as in the seductive ‘call’ of the Hebridean island in O’Neill’s music for Barrie’s Mary Rose (1920) and the peremptory orchestral summons from the beyond which Britten wrote for J.B. Priestley’s Johnson over Jordan (1939); and on the smallest scale too, as with the use of single instruments in Strindberg’s Ghost Sonata (1907: a mysterious golden harp) and Arthur Miller’s Death of a Salesman (1947: an invisible flute) to symbolize the desirable ‘otherwheres’ of those plays.
Incidental music, §3: Within dialogue scenes
Instrumentalists in the spoken theatre from the 14th century to the mid-18th did not only perform preludes and interludes, accompany songs and dances, and underscore supernatural happenings. On stage, behind the scenes or in the pit they also provided ‘realistic’ music: music for ceremony (flourishes, triumphal entries, dead marches, etc.), for hunting scenes, for episodes of psychotherapy and stimulus (sweet airs to cure disordered brains, induce sleep, increase desire) and for war: the whole repertory of alarms, charges, retreats, parleys and the like needed for a theatrical battle. Such music does not sound of its own accord; it is almost always attributable to and commissioned by specific human or supernatural agencies within the fiction of the play itself. It was for the later 18th century to add a further resource to a spoken drama in the shape of ‘unattributable’ music: music rising from the orchestra pit which seems to come directly from the minds of a play’s characters or indeed the mind of the playwright, heightening the mood of certain scenes by sounding unbidden between or beneath the characters’ speeches. Several factors may have contributed to the growth of this phenomenon: music’s earlier use to support dumbshows and pantomimes; the spilling over of entr’acte music into adjacent dialogue scenes; the influence of ritornello and recitativo stromentato in opera. But the dominant influence was a new musical form, the mélodrame, as pioneered by J.-J. Rousseau and Georg Benda in the 1760s and 70s (see Melodrama). Spoken plays were soon borrowing and building on its techniques and related ones, and the way was open to a century and a half of within-the-scene mood music. This came in three principal kinds:
(a) First, there were the orchestral scores for the new kind of spoken drama which came itself to be called ‘melodrama’: a genre of Romantic theatre highlighting strong situations, black-and-white morality and sensational turns of events, set often against exotic backgrounds. The scores for melodrama in this sense of the word (by specialist or semi-specialist composers such as Adrien Quaisin, François Amédée, Henry Bishop, Etienne Singla, Robert Stoepel and Henry Sprake) comprised not only the by-now-expected overtures, entr’actes, dances and songs, but also music to underline all of a play’s highly charged dialogue episodes, significant entrances and exits, stirring incidents, strong scene ends and striking tableaux. The music supplied could be extensive (making particular use of the varied repetition of a set of melodic motifs associated with particular characters); but in general it did not force itself on an audience, operating instead in a way rather similar to the period’s versatile new stage lighting by gas, limelight and eventually electricity. Still, like that lighting, it was a crucial determinant of a play’s impact.
(b) The second sort of mood music was something of a poor relation to the first: ‘off-the-peg’ melodrama music, so to speak. For this, a theatre’s musical director would recycle the same all-purpose collection of tried and trusted musical devices over and over again from one melodrama to the next. He might assemble his collection personally, or buy a convenient passe-partout compilation like one advertised (quite late in the tradition) in 1912: ‘Incidental Music Suitable for Lively Rise of Curtain, Entrance of Characters, etc., Hurry, Combat, Apparitions, Pathetic Situations, Martial, etc., etc. Price Piano 4s., Full Orchestra 5s.’ Thereby in part hung the bad name that melodrama began to get in the later 19th century.
