1
INTRODUCTION
Something about singing
Oh sing unto the Lord a new song: sing unto the Lord, all the earth.
(Psalm 96:1, c.l BC)1
Orpheus with his lute made trees, And the mountain tops that freeze, Bow themselves, when he did sing.
(Sung by a maid to Queen Katherine in Henry VIII, 1613)2
Sie kämmt es mit goldenem Kamme, Und singt ein Lied dabei; Das hat eine wundersame, Gewalt'ge Melodei.
[She combed her hair with a golden comb And sung a song as she did, That had a wonderful Powerful melody.]
{Die Lorelei, 1822)3
Shout! Shout!
Up with Your Song!
(Suffragette anthem, 1911)4
There are many songs about songs, and thus there are and have been many singers singing about singing. Such songs and singing are rarely ironical, self-reflexive, estranging or postmodern. They are mostly whole-hearted, full-throated, unabashed
t
2 Introduction
affirmations of the feeling of singing. Some harness singing to other feelings: worship, love,, lamentation, high spirits. Many sing of singing for the sake of singing, just as some give way to pure semantic-free yocal utterance: a hey and a ho and a hey-nonino, ta-ra-ra boom-de-ay, da-doo-ron-ron-ron da-doo-ron-ron. Singing about singing declares that there is something about singing. Films can build and expand upon that sense of something, not least in the way space and time are handled in relation to it, in the process indicating and contributing to a given culture's perception of what that something about singing is and what's at stake in it.
Song is often apprehended as something magical. Angels sing and bring to earth a glimpse of the transcendent: God, eternity, heaven, paradise. Sirens, mermaids and the Lorelei sing and lure men to a watery grave. Sung mantras take the singer to a higher spiritual plane. The magical power of song lingers at the core (or second syllable) of two of the standard English words for magical effect: enchantment and incantation. Song is the medium of choice of religion and witchcraft. Even if we do not believe literally in the supernatural, we may still recognise in such images and practices a sense of the power of vocalising in a way different to speech. We know it too in more everyday terms: lullabies send the fretful to sleep, serenades seduce, psalms and hymns bring believers closer to God and each other, calls summon believers to prayer, football chants create solidarity and encourage victory, marching songs keep soldiers marching and work songs workers working.
There are many ways of accounting for this sense of song. In a series of aphorisms on the voice, the seat of song, Meredith Monk (1976: 56) wrote of it as 'a direct line to the emotions'. Song, like all art, is at the intersection between individual feeling and the socially and historically specific shared forms available to express that feeling, forms that shape and limit and make indirect what can be expressed and even with any degree of consciousness felt. There is something outside culture — the body has its ways, the mind its operations, the physical and psychological senses their responses - but the moment there is utterance we are in the realm of culture, of indirection. Yet we may all the same apprehend song as a direct line to feeling. Song is close, perhaps closer than most other aspects of music or most other arts, to the pre-semiotic: to breath and the unpremeditated sounds of crying, yelping, whimpering, and to such not-quite-spontaneous (or not always, perhaps not often) vocalisations as keening, laughter and orgasmic yells and the special registers of whispering, hissing, calling and shouting, as well as to mumbling and stuttering. Because birds, whales and other creatures sing, we may feel our singing connects us to nature in a special way. Song may well be, or be felt to be, one of humanity's first means of communication (Potter 1998: 198); it may take us to the first voice we hear, with its coos and teases as well as its lullabies.
The sense of song's venerability - its connection to first and abiding sounds — may also inform the experience of secular transcendence in song identified by Mark Booth in The Experience of Songs. He suggests (1981: 206) that song provides a
1'glimpse of authentic togetherness . . . offering us the experience of unity with what seems to lie apart from ourselves'. This is not primarily an argument about community or mass singing, or singing in unison or harmony or contrapuntally,
Introduction 3
all of which enable different experiences of togetherness. Rather, in song and singing 'musical tones transform speech away from the presupposition of separations among speakers, hearer, and object of discourse' (18) and thus dissolve the boundaries between the self and others and take people into a realm of reality other than the separate and sequential experience of time and space in ordinary reality.5 This provides a gloss on the specialness of songs and singing as well as opening out onto the issues of their relation to time and space.
A number of factors, beyond the various kinds of explanations just given, account for the specialness — or at the least, the particularity — of song and singing: the physical pleasure of singing, its uncanniness, its extremely rich semiotic mix of words, music and voice.
Not everyone takes pleasure in singing, doing it or listening to it, and many wish they could do it better. Yet it is common to hum, whistle, la-la as you go about your business — sounds that serve no purpose and generally express nothing very much in particular beyond the enjoyment, the comfort or cheer of making them. Singing in temple or church, in carriages, coaches or sleighs, at football and other sports matches, in drunken array, there is nothing unusual about any of these, and, if they do also sometimes have other ostensible purposes (prayer, urging on), they are also a joy. Only in very particular circumstances (at religious gatherings, perhaps, or in situations where you are pressurised to join in or. be a spoilsport) does anyone have to sing: you do it for the sake of doing it, and pleasure in others singing in part derives from one's knowledge of the pleasure of singing. Gino Stefani (1987: 23—4) argues that pleasure is intrinsic to singing, because of what is physiologically involved in both producing and hearing song:
A singing voice corresponds to the principle of pleasure, as a talking voice does to the principle of reality. Oral melody is the voice of pleasure: nature teaches us so from childhood. Why? First, because of the conditions of production: a pitched voice requires relaxation of the muscles involved in phonation, and this implies an emotive state of quiet, peace and tenderness in the person. [. . .] Secondly, pleasure is also to be found in perception: not only because of the transmission of this state of ease through the voice, but more radically because such a type of voicing requires less effort to listen to; in fact, singing is an easier message for the brain to decode than is speaking. Just as the line of speech, filled with articulations by consonants, is similar to the movement of walking, so the line of vocal melody, being continuous, flowing, relaxing and free from resistance, is similar to the movement of flying. Flying and dreaming: the longing to escape, the pleasure of escape from harsh reality; movement in slow motion, wide and fluid; power without obstacles.
Part of the pleasure of singing — whether in terms of how well one sings or of one's delight in others' ability — resides in mastery. Songs can be more and less demanding to sing, be sung with more and less finesse, security or surprise. Simon Frith (1998: 193) writes of
4 Introduction
the sheer physical pleasure in singing itself, in the enjoyment a singer takes in particular movements of muscles, whether as a sense of oneness between mind and body, will and action (a singer may experience something of the joy of an athlete) or -through the exploration of physical sensations and muscular powers one didn't know one had.
Judy Garland's cocked eyebrow and little laugh at the end of the 'The Man That Got Away' (see Chapter 3), a song whose lyrics and underlying tread are grim and despairing, indicates the pleasure in a song well sung, the voice tone mellifluous, the high notes reached and held, the rhythms of the melody ridden with ease. On the other hand, Lata Mangeshkar (like perhaps Joan Sutherland or Ella Fitzgerald) has a voice of seemingly effortless pitch and purity, the very lack of any sense of effort being part of what allows her voice to seem sublime and even transcendent (see below).
Song and singing are such common activities that it may seem perverse to suggest that there is also something uncanny about then, but this too relates to the sense of their specialness and their lending themselves to the supernatural perceptions and practices noted above. In part, the very fact that song is sound opens up the sense of the uncanny. A recent collection of philosophical essays on the perception of sound (Nudds and O'Callaghan 2010) explores the problem of deciding where, in perception, sounds are: at their source? in our ears? all around — that is, all around? moving in a particular direction or flooding space? Speech, animal and machine sounds are so anchored in their source and in practical identification that this may not bother us most of the time, but other natural sources are less straightforward. Where, for instance, is the wind that it produces such sounds, the wind, which is readily associated with breath and song? Song compounds this by how far the sound of singing can travel, even without the aid of mechanical amplification. There is something almost freakish to stand near an opera singer, especially one of slight or elegant build, and observe what an extraordinary, big, reaching sound they can produce. Contemporary audiences were astounded at Bessie Smith's 'ability to fill an auditorium with her voice alone, unaided by a microphone' (Antelyes 1994: 216), with this perhaps becoming all the more astonishing in the context of the possibility of amplification. As Danny Barker observed, Smith 'could fill it up with her muscle and she could last all night. There was none of this whispering jive' (quoted in Antelyes). Yet mechanical amplification also both normalised and made possible other uncanny perceptions, notably the fact that one could hear the intimacy of crooning and that 'whispering jive', even if one was not near it. Much of the handling of space in relation to song in films at once exploits the magic and downplays the strangeness of this ability to hear song where and how it might not necessarily be heard.
Song may also be formally uncanny. It imposes on vocal production musical time - rhythm, tempi, overall length, the sense of melodic ending - over against both the underlying, ongoing, undivided flow of time and the organised, analytical time (minutes and hours, deadlines, timetables) of practical life. There is something
odd about harmony, whereby two (or more) distinct notes exist as 'apart but simultaneous', and of'tune, where tones move around somewhere and come back to rest in some invisible place' (Booth 1981: 18). Song, as befits an oral art, makes great use of repetition and redundancy and thus has an overall tendency towards a sense of stasis, towards not going or getting anywhere, to a sense of tableau, of suspended time (Booth 1981: passim). This is highly suggestive in the context of time and space in film.
Pleasurable and uncanny, song is also a particularly rich semiotic mix for the statement of feeling. Because they have words, songs can name and ground emotions; because they involve music, they can deploy a vast, infinitely nuanced range of affects; because they are vocally produced, they open out onto physical sensation.
The words of songs can do more than name and ground emotions. They can tell stories, describe landscapes, promote political positions, make philosophical statements. However, even when not done in song, all of these, like practically all human utterance, have an emotional dimension, and this is usually even more to the fore in poetry, of which song lyrics are a particular form (to say nothing of the long tradition of setting pre-existing poems to music).
Words name emotions. To take only one of the more favoured emotional topics and only English language examples:
Love My Love I Love You Do I Love You? Love the Way You Lie With Love from Me to You What's Love Got to Do with It? Love Lifts Us Up Where We Belong P.S. I Love You In Other Words, I Love You
Songs do not need to use the word for an emotion in order to indicate it very precisely: there are many love songs without the word 'love' in them. There are also categories of song defined by their emotional address: hymn, lament, lullaby, serenade.
Naming an emotion indicates and specifies it. It brings it into verbal discourse, into the most privileged form of recognition, reflection and scrutiny in modern cultures. On the one hand, it limits the emotion — it's love, not anger or depression or just feeling good. On the other hand, it indicates something vast. The song 'Love', after listing the many things love can be, such as amusing, insane, a joy for ever, a dirty shame, concludes that 'love is almost never ever the same'. (Curious, that 'almost' with the insistent 'never ever'.)6
The simplest lyrics provide further nuance to the emotion. The Perry Como hit 'I Love You (And Don't You Forget It)'7 merely repeats these words over and
6 Introduction
Introduction 7
over. There is wit here: repeating the same refrain so often means it is not liable to be forgotten. There is also self-reflexivity in the one other line the song has: 'That makes seven times that I said it'. There are also hints of something else. 'Don't you fprget it' is a phrase usually used a little threateningly, but the repetition here may also hint at anxiety, that maybe thejoved one will forget that they are loved. This is already a complex set of emotions, suggesting that, beyond the declaration of love, it is not altogether clear what other emotions are fully or at all in play. The light-heartedness of the song, potentially in the wit and self-reflexivity, more evidently in the up-beat tempo of the music, itself complicates matters, for light-heartedness and humour are often a way of dealing with difficult and aggressive emotions. All that in an apparendy straightforward song in the supposedly simplest of song genres.
Words also ground the emotions they state, which in turn means they indicate space and time. At the simplest and also most inescapable level, they entail grammar, including person and tense, both of which are space-time dimensions. Person: / love you, You're just in love, Your eyes are the eyes of a woman in love, She loves you, He loves me, Love lifts us up. Tense: I love you, Love lifts us up, When I fall in love (it will be forever), It was a lover and his lass, Love me (that's all I ask of you), If I loved you (with the examples encompassing present, future, past, imperative, subjunctive). All imply spatial positions - for I or you or she or he must be somewhere - as well as temporal ones. Words may make this grounding more specific, indicating where the singer is or was, what time of day or the year it is and so on. The less spatial and temporal indicators the lyrics give, the more there may be a sense of the emotion floating free of such concerns, perhaps to suggest transcendent states of love, ecstasy, spirituality.
Words do not only have semantic dimensions. They have sounds and rhythm, with potentially wide affective qualities, and they are always, even when spoken, coming out of a body, involving effort, muscular activity and breath, sensations located in space and taking time to produce. With music, the affective is to the fore.
Music deploys a very wide and nuanced range of affective timbres, vasdy beyond the reach of language (or musical notation) to describe, nor readily separated into analysable components. Suzanne Langer (1953: 27) begins to give a list of music's affective qualities that lie before or beyond any more specific, nameable emotion: 'forms of growth and of attenuation, flowing and stowing, conflict and resolution, speed, arrest, terrific excitement, calm, or subtle activation and dreamy lapses -not joy and sorrow perhaps, but the poignancy of either or both'.
This may be linked to the flux of feelings experienced in the womb and in very early childhood, before these have been organised into more discrete categories. It is perhaps comparable to the kinds of pleasure, interest and curiosity we find in colours, sounds, scents, tastes, in flowers and, most pertinently, birds before they have geographic or symbolic associations or in jewels before they connote wealth and possession. It is suggestive of a capacity to make correspondences between the senses, if not necessarily synaesthesia in a strict sense, at any rate a perception of
shared logical form' (Langer's term). The interest, pleasures and displeasures of apparently purely musical sounds may be related to the qualities of the human voice, and its shapes and rhythms to those of human movement (gait, bearing, carriage) (Davies 1994: 206).
Musical affects are known, familiar, everyday and in large measure learned, that is, cultural; there is for the most part nothing mysterious or ineffable about them. We may nonetheless feel that there is something special about music qua music (or, for instance, the formal qualities of paintings or poems, the fall of a dress, the shape of a flower, the song of a bird) partly because its affects cannot be put into words. It gives voice (literally in the case of song) to those affective dimensions for which practical and analytical discourse have no language.
In practice, music's broad affective potential has been harnessed to specific aspects of feeling, to emotions. We do not need words to recognise, with a high degree of uniformity within a culture,8 that a given piece of music is happy or sad, inspiring, spiritual, sexy or romantic. Conventions have developed, which we learn so early they seem natural, to express all these and more, particularly valuable in theatre and film. These conventions also help to identify genre, place and period, even to the extent of certain musical modes suggesting landscape or the sea, the 'Orient' or the jungle, the ancient world or the Middle Ages (all of these based, but to very varying degrees of attentiveness - from the philological to the insultingly imprecise - on the sounds and musics of these places and times). All of this is mobilised by the music of song.
Music acts on our bodies: a rhythm prompts our feet to tap, a metre makes us want to dance; very deep notes seem to resonate through us, very high ones to make us wince; we readily drum with our fingers or play air guitar, or karaoke along. Perhaps with no other aspect of music is sensation so readily available as with song. Anyone who can sing, no matter how badly, knows the effort involved, the challenge of notes too low or high for our register, of having enough breath, the energy of keeping the rhythm going, the excitement of hitting the right note, of holding it. We probably register the way we have to hold our body differently, or the way certain cadences (that we might label sad) seem to demand that our body droop, others that we throw our hands exultantly up in the air. Even without seeing someone singing, we can sense much of this from the sound alone, from our recognition in the sound of our own knowledge of producing such sounds. In films we see singing.
If it is physical sensation that is the most distinctive contribution of vocality to song, yet voice too has emotional-semantic and affective dimensions. Delivery and tone - hardening, softening, louder and quieter, faster and slower - suggest both specific emotions and convey affective qualities. Voices carry markers of gender, class, ethnic, regional, geographic and other socio-cultural differences as well as the grain of individuality, while affect is carried in vocal timbre - smooth, rasping, bright, warm, husky, raw and all the other terms we might have recourse to without anywhere near exhausting the voice's affective range. Specific modes of song are commonly sung with particular styles of vocal delivery - such as folk, crooning,
8 Introduction
art song, gospel — which themselves carry the values associated with that kind of song, 'its historical provenance and common associations.
Song, then, involves words, music and voice, emotion, affect and sensation. In some traditions, and specific songs, one or other element may predominate. Words are more important in comic, narrative-'and political song, whereas for many other songs, the kind we identify in terms of love, longing, happiness, sadness, despair and so on, it may be enough that the words broadly indicate the emotion. In yet other cases, vocal delivery may be as or more important than either words or music: opera and soul, for instance. No song, however, fails to involve all three elements in interaction (if we leave aside certain kinds of wordless singing — vocalese, scat - as perhaps not really 'song' proper).
Pleasure, uncanniness and semiotic richness together constitute the something about singing. This may be felt as more or less (but seldom not at all) special. White so-called integrated musicals, the focus of the first essays in this book, are based on a perception of songs as special to the point of exceptional, such [hat it is a challenge to incorporate them into the unspecialness of spoken narrative. However, other periods and cultures may make the difference less marked. In Elizabethan drama, language is often rhythmic and poetic before it is clearly lyrical, and it is not always clear from the texts of them what was meant to be sung and what declaimed (Dubrow 2008: 216-7). African-American culture too suggests less of a divide: the relation of black preaching to gospel/soul music (exploited for instance in Hallelujah!, USA 1929), and of jive talk and the dozens to blues and jazz singing, seem in each case like a continuum rather than a separation, and rap unequivocally occupies the liminal space between speech and song. Isaac Hayes's vocals for the main Shaft theme (see Chapter 8) are sometimes near spoken and sometimes lift off into soul delivery, and many of the films discussed in Chapters 7 and 8 draw on the sense of music as 'the thing that is always there'. As discussed below, there is extreme fluidity between song and narrative in Italian cinema, while in Hindi cinema, where there is always song and dance, nonetheless the singing voice preserves a sense of specialness. There is never anywhere not something about singing.
All of the above are reference points for film and some — how singing fills space, its relation to time, the spatial and temporal co-ordinates of lyrics, for instance — already open out onto the question of space and time. I want to explore this some more here, specifically in relation to film. I begin with the overarching organisation of space and time in the vast majority of films (and certainly of the kind discussed in this book) — narrative — before turning to the fact that film involves looking at singing and to film's handling of time and, especially, space. Songs in films take up space: they literally take place in space and often in a particularised sense of space, but they also take time and occupy the space-time of the narrative. It is in all these senses that the essays in this book are about the space of a song.
Introduction 9
In film (and plays) songs take place in the course of a narrative also told by means of speech, gesture and action. There are traditions of theatre and film (e.g. melodrama and Hollywoodian underscoring) that do have more or less continuous accompanying music, but song in such cases remains distinctive since it is a departure from speech towards the otherwise largely unacknowledged realm of music. Opera and occasional musicals (e.g. Les Parapluies de Cherbourg, 1964; Jesus Christ Superstar, 1971; Les Miserables, 1980; Starlight Express, 1984; Rent 1994) are through-sung, but even here there is a distinction between arias (or, in the musicals, songs proper) and recitative, although, as both are sung, the closeness between them creates a dynamic- different from the shift from speech to singing in the works discussed here.