(c) At the other pole of mood music was musical mélodrame as it featured in plays other than melodramas. Quite early in the 19th century E.T.A. Hoffmann had written some observations on the use of choruses and mélodrames in spoken plays (Allgemeine deutsche Theater-Zeitung, 17–20 May 1808), warning against uncritical profligacy with under- and inter-speech music, and stressing the need for taste and economy in its deployment. It is in this spirit that most of the major composers who used mélodrame as an element in their scores for tragedies and other kinds of poetic drama do deploy it. Thus there is only one, four-minute mélodrame in the whole of Mozart’s Thamos score, and only one – at a place indicated by the dramatist – in Beethoven’s for Egmont (though there are more in his König Stephan music). The three in Weber’s 1821 score for Wolff’s Cervantes play Preciosa all focus strongly on the heroine. Mendelssohn in his music for A Midsummer Night’s Dream (1843) restricts mélodrame to scenes involving fairy characters, as Elgar later restricts it in his score for Pearn and Blackwood’s fantasy The Starlight Express (1915) to dialogue concerned with imagination and ‘star-sympathy’. All but one of Schumann’s mélodrames in his music for Byron’s Manfred (1848) involve invocations of, or communion with, supernatural or cosmic presences, as all but one of Bizet’s for L’Arlésienne underpin intimate scenes for the two brothers at the centre of Daudet’s tragedy. Where theatrical effect is concerned, ‘less’ in each of these cases arguably means ‘more’.
An overture, songs and entr’actes, a mélodrame leading to music for a supernatural vision, death scene music, a concluding Siegessymphonie: Beethoven’s score for Egmont lacks only dance and chorus music to make it a conspectus of all the ‘incidental’ modes known up to his time. His authority in the 19th century came near to making a set of such pieces in all or most of these modes a canonical ‘form’: a form, indeed, that was worth attempting even if there was no prospect of an actual production of the play to which one’s set was incidental. Thus Schumann had no commission when he imagined an ideal performance of Byron’s Manfred and composed his extensive score for it, detailed down to a two-bar cue to cover the exit of a witch; Sullivan chose as his 1861 graduation exercise from the Leipzig Conservatory an unsolicited set of Tempest pieces; and in 1905 Busoni wrote his set for Gozzi’s Turandot on the pure off-chance that it might one day be useful for a staging of the play in Italy, where since the birth of opera (and outside the commedia dell’arte) incidental music had not flourished. (Weber’s Turandot score of 1809 was not a contender, since it was written for what Busoni considered the over-germanized adaptation of the Gozzi by Schiller.)
If the Egmont set became canonic in status, it was also quasi-symphonic in its method: the final section of the overture is reprised at the very end of the play and one of the heroine’s inset songs is developed in an orchestral entr’acte. In emulation, Schumann and Sullivan (Bizet also in L’Arlésienne) allude to motifs from their overtures later in their incidental sets, Schubert in Rosamunde makes substantial links between his entr’actes and an inset chorus and ballet movement, and Mendelssohn in A Midsummer Night’s Dream mines his overture (independently written 15 years before) and two of his entr’actes for mélodrame material. The drift towards symphonic procedures in these and other composers (something Bernard Shaw describes wittily in the 1892 article on incidental music referred to above) is matched in the late 19th and early 20th centuries by a drift towards Romantic symphonic scale. Thus the scores by Grieg, Debussy, Elgar and Sibelius for, respectively, Peer Gynt, Le martyre de St Sébastien, The Starlight Express and The Tempest are all well over an hour in length – two indeed are close to an hour and a half – and all ideally call for a large orchestra with at least two solo singers and (the Elgar excepted) a chorus as well.
However, the last of these to be written, Sibelius’s Tempest score of 1925, proved to be something of a ne plus ultra. Full sets calling for such large resources and with such extended components became rarities after that, and were all but extinct by the middle of the century. Several factors contributed to this: the increasing reluctance of managements to maintain large pit bands for spoken drama; developments in stagecraft which tended to render lengthy inter-scene music (useful in the 19th century to cover elaborate scene changes) logistically unnecessary; and the spread of the feeling that anyway such music was often aesthetically superfluous (‘beastly tootlings between the acts’, as one objector put it). Modernism had its say too. Four years after Sibelius’s graphic storm-prelude to The Tempest, twice as long and for forces well over twice as big as Locke’s equally graphic ‘curtain tune’ to the play of 250 years before, Paul Claudel wrote in an essay on ‘Modern Drama and Music’ that his preference where theatrical storm music is concerned was in fact not for a large, hidden symphony orchestra but a single and thoroughly visible Japanese kabuki performer ‘with a tiny cup of tea by his side and in front of a tremendous drum, which it is his role to beat’.