The relation of songs to narrative in plays and films varies enormously. They can just be an interruption, a change of pace, a pause for something different. They may set a mood or else function as scene-setting, establishing a locale as a particular kind of club or party or occasion, or even just when and where a scene is taking place. They may act as commentary or reflection on the action, or they may -and this is a common function in movies — be expressive in relation to character.
Before turning to this last, I want here to say a bit more about some of the other possibilities just mentioned, considering a rather different body of work from that focused on in this book. The debate on the issue of song and narrative is lively — even at times vitriolic - in relation to Elizabethan and Jacobean drama, including Shakespeare. Louis B. Wright (1927: 262), for instance, is robust:
A careful presentation of all the songs in Shakespeare from the point of view of the practical Elizabethan playwright [. . .] would show that many of the songs, frequently explained on grounds of subtle foreshadowing, atmosphere, or characterization, are chiefly nothing more than diverting entertainments.
Diana Henderson (1995: 160) notes how commonly songs in Elizabethan plays, 'seem to be part of the theatrical display, important as moments of virtuosity, as mood-setting, melodious interludes rather than verbal commentary'.
At the end of Love's Labour's Lost (c.1598), a play characterised by the particular refinement of its language, two songs by the 'learned men', the curate Sir Nathaniel and the schoolmaster Holofernes, are presented by the 'fantastical Spaniard' Don Adriano De Armado and performed by (presumably) him, the clown Costard, the page Moth, perhaps Sir Nathaniel and Holofernes and quite possibly sundry others. The songs are those of Spring ('When daisies pied and violets blue . . .') and Winter ('When icicles hang by the wall . . .'). Wright (1927: 262) comments that these songs, 'necessitating a return of the clowns to the stage, certainly left the audience . laughing, but the relation between the songs and the preceding play is slight'. Henderson (1995: 206—7), on the other hand, suggests that the very contrast between the language (and, presumably, music and singing style) of these songs and the preceding courtly dialogue acts as a commentary on what has preceded it: '[The songs'] verses relate routine and unidealised sensory experience yet exploit the
10 Introduction
harmony of sound and form to do so. They appear different from the self-conscious poeticizing in the play because of their greater referentiality to nature'. Thereby also, precisely because the songs are there for their own sake and not to express the personalities of either the main characters (a group of aristocratic young men and women) or the performers, the songs also act as 'a gesture at the immediate community between actors and audience rather than the idealized community of love's labour's won'.
Both Wright and Henderson acknowledge that the songs in Shakespeare (and other drama of the period) may be diversions and virtuoso displays. However, without pressing the argument towards expression of character or immediate commentary on what is taking place, Henderson shows that they may also contribute to the overall feeling of a play. Something between an entirely unified and a throw-everything-in organisation can be posited.
In the case of Twelfth Night (1601—2), for instance, it would be as wrong to say that the songs, sung by the jester Feste, are 'thoroughly extraneous [or] only slenderly functional, if at all' (Wright 1927: 263) as that they are wholly integrated. (to use a term much favoured in certain critical discourses on the musical, touched on in Chapter 2). If considered purely in terms of the literal meaning of the lyrics, they do often appear extraneous, but it may be the affective suggestiveness of the lyrics, and the kind of music they need to deploy — the kind of song they are - that is important. Feste's first song is 'O Mistress Mine' (11:3), sung in response to a request for a love song from Toby Belch and Andrew Aguecheek, but there is no reason to supposed that Feste or Andrew are in love or that the song gives voice to Toby's enthusiasm for the maid Maria. At most it is what suits Andrew and Toby's maudlin drunken mood.9 Feste's second song, 'Come away death' (11:4), is sung in the person of a man whose beloved is straying from him and who intends to kill himself. Orsino comments that this is a song favoured by unmarried women, presumably one of his moments of being cynical or despairing about women's love and the pleasure they may take in how much men suffer for love of them. Orsino has specifically asked for this song, as the one heard the night before, 'Methought it did relieve my passion much'. He might just conceivably make the comment about unmarried women to hide the fact that the song expresses his feelings of despair at Olivia's constant rejection of him, although Orsino generally loses no opportunity to display his feelings. In short, the song does not express Feste's feelings and can only tortuously be made to express Orsino's, nor does it express the character of feelings in the play as a whole (in which treacherous straying does not occur). On the other hand, it does contribute to the sense of melancholy in the play, a tone limning its different forms of comedy (courtly romance, sentimental cross-purposes, rumbustious old men and, in the Malvolio subplot, the caricature and punishment of Puritanism). Feste's last song, 'When that I was and a little tiny boy', which ends the play, does not need to be taken as him singing about himself: on the one hand, the reference to coming 'alas! to wive' bears no relation to anything we know about the character (there's no reason to suppose he is or has ever been married) and on the other hand, all the other references are, allowing for their
Introduction 11
being cast in the male gender, at such a level of generality (being a 'little tiny boy', coming 'to man's estate' and 'unto my beds') that, on the one hand, they could apply — eventually — to anyone, and on the other apply hardly at all to the characters in the play, none of whom have experience of marriage, leave alone ageing. The last verse effects a return from the fictional world of the play to the actual world of the theatre ('. . . that's all one, our play is done / And we'll strive to please you every day') comparable to that indicated by Henderson in relation to Love's Labour's Lost, and such a return likewise throws a tonal perception over all that has gone before. There is an inherent sadness within the comedy of Twelfth Night (Viola believes her beloved twin brother has just died in a wreck in which she is saved, Malvolio is 'most notoriously abused' and almost everyone longs for someone they can't have), nearly kept at bay at by the high spirits and pretty language but connecting, as Feste's last song suggests, to the recurrent disappointments and raininess of human life.
Songs then do not have to be about character or the immediate situation in I order to be something more than enjoyable but extraneous interludes. However, one of the most common functions of song in Western cinema is indeed linked to character. Heard songs may induce a feeling in a listening character, they may make him or her feel sad or fall in love or they may awaken or bring to the surface a feeling that is already there: They may also express what the singer is feeling. Because songs are frequently in the first person, because they mobilise affects beyond discursive consciousness, because they come out of the body, it is common to consider them expressions, even emanations of someone, of the singer, the character, the composer, the lyricist, even the listener or the collective audience.10 Biopics and liner notes are two paratextual modes especially inclined to emphasise that a song was written and/or performed in — or rather, out of- particular personal circumstances, that it expresses the feelings prompted by those circumstances. Love songs around the world are commonly presented as the means by which one person conveys his or her feeling to another, often as if this is indeed a transparent release of what they are feeling inside; this is very obviously so in developed narrative fictions, but it is residually so in most music videos featuring love songs, and the very use of personal pronouns in love songs implies a narrative. We are perhaps unlikely to consider that the 'you' of Lionel Richie's 'Hello' ('Is it me you're looking for?') or Lady Gaga's 'I Want Your Love' is anyone particular that these performers and composers have in mind, or that they are actually feeling that feeling as they perform or write, but the songs nonetheless mobilise a vestigial scenario in which someone singing of their love to another is expressing what they feel, inside, to another person.
In shows or films in which the main character is a professional performer, the onstage or on-diegetic-screen number may be significant in terms of their success and not necessarily be understood as expressing the particular feeling they are singing about at any given point. However, high points are commonly moments when what is being performed in public is also presented as an expression of the performer's inner self. Most of Fanny Brice's onstage numbers in Funny Girl (USA
12 Introduction
1968) have nothing much to do with her state of mind, but the last number, 'My Man',11 when she has agreed with husband Nicky that they must separate, is to be seen as a all-out expression of her love and despair. The words of the song are actually about an ongoing but abusive relationship that, despite everything, the singer clings to, which is not the exact situation in which Fanny finds herself, but the number draws on the song's torch quality to make it a song about what Fanny is feeling (in a way untrue of actual torch singing, including Brice).12 Characteristically, Barbra Streisand as Fanny starts very quietly, but whereas this is usually with her beautifully modulated and phrased, here there is hesitancy, notes not held, slightly too long pauses between phrases, even off notes. We have already seen her walk forward to rapturous applause and convey, in her not smiling with pleasure at her reception, in a slight wavering of the head and long pause before starting to sing, that Fanny is not quite sure she can pull herself together enough to sing; when she does start, the imperfections of the delivery suggest the ordeal of singing, an ordeal that at first we may take to be that of the show having to go on, of performing in the face of personal tragedy, but which becomes the expression of that tragedy. Facial expressions strongly underline specific lyrics: on 'When he takes me in his arms', Streisand's expression suggests the ecstasy of being so taken and then snatches it back in the recognition that such taking is no longer available to Fanny. All of this runs through the performance the first time through the song, but when she begins the second time through Streisand moves to her smooth belting mode. Fanny is now back in control of the number. Slight alterations to the tune (e.g. on 'He'll never know', 'The world is bright alright', with a pitch-perfect soaring note on the last) and an unusual roughness on 'What's the difference if I say I'll go away' intensify the emotion expressed, underlined by a move to a close-up showing tears running down her cheeks. Taken all together, the song declares, pace some of the lyrics, 'I still love him even if I don't have him' and the whole number says this is what Fanny is really feeling, even though she is performing this on stage (many fans may also feel that it is what Streisand was feeling at the time of recording the song, even if there is no direct relation to her own love life).
The Hollywood musical biopic often works towards moments such as these. Some of the power of 'My Man' derives from the fact that all the other onstage numbers in Funny Girl are not about Fanny's inner feelings, not expressions having to do with her personal life: 'My Man' is the coming together of the private and the public, a disclosure of the self central to the illusions and pleasures of much modern Western popular culture. Biopics do not have to work like this, however. In Gainsbourg (France 2010), for instance, the eponymous Serge's songs relate to his feelings, but not directly or as an outpouring of feeling. The relation is oblique or glancing. When Boris Vian sings his and Alain Goraguer's 'Je bois' and Serge interpolates his own 'Intoxicated Man' (whose opening words, in homage to Vian, are 'Je bois'), what is evoked is their camaraderie and pleasure in singing and word play at least as much as anything specifically to do with drinking — the reasons for doing so in Vian's case, the effect of it in Gainsbourg's. 'La Javanaise' is a song already written before Serge meets Juliette Greco (though many may know
Introduction 13
Gainsbourg wrote it for her); its tentative and playful character suggests, guardedly, his feelings for her (in love with her without ever having met her, insecure regarding his own attractiveness); it sings of a passing love ('We loved each other/For the length of a song'), which might be a comment on the significance of Greco in Gainsbourg's life but is not an expression of how he feels, as it occurs in the film, when he plays it for her for the first time at her place. The close-up of her at the window as the rain pours down, after he has left (summoned by his rightly suspicious wife), leaves the significance of the moment with Greco rather than composer, singer and biopic subject Gainsbourg. Invited to compose for the adolescent France Gall, Serge overcomes his disdain for pop music for clearly paedophiliac reasons, even acknowledging, in the context of his reading Lolita every night, the pleasure of corrupting youth; the resultant hit, 'Baby Pop', is cynically popcorn but performed with gusto by Gall, eventually stripped off in front of him, the song thus functioning as both a full-of-fun pop song and a celebration of baby doll pleasures (with, however, words that are less straightforward, stressing how brief such 'baby pop' fun will be before the dreariness of adulthood sets in, giving perhaps a rather hysterical edge to Gall's delivery of the 'Baby Pop' refrain). 'Le Canari est sur le balcon', which we see him rehearsing with Jane Birkin, seems, other than a reference to London, to have nothing to do with their relationship in words or musical atmosphere; we learn of what would become his greatest and most scandalous international hit, 'Je t'aime (Moi non plus)', after he and Jane Birkin have recorded it, and while it might express their sexual pleasure in each other (notably when played over them in the bath together), the ambiguity of the title ('I Love You (And Neither Do I)') and the excessiveness of the recording with its heavy panting may not be asking for the song to be taken entirely straight.
This oblique approach relates to other aspects of the film. First, Gainsbourg does not for the most part use the original recordings, with the exception 'Je t'aime (Moi non plus)', which is not shown being recorded (and thus we don't see Serge and Jane singing it). The actual vocal performances of the other songs mentioned in the previous paragraph are by Lucy Gordon (as Birkin), Eric Elmosnino (Gainsbourg), Sara Forestier (Gall), Anna Mouglalis (Greco) and Philippe Katerin (Vian). On the one hand, compared to, for instance, TheJolson Story (USA 1946) (Al Jolson), Tina: What's Love Got to Do with It (USA 1993) (Tina Turner)13 or La Mome (La Vie en rose) (France 2007) (Edith Piaf), this means that who we see singing is who we hear singing, situating the song in the moment of the characters' lives. On the other hand, unlike say Barbra Streisand as Fanny Brice or, at any rate as far as the producers must have assumed, Diana Ross as Billie Holiday (Lady Sings the Blues, USA 1972), the seen actors are playing the parts of people with very and still familiar famous voices, so that we can know that, good an impersonation as it may be in some cases, this is not the voice of the person referenced by the film. When we do hear Gainsbourg himself singing ('La Valse de Melody', 'L'Hippopodame'), we do not see Elmosnino as Gainsbourg singing. Second, the whole film has a cartoonish quality, most obviously in the use of Serge's early drawings and then of larger-than-life puppets to suggest Serge's alter egos but also
14 Introduction
Introduction 15
in some of the performances, including Elmosnino's mannered Gainsbourg and Laetitia Casta's exuberant Brigitte Bardot (making their duet, 'Comic Strip', at the height of their affaire, at once perhaps the most direct expression of feeling in the film,- precisely because the most abandoned to a cartoon aesthetic, but also disconcerting — the performance.prompted as it seems to be by Brigitte's saying uneasily to Serge, when he tells her he has a comic strip song for her, 'Is that how you see me?'). Third, compared to a Hollywood biopic (or indeed La Mome), Gainsbourg has a rather elliptical structure, often missing what one might take to be the high or turning points in the protagonist's life: the first meeting with Bardot, his initial encounter with Birkin, the birth of his children, the moments of composition or of acclaim. In most musical biopics, these would be moments of plenitude ripe for amplification, but Gainsbourg largely sidesteps them.
All this may relate to the sketching in of Gainsbourg's Jewish background. There is no suggestion that there is anything Jewish about his music, but what is brought out is his readiness to please at school (by drawing erotic cartoons for his classmates and even one of the teachers) and his denial of his Jewishness at the Catholic boarding school he attends during the war. This might give a particular resonance to the rather chameleon-like quality of his work, moving from the more traditional chanson mode of'La Javanaise' and 'Intoxicated Man' to the candy floss of'Baby Pop' and, via a pop cover of a theme from Dvorak's New World Symphony in honour of Bardot ('Initials BB'), on to the heavy metal of 'Nazi Rock'. The latter, first bursting in as Serge kneels over his father's corpse and eventually fading away as he cradles and weeps over his dead dog, is presumably meant as provocation fuelled by grief, a heavy irony comparable to the young Serge's turning up early on the day of issuing the yellow stars for Jewish citizens under the Occupation so that he can be the first to sport one. This mixture of chameleon and provocateur might be discerned in the reggae version of La Marseillaise, 'Aux Armes et Caetera'. The film emphasises the right-wing outrage that greets this setting of the French national anthem to a Caribbean (and not even French Caribbean) musical form. However, there is also something casually racist, as the film presents it, in this act of appropriation: Serge is told, perhaps with a suggestion of sneer, of the sacred quality of the music for Jamaicans, but, cheerfully reckoning on Jamaicans not understanding French or knowing the anthem, he goes ahead with the song, ripping off a black musical culture in order to be naughty, from within the privilege of whiteness, about a white sacred cow.14
The oblique approach to song and narrative in Gainsbourg does not characterise the films discussed in the essays here. Songs are often expressive of what characters are feeling: Esther (Judy Garland) in Meet Me in St. Louis indulging her longing for 'The Boy Next Door'; Jerry (Fred Astaire) really telling Dale (Ginger Rogers) that he loves her ('Cheek to Cheek' in Top Hat) and Tommy (Gene Kelly) the same to Fiona (Cyd Charisse) ('The Heather on the Hill' in Brigadoon); Claire (Ann Miller) in On the Town giving vent to her enthusiasm for a 'primitive mate' and infecting others with it; everyone in Car Wash picking up bits of the tracks on the radio and making them their own. Other examples are less straightforward. Esther
(Judy Garland) in A Star Is Bom, in singing 'The Man That Got Away', is not singing about any particular man that got away, nor does Garland sing it as if she is singing about any such in her own life. The caressing intensity with which Lena Home sings 'Can't Help Lovin' That Man of Mine' in Till the Clouds Roll By might suggest that she is really feeling the emotion as she sings it, but quite who she is at that point of the film (the character Julie in Show Boat? an unnamed singer playing that part? 'herself'?) is so confused as to make it impossible to connect with confidence the conviction of the performance to the putative life of the character/performer. With the eponymous 'Shaft' or Curtis Mayfield's 'Pusher Man' in Super Fly, discussed in the final' chapter, the songs may reasonably be taken to express the attitude of the singer/songwriters towards the characters about whom they sing, their feeling about them - respectively, macho-worship (with a tinge of self-awareness) and insinuatingly ironic - but it is as much an attitude towards the type represented by Shaft and Priest as to them as individual fictional characters. In the last section of this introduction I discuss further the case of Pakeezah, where the songs are explicitly about the characters singing them and yet are not best understood as an outpouring of emotion a la Fanny and 'My Man' in Funny Girl.
Car Wash, and those moments in blaxploitation movies when characters may be said to be attending to the music that is playing, are unusual in this book in that the characters themselves do not sing (though they occasionally mime or dance along). Pop music in film is often not sung by the characters and yet is foregrounded in a way that associates the song with a character in terms of their sense of self. The films — sold in some measure on the presence of the pop soundtrack - work with the way that pop songs are used by fans in terms of self-expression. Ian Garwood (2006) shows that this process in the films is by no means straightforward: the mise-en-scene of'Everybody's Talkin" in Midnight Cowboy (1969) and 'Stayin' Alive' in Saturday Night Fever (1977) suggest an aspiration or self-image in the characters to whom they are attached — Joe (Jon Voigt) and Tony (John Travolta) respectively - at some odds with their actual situation. Baby It's You (1983) critiques the ideal identities embodied in its various tracks (by among others the Dixie Cups, Frank Sinatra, Dusty Springfield and Bruce Springsteen). Jackie Brown (1997) moves towards a more positive conclusion that suggests 'that the title character deserves to be aligned with the ideal identity proposed by the song ['Across 110th Street']' (Garwood 2006: 160). In short, the way in which songs may be an expression — an emanation, a declaration, an indication - of what a character is feeling or what a character is, is not secure, predictable and consistent, but nonetheless the sense of some relation, direct or travailed, between song and the intimacy of individuality is a dominant trope of Western cinema.