The idea of using a visible and minimal band for modernist stage music had already been embodied at its most radical in the single (and musically versatile) figure of the Peuple de Zanzibar in Apollinaire’s ‘drame surréaliste’ of 1917, Les mamelles de Tirésias. And later in the century small ‘chamber’ ensembles, often hidden behind the scenes (as orchestra pits were going out of fashion outside the operatic and balletic world), but sometimes appearing on stage as a static or even actively moving part of the play’s visible presentation, came more and more to be the norm for incidental scores in such theatres as chose, and could afford, to use ‘live’ music. These ensembles, especially when visible to audience and actors, could mesh their music intimately and extensively with their host plays, as for instance in Birtwistle’s score for the Oresteia of Aeschylus as directed by Peter Hall in 1981 – ‘composed around’ (Hall’s phrase) the actors’ verbal responses to the text in rehearsal – and in the music, its materials prepared by Toshi Tsuchitori, that was partly improvised during the performances of Peter Brook’s 1985 staging of the Hindu Mahābhārata.
However, resources for another kind of integration of music and stage had by then become available through 20th-century advances in acoustics and electronics. Between the 1920s and the 1980s these generated the electrical gramophone (with short-playing, then long-playing records, then compact discs) and along the way the tape recorder, synthesizer, computerized sampler and advanced theatrical sound system. This technology could be used simply to play recordings of existing incidental scores, or indeed of operatic and concert music made incidental (sometimes with strong possibilities for irony) by its new use on disc or tape in the playhouse. It could also be used to transmit recordings of specially composed scores for singers and instrumentalists performing in traditional ways. Beyond that, the technology could create its own music (musique concrète, electronic and computer music), either in a ‘pure’ form or blended with sonic ‘atmospheres’ and other sophisticated developments of what traditionally had been the territory of theatrical ‘sound effects’.
So was produced the late 20th-century electronic equivalent, and to an extent descendant, of the assiduous orchestral scores (supplemented with ‘live’ sound effects) of 19th-century melodrama and spectacular drama, able to support or point up any word, mood, movement or visual aspect of a play’s mise en scène. The descent was a direct one in that 19th-century practices had had an impact on the extremely complex musical contributions to the stagings of classic texts directed in the first decades of the 20th century by such consequential figures as Reinhardt (who worked with several composers, Humperdinck especially) and Meyerhold (Mikhail Gnesin, Vissarion Shebalin etc.); and these in their turn influenced later creative directors with more various and ‘advanced’ sound sources at their disposal. A more oblique lineage was by way of the cinema. Melodrama music of the 19th century was the major influence on the music accompanying silent movies, and the latter’s conventions infiltrated the soundtracks of the first talkies, whose influence was later to be felt in the theatre. Film music, however, is another topic, if a closely related one.
Once the sequence of performances of the play for which it was composed has come to an end, incidental music has rarely had much permanence in a theatrical context. Plays often fall out of the repertory, and where they do not, revivals have seldom felt duty-bound to revive all or any of the original score. For example, the surviving music written for The Tempest in its various forms on the London stage in the first century and a half of its existence reveals a complex process of evolution, with contributions from at least ten composers; Racine’s Athalie had its choruses set at least five times; and Maeterlinck’s Pelléas et Mélisande generated at least three different sets of incidental music (by Fauré, Sibelius and William Wallace) in the first dozen years of its stage life. Even those few sets of incidental pieces which have come in particular ages to seem all but inseparable from their host plays tend to find themselves discarded eventually by progressive directors: most notably (after about 70 years in each case) Mendelssohn’s Midsummer Night’s Dream score, dropped by Granville Barker in 1914 – Cecil Sharp found him English folkdance and song with which to replace it – and Grieg’s score for Peer Gynt, dropped in 1948 by Hans Jacob Nilsen who preferred to commission new, tougher and rougher music from Harald Saeverud.