Films characteristically not only embed songs in narratives, they also show us people singing. We are accustomed in our time to hearing singing without seeing it, and even before such mechanical audio reproduction, looking at singers - in temple or church, during a feast or carnival, at court, in the light-filled, interaction-busy audience spaces of opera houses and music halls, in night clubs - may not have been so focused as it is by film, or the darkened stages and silenced auditoria
16 Introduction
of modern theatre and concert going or, most recently, television singing competitions such as The X Factor and Pop Idol. On the other hand, both the pop/rock gig, with its distant performers and enormous screens, and pub karaoke suggest other relations of space trj the sight of singing.
In music videos, where there can., be a maximum disregard of space-time co-ordinates, with little or not attempt to construct, let alone reproduce, a coherent space-time continuum, nonetheless when we see the person we hear singing, we usually also see them singing and at the right point in the song - despite a surrender to virtuality, the bond of the mouth and the voice remains intact. This suggests the hold of the sense in Western tradition that singing is in a peculiarly intimate relation to the body and thereby to the capacity and often the person of the singer. To suggest any break with that relation may be considered a violation.
At the premiere of the film The Dancing Cavalier towards the end of Singin' in the Rain (USA 1952), the audience calls out for the star, Lina Lamont (Jean Hagen) to sing, having just heard her sing so dehghtfully in the film. As we know, Lina has a horrible voice, speaking and singing. Backstage, Don (Gene Kelly) and Cosmo (Donald O'Connor) persuade Kathy (Debbie Reynolds) to sing 'Singin' in the Rain' for Lina while the latter mimes to it. Kathy is furious: she has already dubbed Lina's voice for the film and, at Lina's insistence, had the screen credit for doing so withdrawn. However, she does as she is told, and as she sings and Lina mimes, Don and Cosmo raise the curtain behind Lina to reveal to the audience the true source of the voice.
Cinema, in the actual mechanical separation of audio and visual tracks, has always allowed for a separation of voice and body in the presentation of song. Seeing Kathy singing 'Singin' in the Rain' is an affirmation of Hollywood's good faith, that it is not involved in severing the voice from the body by technology. Part of the force of the moment comes from its narrative significance. First, it is the moment of Lina's comeuppance. She has throughout the film been portrayed as a shrill, selfish, mean diva; now at last she is exposed to the ridicule of the onscreen audience. In particular, this is pay-back for her meanness in not allowing Kathy to have any credit for the singing. Second, it is the moment that affirms Kathy's stardom. When the curtain is raised to reveal Kathy singing behind Lina, she runs off stage into the auditorium in tears, but Don calls out from the stage, 'That's the girl whose voice you heard and loved tonight! She's the real star of the picture!'. Kathy has been a showgirl and bit player up to this point, but now she is recognised as a star, her quality affirmed by the man she loves. This relates to the third narrative investment in the sequence. Kathy feels utterly betrayed by Don's putting the interests of the studio and Lina before his love for her, a love he has earlier wanted to keep from the studio and the public, because he and Lina have been promoted as an ideal off-screen couple. To have to dub for Lina at this juncture is a humiliation that also means, as far as Kathy is concerned, the end of their relationship ('I never want to see you again'). The anticipated — indeed generically required - resolution of the film, Kathy and Don, Gene and Debbie, together, is thrown into crisis. It is resolved in the moment of truth, the revelation of Kathy's
Introduction 17
voice. The moral and emotional anxiety set in train by his making her fake things again for Lina (and behind that the shade of the possibility that, this being a film, any voice might belong to any body) is resolved by the truth of the sight of singing.
The sense of the importance of the moment is reinforced aurally and visually. In the cross-cutting between shots of Lina miming and Kathy singing, there is a change of sound quality, so that Kathy's voice seems closer, more intimate, less touched by the apparatus of mechanical reproduction than it sounds when Lina is in shot: we are privileged to get a more real Kathy than the audience in the theatre does. Kathy stands straight, only moving slightly, singing close to the microphone: this is professional, aurally intimate and also a refusal to affirm bodily the cheeriness of the song (Figure 1.1). Lina, on the other hand, makes a show of singing, standing somewhat away from the microphone and making any number of singers' gestures in accordance with the tone of the song. At the start of the number, even before the singing begins, Lina moves her hands and arms rhythmically in a gesture sent up by Don, Cosmo and studio head R. F. (Millard Mitchell) standing in the wings (Figures 1.2, 1.3). The truth of Kathy's singing is affirmed in tandem with the falsity of Lina's.
Singin' in the Rain is nothing if not complicated and not least in its blithe treatment of the authenticity of the voice. This is true first internally to the film. Kathy is only finally convinced of Don's love when, after calling out to her from the stage,
I
FICURE     MGM Ra'n" CatHy Pebb'C Rcyn°lds) {Si^n'm lh" Ram>
18 Introduction
Introduction 19
FIGURE 1.2 'Singin' in the Ram': Lina (Jean Hagen) (Singin' in the Rain, MGM 1952)
he starts to sing 'You Are My Lucky Star'. She turns and walks back towards,him: when they meet on the stairs by the side of the stage, she picks up the song from him. Their love is now celebrated in public, the revelation that she is the true star is fused with the revelation of their true feeling. Yet the sense that this is, also, a performance is suggested by a number of details. Cosmo, all along adjunct to their relationship, climbs down into the orchestra pit to take over conducting, servicing their performance of love. As Don starts singing, he is standing in front of a screen, half on screen, half onstage. When he and Kathy move back onto the stage, the screen has disappeared and they sing before a curtain: any whiff of screen image has disappeared in the face of the live performance (though what they are enacting is, as we're about to see, the screen image they will become). At the end of their singing, the camera moves in on Don and there is a dissolve to his face on the billboard for a film starring him and Kathy; before it stand Don and Kathy themselves, looking up at the film that affirms their love and their stardom (and their love as stardom). The title of the film in question is Singin' in the Rain. The fusion of truth and performance in the premiere theatre is finally accomplished in the film they have made, which has the same title as the film we have been watching.
So many slippages to affirm the truth of what we are seeing suggest just how important it is and perhaps how anxiety-inducing is any sense of it not being so. The final twist is external to the film. It is widely rumoured, and claimed by Debbie Reynolds herself (1989: 89),15 that when Kathy dubs for Lina's 'Singin' in the Rain'
(and for 'Would You?' in The Dancing Cavalier), the voice is that of Betty Noyes, not Debbie Reynolds (while Jean Hagen herself dubbed Kathy dubbing Lina's speech on film). Just as the curtain rises to reveal the reunification of body and voice in Kathy, only for the film to complicate this by the play on performance and screen, so the film's openness and reassurance about dubbing masks its own deceitfulness about it. In their production history of the film, Earl J. Hess and Pratibha A. Dabholkar say that it is Debbie Reynolds' own voice in the final 'Singin' in the Rain' number, but that Betty Noyes dubbed for Reynolds on 'Would You' and 'You Are My Lucky Star' (2009: 145). The first of these last two is a sequence in which Kathy dubs the soundtrack for Lina, and includes cross-cutting between her and Don that affirms their love: in other words, in this double moment of truth, of the voice and of love, of'"singing from the heart" [which] earns the right to the sound track' (Altman 1987: 258), the voice is faked. The same is true of You Are My Lucky Star', the song Don and Kathy sing together at the end, the triumphant public demonstration of the truth of their love and (singing) stardom. Whatever the truth about what Noyes, Hagen and Reynolds actually sang, anyway you look at it leads (by the norms of the Hollywood musical) to deceit.
Peter Wollen, Susan Smith, Carol Clover and Steven Cohan have explored the ideological anxieties in aU of this. Clover (1995: 723-5) and Cohan (2005: 234-8) point to the multiplicity of deceits or near-deceits in the film: the virtual plagiarism
4 „„ä ii-
FIGURE 1.3 'Smgm' in the Rain': R. F. (Millard Mitchell), Cosmo (Donald
O'Connor) and Don (Gene Kelly) (Singin' in the Rain, MGM 1952)
20 Introduction
of Cole Porter's 'Be a Clown' for 'Make 'em Laugh', Kelly dubbing Reynolds's taps in 'Good Morning', the re-use of sets and costumes, citations of previous Kelly movies, Jean Hagen as a parody of the already parodic Judy Holliday, Don's early career as a stuntman and thus double, his recounting of his career to a reporter in terms of his maintenance of 'dignity' belied by images of undignified slapstick and brawling, his mobilisation for 'You Were Meant For Me' of wind machine, dry ice, lighting and studio set 'to create the romantic setting which allows [him] to confess his love, the traditional product of the dream factory' (Wollen 1992: 17).16 As Wollen notes, 'the site of deception [is] the image of the woman' (56) and Smith (2005: 70—5) explores the way that the film's affirmation of one woman's voice-body unity is at the expense of another woman and under the control of men (who, even at one point literally, pull the strings). Clover emphasises the degree to which the film occludes the centrality of African-American culture to the dancing in the film, noting both the slips that reveal awareness of it (a reference to 'Mammy', dancers in blackface, the significance of the street setting of the title number) and Kelly and O'Connor's familiarity with black dance. Such occlusion is near ubiquitous in HoDywood cinema but it is striking in a film that seems to be displaying the Hollywood musical's foundations. In this perspective, the film's 'real subject is not white women's singing voices, but black men's dancing bodies' (1995: 737).17 Cohan (2005: 239-42), noting the supposedly asexual Cosmo and the camp of Jean Hagen as Lina (and of the prissy language coach), points to a further occlusion, that of the gay men also so central to the development of the musical, a fortiori in the MGM Freed unit.
Yet another perspective opens up when one sees the Italian release version of Singin' in the Rain, Cantando sotto la pioggia. Here all speech is dubbed but all the singing is retained from the American original. Thus when Kathy opens her mouth to sing, it is, as it were — and as long as you don't know the back story concerning Betty Noyes — even more the real Debbie Reynolds, who when speaking in the rest of the film has the voice of the practised (though extremely young — fourteen at the time) Flaminia Jandolo.
Dubbing is widespread in Italian and Indian cinema, creating a different relation between the sound and sight of singing, a point to which I return below. It was much less common in Hollywood. As long as films, and a fortiori musicals, were centred upon singing stars, whose records they were also promoting and selling, singers sang for themselves, although this actually meant dubbing themselves so that when filming attention could focus on how they looked. The norm of self-dubbing remains in place in pop, rock, soul and rap musicals, although the signs of effort may be less concealed, appreciated as part of the emotional and physical impulse of the performance, while out-of-sync self-dubbing contributes to the aesthetic of punk musicals (Laderman 2010). Dubbing actors with others' singing was more common in the Broadway adaptations that dominated Hollywood musical production in the 1960s and 70s, where films sought to retain the Broadway sound, familiar from best-selling cast recordings, but fitting it to Hollywood facial and bodily attractiveness (Siefert 1995: 57-9). As in these cases, dubbing by others
Introduction 21
in Hollywood cinema has mostly been concealed. Many now know that Rita Hayworth was dubbed by many different singers, some of whom were quite well known (Anita Ellis in Gilda, 1946, Down to Earth, 1947; Jeri Southern in Fire Down Below, 1957), but there was no recognition of this in the credits for the films or the attendant publicity. Likewise uncredited, Marni Nixon dubbed Deborah Kerr in The King and I (1956), Natalie Wood in West Side Story (1961) and Audrey Hepburn in My Fair Lady (1964).18 Mario Lanza, a very famous voice as well as image, quarrelled with MGM over the director assigned to The Student Prince (1954) and was visually but not vocally substituted by Edmund Purdom (though it has long been gossiped that it was really because Lanza was becoming too fat), but this is an unusual case in Hollywood, perhaps made possible by the almost highbrow status of the vehicle (an operetta first performed in 1924). The concealment of dubbing in most cases suggests the need to promote individualism as the great guiding value of civilisation, by affirming the survival of the individual's most intimate, distinctive and special mode of expression, the voice of the body in song.
Sometimes Hollywood felt uneasy about the match of voice to looks and/or acting ability. If the singing star wereperceived to lack screen presence, then they might well not have much of a movie career, the fate of massively popular (and not uncomely) recording stars such as Dinah Shore and Vic Damone. Judy Garland was, notoriously, regarded as unattractive, a theme often built cruelly into her films. Astonishingly, in the light not only of his subsequent career and lovers but also of his 1940s bobbysoxer following, Frank Sinatra too was not considered to be, in looks, up to the level of his voice, and in the early films has difficulties attracting and eventually getting the girl. Hollywood persisted with Garland and Sinatra, because, while the idea of the rightness of a certain look with a certain sound (beauty with beauty) was important, that of the truth of the voice and body belonging together was even more so.
Song is an art of time and space. The way film, also an art of time and space, handles song works in tandem with these dimensions of song, selectively, more and less consciously,19 especially when foregrounded in a film and not merely present ambiently, as background or local colour. In the first few decades of film there was much interest in the exploration of analogies between film and music, between the way shapes and images might be seen as being like sounds and melodies and especially the correspondence between visual and musical movement; this is evident in silent cinema (Germaine Dulac, Viking Eggeling), even contemporaneous painting (Wassily Kandinsky), as well as in the early sound period (the Oskar Fischinger films setting abstract moving shapes to pieces of classical music and jazz, the Eisenstein-Prokofiev collaborations).20 Some of the theoretical underpinnings of these may have been challenged and the avant-garde aesthetics found uncongenial in mainstream quarters, but, especially in the handling of song, something of the practices deriving from these perceptions remains in place.
Song is produced and heard over time - notes following one another and interconnecting - and at a point in time — notes sounding together (harmony, polyphony, counterpoint) (Davies 1994: 232). As noted above, song prioritises
22 Introduction
musical time over that of practical — speech, action, story — time. Musical time may be measured (as tends to be the case in Western music) or unmeasured (as in the drone of Indian music), regular or irregular (modernist classical music and jazz), but whatever it is, in song it takes over from the rhythms of speech and ordinary movement. In film this can mean the organisation of the image — and especially
(camera and performer movement and editing - in relation to the rhythms and contours of the music. The simplest and most evident instances are suggested by breaks in the music, cutting at the end of verse or phrase. The editing of the opening lines of'If My Friends Could See Me Now' in Sweet Charity (USA 1968) responds to its first separated notes/words by cutting after each one ('If/they/could') but then holds after the next cut (after 'could') in response to the exuberant take-off into melody proper ('/could see me now, that little gang of mine'). The editing of the whole number takes its cue from the jerkiness of the song's rhythms and repetitions (notably in a sequence in which each immediately succeeding 'They'd never believe me' cuts to Charity (Shirley MacLaine) in a new pose in a new part of the room with a new prop); in the process, time and space are often violated in a way that also suggests (in line with the film's procedures overall)21 the precariousness of Charity's happiness. In some of the Lena Home numbers discussed in Chapter 6, camera movement responds, sometimes to the meanings of words and gestures (as in 'Why Was I Born?'), sometimes to the shape and climax of the melody (as in Where or When'). The sweeping camera movements for the joined numbers 'Peethe gharaaj' ('My Foolish Heart Sings with the Breeze') and 'Matwale jiya' (Life Goes By as We Cut the Crop') in Mother India (discussed below) - notably as they create a visual rhythm from waving plants as well as registering the rhythmic actions of sowing, cutting and sifting — work with both the quite fast rhythm and the sweeping violin phrases. The opening sequence of Shaft, discussed in detail in Chapter 8, is a particularly clear instance of filmic (onscreen and camera movement, editing) and musical accord.
Songs also have an overall internal logic: they either come to an end, in the satisfaction of tonal resolution, or fade away, in exhaustion or the re-imposition of outside time (or recording length) constraints, suggesting that the music has not ended. Musicals are usually adept at moving into a song within a scene but very often at the end of a song there is either a pause or marking of some kind rather than simply a subsiding back into narrative time, or else the scene ends when the | song does, as if the musical closure is so final that there can be no immediate return to the narrative time out of which it emerged. At the end of the first 'It Only Happens When I Dance with You' number in Easter Parade (USA 1948), in which Don (Fred Astaire) seeks to persuade Nadine (Ann Miller), his dancing partner and lover, by singing and dancing with her not to leave him and the act, the pair subside into the pleasure of having danced so well together. The spell is broken by the arrival of their friend Jonathan (Peter Lawford) who says, 'I hope I'm not interrupting something', a standard line on an entrance but one that also signals here the abrupt return to spoken, undanced narrative.22 To take some other examples discussed in these pages: at the end of 'Cheek to Cheek' in Top Hat there is a
Introduction 23
longish pause while Dale and Jerry collect themselves from the deliciousness of the dance and return to the sober, and for Dale disquieting,23 reality of their situation. 'Prehistoric Man' in On the Town ends with the group in a held, end-of-number, to-camera formation (Figure 1.4). To get back from this to the narrative, Ozzie lets out an enthusiastic cry ('Yowlee!'), a non-verbal vocal sound picking up on the high spirits of the number (and an idea of a caveman yell) but also allowing for a transition to non-musical utterance. 'The Boy Next Door' and 'The Trolley Song' in Meet Me in St. Louis both end the sequence in which they occur (the former especially decisively by Esther letting a lace window curtain fall in a way that resembles a theatre curtain closing). In contrast, many songs in Mother India bleed back into the narrative, with for instance the arrival at the groom's home, Radha fainting, pregnant, at the end of the harvest and so on. Songs in blaxploitation movies (which are of course not performed by the characters, unless they occasionally join in as in Car Wash) are different again, since the music does not work to a point of closure. It often fades away (or in Car Wash segues into the next number); the music for the opening sequences of Shaft and Trouble Man, by musical logic, would just carry on and be 'always there', but in both cases it is brought to a halt with a chord signalling finality and the start of narrativity proper, while also implying that, but for this rude musical and narrative interruption, the music could have just gone on.
FIGURE 1.4 .Prehistoric Man': Gene Kelly, Frank Sinatra, Ann Miller, Jules Munshin Betty Garrett (On the Town, MGM 1949)
24 Introduction
Introduction 25
Because of the importance of repetition and redundancy in song (as noted above), it tends towards a sense of temporal stasis, of not going anywhere, a much weaker sense of movement than in more elaborated musical forms. This allows for that feeling of time being suspended, standing still, that can inform song sequences. Screen time - how long the song is. on screen - coincides with musical length but not necessarily with putative diegetic time (how long one can suppose it would take for an action to occur). Much the most common response to this is the long take, a whole number filmed in one take (e.g. 'The Man That Got Away' discussed in Chapter 3 or 'Can't Help Lovin' That Man' in Chapter 6) or by means of continuity editing that seeks to preserve natural space-time co-ordinates (e.g. 'The Boy Next Door', discussed in Chapter 2, or 'Cheek to Cheek' in Chapter 4), an affirmation of the unity of screen and diegetic time.
However, the potential mismatch between screen time and putative time is the basis for the possibility of conveying a sense of time expanding or contracting in the length of a song. In the course of a number a great deal of diegetic time may pass (as in the journey numbers in The Sound of Music and Funny G;V/,*or Radha's search for Shyamu to 'In Every Town and Country I Search for You' in Mother India, all discussed below), but it is musical time and length that determines screen time. It is harder to demonstrate examples of song time taking up more screen time than diegetic time: love songs, indeed most songs, to say nothing of arias, might be thought to take up longer than the basic information and emotion they are conveying, but while it may literally only take seconds to say 'I love you' or 'I am happy' or 'I am dejected', there is no readily identifiable putative time for how long it takes to feel it or to convey it. The fact that songs do often seem to be taking their time conveying what would be quite swiftly conveyed in non-singing narration does perhaps relate to the sense of time standing still or engulfing one that is characteristic (or thought or wished to be) of powerful or transcendent feelings.