True, stage music has sometimes been kept alive in the theatre by transference from one show to another, as when Grigory Kozintsev re-used Shostakovich’s music for his 1941 production of King Lear, only slightly supplemented, in his 1954 Hamlet. In the name of organic unity, there have been occasional successful pleas since the time of Arnold Dolmetsch’s association with the Elizabethan Stage Society in the 1890s for the re-establishment of the original (or something like the original) incidental music in the case of modern revivals of plays from the past, and less frequently (and less successfully) for special revivals of classic plays incorporating notable scores that they have inspired in intervening periods – though Stravinsky surely had a point when he opined that a modern production of Shakespeare’s Dream with Mendelssohn’s 1840s music would have to be ‘clothed in the provincial German court style of the period’. But by and large, where ‘live’ performance is concerned, most incidental music has been bound for oblivion, unless, that is, some new use can be found for it outside the theatre. Several new uses have in fact been found.
One has been the concert presentation of a complete theatre score with a specially written linking narration performed by an actor: a verbindender Text, early instances being the Egmont narratives by Friedrich Mosengeil (1821) and Franz Grillparzer (1834). Another has been the publication of incidental items for use in non-narrative contexts: for example the several printed collections with such titles as ‘the newest and best songs sung at the court and at the public theatres’ appearing in the English Restoration and aimed at domestic music-making and the new-fangled public music-meetings. In 1697 London saw an ambitious and influential extension of this to purely instrumental music: the publication of A Collection of Ayres, Compos’d for the Theatre, which posthumously printed Purcell’s pre-musics and act tunes for 13 shows (nine plays and four semi-operas), reordering the items into effective free-standing suites, each with an overture followed by a sequence of airs, dances and sometimes ‘song tunes’ arranged from inset songs. The collection is a notable early instance of the idea of a published sequence of incidental pieces meant for non-theatrical performance. This idea had a brief English vogue around 1700, but did not become common until the concert-loving 19th century. There was a growing taste by the 1830s for the concert performance of orchestral stage music – for instance, Henry Bishop had a great success with a programme of his own in Manchester in 1839 – and from the 1870s the published concert suite named after its source play becomes a standard form. Composers preserving some of their incidental music in this way (sometimes reorchestrating it in the process for the bigger forces available in the concert hall) include Bizet, Fauré, Grieg, Sibelius, Nielsen, Vaughan Williams, Korngold, Roussel and Milhaud.
Further, some incidental items have gained new lives as concert miniatures, almost wholly divorced from their theatrical past (the Valse triste of Sibelius, once part of Arvid Järnefelt’s Kuolema; Debussy’s Syrinx, from Gabriel Mourey’s Psyché), and substantial sections from scores originally for the spoken theatre have been given new identities in larger forms by composers not wanting to lose sight of good material. In the 18th century, for example, Handel, loath to see the quite extensive score he had written for Smollett’s unperformed (and now lost) play Alcestis go to waste, reworked it into the all-sung Choice of Hercules; Haydn used his music for Regnard’s Le distrait in his Symphony no.60 in C; and Mozart seems to have had a hand at least in the scheme to provide the sun-worshipping choruses he wrote for Gebler’s Thamos with Christian texts which might encourage frequent church performance. Similarly, in the 20th century, Elgar used elements of his music for Binyon’s play Arthur in the sketches for his unfinished Third Symphony; Ibert turned his score for Labiche’s Un chapeau de paille d’Italie into the independent orchestral Divertissement; and Britten recycled his music for a nightmarish dance of typists and office clerks in Priestley’s Johnson over Jordan as the march variation of his Diversions for piano (left hand) and orchestra.