In the case of the extended production numbers in, especially, the MGM musicals, presented as imagined or dreamt (e.g. The Pirate, 1948, An American in Paris, 1951), or proposed (Don's idea for 'Broadway Rhythm' for The Dancing Cavalier in Singin' in the Rain, of which, after we have seen what he has presumably verbally described, studio head R. F. remarks, 'I can't quite visualise it'), or supposedly taking place on stage ('Girl Hunt' in The Band Wagon 1953), the editing follows an oneiric logic of time (and of course space). In much music (and notably hip-hop) video, where there is no attempt or wish to create a sense of time (and space) beyond that of literal screen time, the editing comes closer to the non-time of the associative editing characteristic of Soviet montage cinema.24 In other cases, there can be a linking of images (including their internal rhythms and that of the cuts between them) in sympathy — or even sometimes wantonly at odds — with a song to evoke a different sensory experience. The realistically filmed (that is, continuity edited), individuated dancing to 'Night Fever' in Saturday Night'Fever gradually gives way to spontaneous formation dancing, discontinuous cuts, dissolves, overhead and very low shots and dry ice, conveying the dreamy, merging loss of
self on the disco floor. The 'Perfect Day' sequence in Trainspotting (UK 1996) (cf. Smith 2002: 67—70) sets Lou Reed's light, grainy voice and laid-back delivery, with a sostenuto string accompaniment becoming more epic for the chorus,25 against Renton (Ewan McGregor) injecting himself, sinking back in easefulness and appearing to sink blissfully into the floor, but then hauled out of the house, into a taxi, onto a hospital trolley, until, against Reed's still softly sung but ominous, repeated words, 'You're gonna reap just what you sow', he begins the violent, painful process of recovering from an overdose. The music-image is a juxtaposition of the dreaminess of the song, suggestive of the drug experience also hinted at in some of the visual imagery, with the grim reality of the overdose and treatment; but the song - in the insinuation of the voice, the irony of the epic chorus and the final repeated words — also itself provides a starker gloss on the perfect day experience, while the temporal contraction and odd spatial camera positions also suggest the dim mental registration of the rescue and treatment as part of the drifting drug experience itself.
Song takes up space as well as time. Music itself is a spatial as well as a temporal art, perceptually and also literally.
Musicologists in recent years have often been uneasy about the perception of music in spatial terms,26 and it may indeed be a culturally bound perception.27 However, it has been important, not least in relation to attempts to organise cinematic space analogously to music. The tempo of music is apprehended in terms of movement, a notion necessarily spatial as well as temporal. We speak of melodies going and voices being higher and lower, and we recognise (whether we have the vocabulary to describe it or not) that 'a pitch one octave above another is "the same" as it' (Davies 1994: 230); there are patterns of tension and release that 'resemble in their structure kinaesthetic aspects of movement as well as the teleological character of human action' (233). Western music especially (and a fortiori its song) is organised around the dualistic, implicitly spatial principle of figure, the tune, and ground, the accompaniment (and such variations as soloist and band, ■voice and chorus) (Van Leeuwen 1999: 21).28 Music may have arisen from, or at any rate be felt to be closely connected to, dance, and song may likewise be assumed to be either the origin of all music or its most basic mode,29 and dance and song are ineluctably produced by bodies, in space.
Song is physically produced somewhere (though certain forms of mechanical production and, even more, reception — such as Muzak and the iPod — make this less evident). This somewhere is a location in, and in relation to, space: the sound comes from somewhere, and commonly one that we look towards, and it fills space. It is also in a geographically and historically specific place. Recent work in the I geography of music (e.g. Connell and Gibson 2003; Leyshon, Matless and Revill 1998) notes the way that music is produced in very specific locations ('whether in the subways of New York, the rainforests of New Guinea or the clubs of Manchester', Connell and Gibson 2003: 11), themselves linked to (imagined or otherwise) regional and national aural landscapes, and then this locational rooting is used to categorise, characterise, understand and enjoy it, to 'place' it. These aural
26 Introduction
associations and modes of hearing (as we may speak of hearing Naples in Neapolitan song or the ghetto in rap) geographically anchor music when taken up for wider, above all, mass consumption, even while moving away from local embeddedness, crossing borders, going global. Songs are also performed and consumed in different kinds of spaces - bedrooms, pubs, concert halls, stadia - and different kinds of locations (e.g. the inner city, the concert, the community hall) (Connell and Gibson 2003: 15), which create different perceptual dynamics.
Sound also takes up space. A song may fill up a huge space, either by virtue of the singer's powers of projection or by amplification. Which of these it is alters the dynamics of the experience. Bessie Smith's ability to fill space without amplification, mentioned above, was an exuberant, commanding occupancy (all the more remarkable for being achieved by an African-American, and not young, woman). A hard rock singer, shouting and screaming into a microphone against the din of the dense, swelling roar of the band, is both satisfying and thrilling and perhaps also expressive of the travails or triumph of the voice in the urban environment (Tagg 1990: 111-12).
The sound of song may carry over a long distance. In film, this may be an important narrative device, as a person hears a song before they know where it is coming from or are sure who is singing it; it may also be exaggerated magically, as in the passed along 'Isn't It Romantic?' in Love Me Tonight (discussed below). In La Haine (France 1995), DJ Cutkiller (appearing as if one of the lads on the estate) mixes samples of two rap tracks (one American, 'Sound of Da Police' by KRS-One, the other French, 'Nique la police' by NTM) with the opening of Edith Piaf's 'Je ne regrette rien',30 all repeated over and over, the sound carrying out of his bedroom over the surrounding estate, the aerial camera floating backwards from his window, first pointing downwards, then ahead and out across the estate towards Paris, with the mashup still carrying on at the same sound level as the camera moves further away from its diegetic source and up into the sky.31 The music embodies a sense of hybridity: an actual African-American track, a French one influenced by African-American music, and a classic French chanson, mixed by a French DJ with an English language - but surely American sounding - name who is of Maghrebin origin, with the original chanson popular (as Vincendeau, 2005: 77, points out) among ultra-right paratroopers during the Algerian war. This musical hybridity corresponds with the hybridity of the milieu the film depicts. There is a sense of defiance, in the music's hybridity and in the context of the previous nights' events, 'a rallying cry for resistance against police brutality' (Higbee 2006: 82); the camera suggests the (not literal) carrying power of the music, although the dreaminess of the movement, the fact that at the end of this (very short) sequence the estate and the sky fill the screen with only a sense of the centre of Paris in the distance, may also suggests the limits of how far it, and the call to resistance, can really travel.
A song can also be sung quietly, just by one person to another, something that might not be heard by anyone else. Mechanical reproduction is able to capture this and make it audible to others, and crooning was one of the first vocal
Introduction 27
developments to exploit this possibility of intimacy at a distance, that simultaneously draws us nearer to the voice and signals that we are in fact no nearer than we were before to the body producing it, a fundamental dialectic of the mass medium and especially the star/celebrity system (cf. Booth 1981: 186; Cubitt 1991: 48). Films can convey this sense of the small space of a song. When Fred Astaire as Jerry croons 'Cheek to Cheek' to Ginger Rogers as Dale in Top Hat, they are on a crowded dance floor, yet his light voice and easy delivery suggest she is enveloped in the cocoon of his wooing; when Judy Garland sings one of her trademark sad songs ('Over the Rainbow' in The Wizard of Oz, 1939, 'But Not for Me' in Girl Crazy, 1943, 'Better Luck Next Time' in Easter Parade), even though her strong, sometimes strained, voice adds intensity (including loudness) to the performance, the mise-en-scene and close camera position all reinforce the sense of a private soliloquy.
The opening sequences of the early sound films of René Clair, Sous les toits de Paris (1931) and Le Million (1931), do something else again with song and space. In a very long take (guaranteeing the maintenance of natural duration), the camera moves across elaborately constructed sets of Parisian neighbourhoods towards, respectively, a street singer and a dance hall. In Sous les toits the camera finally cranes right down into the street, while in Le Million there is a cut, which creates a point-of-view shot of two men who have scrambled across the rooftops and are now looking through a skylight at the dance hall beneath. Sound perspective is carefully controlled so that we seem to be drawing nearer to the source of the sound: the song and dance music do not fill the air but rather clearly come from somewhere, a space we are invited to be enticed into, entering a privileged space of warm, communal musicality.
Songs also come out of people's mouths. The significance of this in terms of seeing singing and dubbing has already been discussed above. Coming out of the body involves breath and thus also posture, and then often gesture and stance. Singers may underline words or music or both with bodily movements: Shirley Bassey, for instance, is very word-orientated, whereas Judy Garland's body language seems more responsive to musical phrases and Dusty Springfield's distinctive gestures, while sometimes prompted by words or music, also developed into a baroque language working in counterpoint to the music and lyrics. Not moving, especially as the effort of singing tends towards a degree of movement, may also be expressive, suggesting tension, concentration, repose or other affects; the Lieder singer Matthias Goerne makes only the barest gestures and stands perfectly still, giving the impression of his whole being bound up in the intensity of the song.