On the other hand, incidental music has also provided composers with the stimulus and/or part of the substance for an even more ambitious stage work on the subject of the host play. Tchaikovsky and Debussy thought well enough of their big scores for, respectively, The Snow Maiden and Le martyre de St Sébastien to plan turning them into operas, though the appearance of Rimsky-Korsakov’s Snow Maiden put an end to Tchaikovsky’s project as terminal illness did to Debussy’s. But Humperdinck’s very elaborate mélodrame score for Ernst Rosmer’s Königskinder (1897) did develop into his opera of the same name (1910); Busoni’s music for Gozzi’s Turandot was the substantial basis for his Turandot opera of 1917; and Vaughan Williams’s involvement before World War I with the provision of music for Shakespeare’s ‘Henriad’ and Merry Wives of Windsor at the Shakespeare Memorial Theatre, Stratford-on-Avon, led eventually to his Falstaff opera, Sir John in Love (1924–8). But perhaps it is the travels of a few bars of a later Shakespeare score that best exemplify the afterlives of incidental music. Tippett’s music for a production of The Tempest at the Old Vic Theatre in London in 1962 includes a setting of the invocation ‘Come unto these yellow sands’, which was published the same year with keyboard accompaniment as one of Tippett’s three Songs for Ariel, was then alluded to in the Tempest-based Act 3 of his opera The Knot Garden (1966–9), and appears fleetingly in the orchestral Songs for Dov (1970) which grew out of the opera before being given a final home in the cumulative Suite: The Tempest assembled in 1995. So are incidental music’s revels extended.
N. O’Neill: ‘Music to Stage Plays’, PMA, xxvii (1910–11), 85–99
A. Aber: Die Musik im Schauspiel (Leipzig, 1926)
D. Hudson: Norman O’Neill: a Life of Music (London, 1945)
R.W. Lowe: ‘Les représentations en musique au Collège Louis-le-Grand’, Revue de la Société d’histoire du théâtre, x (1958), 21–34; xi (1959), 205–12
H.M. Brown: Music in the French Secular Theatre, 1400–1550 (Cambridge, MA, 1963)
P. Hartnoll, ed: Shakespeare in Music (London, 1964)
E.T. Kirby, ed: Total Theatre: a Critical Anthology (New York, 1969) [essays on Reinhardt and by Claudel, 162–74 and 197–208]
N. Pirrotta and E.Povoledo: Li due Orfei: da Poliziano a Monteverdi (Turin, 1969, 2/1975; Eng. trans. as Music and Theatre from Poliziano to Monteverdi, 1982)
D. Mayer: ‘Nineteenth Century Theatre Music’, Theatre Notebook, xxx (1976), 115–22
E.J. Dent: ‘Music for the Cambridge Greek Plays’, Selected Essays, ed. H. Taylor (Cambridge, 1979), 26–36
C. Price: Music in the Restoration Theatre (Ann Arbor, 1979)
J. Dutka: Music in the English Mystery Plays (Kalamazoo, MI, 1980)
D. Mayer: ‘The Music of Melodrama’, Performance and Politics in Popular Drama, ed. D. Bradby, L. James and B. Sharratt (Cambridge, 1980), 49–63
G. Woolfenden: ‘The Composer’s Tale’, Backstage: Who Does What in Theatre, ed. J. Cook (London, 1987), 88–94
J. Wood: Music in Caroline Plays (diss., U. of Edinburgh, 1991)
L.P. Austern: Music in English Children’s Drama of the Later Renaissance (Philadelphia,1991)
L.K. Stein: Songs of Mortals, Dialogues of the Gods: Music and Theatre in Seventeenth-Century Spain (Oxford, 1993)
R. Rastall: The Heavens Singing: Music in Early English Religious Drama (Woodbridge, 1996)
N. Brand: Dramatic Notes: Foregrounding Music in the Dramatic Experience, (Luton, 1998),