The next step on from bodily movement is dance. Several of the essays here focus as much on dance as on singing. Dance - at any rate in most of the examples discussed here - takes offfrom song, from both the semantic and the musical material proposed by the song, but it also allows for a much greater occupancy of space and bodily expression. The discussion of the 'Prehistoric Man' number in On the Town (Chapter 5) particularly focuses on the way dance relates to the opening up of space and the implications of this. Chapter 4, on heterosexuality and dance,
28 Introduction
Introduction 29
focuses more on the space of the dancers themselves and especially the relations between male and female dancers. The interaction of the movement shapes and rhythms of dance with camera movement and editing, and how all this constructs sens.es of spaces and the occupancy of space, warrants a treatment in terms of analogy analogous to that between music and the cinematic image, but I have pursued it here only in passing.32
Numbers in film musicals can exploit all these temporal and spatial resources, of music as of film. This commonly involves a sense of the suspension - or perhaps it is elongation — of time, the time that it takes to perform the number, the time that it takes out of the conversations or actions from which it arises. It also commonly involves an expansion of space: Fred leads Ginger off the crowded-with-couples dance floor to a separate pavilion in Top Hat; in the Busby Berkeley Warner's musicals, the sets in the stage shows get ever more elaborate and the chorines ever more numerous, tilling up more and more space and seen from the side, overhead and with a moving camera.
Hello Dolly! (USA 1969) has several numbers that start off from a quite confined area and then gradually involve more space, people and movement: 'Sunday Clothes' moving from the restrictions of the poor clerks' dwelling to the whole town and townsfolk of Yonkers; 'Dancing' moving from the bijou but tiny space of a hat shop to a surrounding park and all who are in it; 'Elegance' exploiting the fact that the characters are too poor to pay for a carriage to get to the Harmonia Gardens by having them dance their way there through the streets.33 There is a nice little gag about all this towards the end. Cornelius declares his love to Irene in a park filled with passers-by in the song 'It Only Takes a Moment'; a drunk turns up and says - illogically, but he is drunk — that he arrived late and didn't hear 'what came after "It only"', at which point everyone in the park joins in with rest of the song. In the stage version, for reasons we don't need to get into, Cornelius delivers the song in court; the judge declares that he hasn't quite got what Cornelius said, at which point a member of the public steps in and says, 'He said' and then bursts into 'It Only Takes a Moment' with everyone else in the court at once joining in. This is a tiny expansion (though it is about filling an august space commonly associated with the words 'Silence in court!') that both makes a gag out of the bursting-into-song conventions of the musical and is touching enough to have Cornelius immediately acquitted.
There are, less commonly, instances where the temporal and spatial expansion is much greater - and far-fetched. One celebrated instance is the 'Isn't It Romantic?' number from Love Me Tonight (USA 1932), where the song is initiated by the tailor Maurice (Maurice Chevalier), singing it to a client, who takes it up as he leaves Maurice and goes outside, where a passer-by picks it up and sings it with the driver of the taxi he takes to the station (telescoped time) and then (a jump in time) carries on with it in a carriage full of soldiers, who themselves take it up even after (another jump in time) they have long left the train and are on manoeuvres, marching along until a fiddler marching alongside them breaks off and carries the tune (telescoped time) to a gypsy encampment, the sound of which is heard on her balcony (spatially
unlikely but close to natural time) by Maurice's beloved Princess Jeanette (Jeanette MacDonald), who reprises it fully (long takes and continuity editing fully restoring a sense of natural time).
In the 1960s and 70s, widely expanded numbers became a relatively common feature of the musical, notably 'Confidence' in The Sound of Music and 'Don't Rain on My Parade' in Funny Girl. Here a huge amount of time and terrain is covered in the time it takes to perform the number and yet Maria (Julie Andrews) and Fanny (Barbra Streisand) are clearly shown singing entirely according to the self-contained temporal organisation of the song: no matter where they are, how far they have travelled, how big a spatial and temporal jump there may be between one shot and another, they can be seen singing at exactly the spot they should be to remain within the logic of the song (which is also to say, the track on the album). The absurdity of this is sometimes emphasised by the evident impediments to singing with which the characters/performers are lumbered: Maria has to handle two heavy suitcases on her journey from the convent to the Von Trapp mansion, at one point getting entangled with them as she tries to get down from a coach, while Fanny is carrying a large bunch of flowers as well as pill-box bag and has to run encumbered with a long, sheath skirt and, at one point, to hail a taxi while at the same time singing. The presence of ambient sounds, recorded at a consistent, natural level — doors opening, footsteps — also contrasts with the impossible volume of the singing (to say nothing of the non-diegetic orchestra), most notably at the end of 'Don't Rain on My Parade', where the big build-up of Streisand's final belt out of the title phrase — with a run-up on the first syllable of 'parade' and huge orchestral forces to back it up, shown in a spectacular aerial shot - comes to an end leaving just the perspectivally more or less correct, and thus very quiet, sound of the chugging of the tug boat taking her to join lover Nicky on his liner bound for Europe.
Yet these are beloved numbers that the audiences who made these films huge box-office hits cannot have found silly. The numbers rightly assume that they can get away with their handling of time and space by virtue of the narrative drive behind them, Maria overcoming her qualms about going out into the world, Fanny determined to live for herself and her desires rather than for the demands of the show (she is leaving a tour of the Follies to join Nicky), with the number ending as she sails past the Statue of Liberty, no less. It has also to do with the star style of Julie Andrews and especially Barbra Streisand (whose subsequent films include comparable numbers such as 'When the Parade Passes By' in Hello Dolly!, 1969, and 'Let's Hear It for Me' in Funny Lady, 1975). Both are notable for the sense of determination, control and energy in their performances, not least in the spectacular quality of their singing (compare, among predecessors, the softer, somewhat more jazz related singing of Alice Faye and Rosemary Clooney, the technically assured but less beautiful belting of Ethel Merman, or Judy Garland and Doris Day, who are capable of both these modes but without the vocal pyrotechnics of Andrews or Streisand). Streisand's final delivery of the title phrase of the song is notable for a choppiness in the rhythm and a sense of the terrific
30 Introduction
Introduction 31
physical effort involved to produce that last, huge soaring 'parade', the effort as well as the soaring conveying the emotional exertion required to fill that space. Aspects of filming in both cases also reinforce the sense of the tightness of the song time-space, enabling the number to operate at something like the level of metaphor rather than literalness.
'Confidence' starts quietly, with Maria/Andrews contained by the archway to the abbey that she is leaving; as her confidence builds through her singing of confidence, the film draws back from Maria walking out into the much wider space in front of the abbey. Throughout, editing and mise-en-scene, as well as Andrews' performance, keep in play the tension between her anxiety and apprehensiveness and her growing confidence. There is a cut from a low angle (thus relatively aggrandising) of her with her back to the abbey, her confidence growing, to a very high angle long shot, that dwarfs her in the image, notably in contrast with, at the other end of the scope screen, a huge baroque fountain. As the number develops, she leaps and clicks her heels together as she moves confidently forward with the camera speedily tracking back from her. Yet when she reaches the Von Trapp mansion, she is once again daunted, and the number ends comically with her seen running-stumbling towards the entrance door -Andrews signalling bodily Maria's mock heroism even while delivering a characteristically high, pure, sustained note on 'me'. At the start of'Don't Rain on My Parade', the camera tracks back from Fanny/Streisand as she walks determinedly away from her fellow performers who have been trying to persuade her not to leave; there is a cut on her pushing at a door and entering a bigger space on 'I'll march my band out'; a series of cuts at breaks in the melody as she takes a taxi ('I'm gonna live and live now/Get what I want I know how/Right on the target and wham') get us nearer to her, from outside to inside the taxi, making her less evidently framed, more immediate; and the final aerial shot, sweeping down from high to near close-up and then back out and up again in one take, exultantly underlines her success in ensuring that 'Nobody, no nobody, is gonna rain on my parade'.
Songs in films take up literal but also temporal and sometimes metaphorical space. They impose musical time and length on spoken and acted narration. They allow voices to fill or carry over space and, in dance, permit - even incite - bodies to do the same.
This can be lovely, but it may also be disruptive, threatening, subversive. If a song may expand on the narrative moment it may also interrupt its drive. Songs may empower the singer.
What gives the song the power to empower? [. . .] It rewrites ontological rules through its connection to ritual. It rescripts conversational rules by substituting an alternative discursive register. And it recasts closural rules through its ability to impose an alternative ending.
(Dubrow 2008: 220)
Song (and hence the singer) can have authority because it 'builds a sense of communal identity' such that the singer 'acquires authority for speaking for and with those around her' (220).34 Thus it can feel 'transgressive' to interrupt a song.35 Mr. Smith cuts off Rose and Esther's final phrase of the title song of Meet Me in St. Louis in their parlour reprise of it, signalling his separation from the family togetherness embodied in the film in singing (see Chapter 2); the harvest song and Pakeezah's last dance, in Mother India and Pakeezah, discussed below, are brought to a halt before they are quite over by momentous events (Radha going into labour, Sahibjaan lacerating her feet).
Whether song will be found disruptive rather that joyous — whether its specialness may seem to interrupt or distract attention or take up space and time inappropriately - is bound up with who is singing what, where, why and to whom. The American musical is notable, for instance, for the space it has given to strong women singing. There is a tension in such films, between the time and space given, to varying degrees, to these figures and, often but not always, elements of confinement and punishment in the narratives and star imagery. Bessie Smith is magnificent in St. Louis Blues (1929), filling the space of the bar in which she sings, her feelings amplified by the clientele who act as a chorus, her song bridging time transitions, yet the song itself and the shards, of narrative also ally this with the timeless, unending fate of women to be besotted by wretched men.36 One of Ethel Merman's greatest hits, from the stage show Anything Goes (1934) - delivered with her peerless security, bang on the note - was a song in which she sings of there being only one person from whom she gets a kick, a person who obviously does not adore her. The story and dialogue of many of Judy Garland's films replay what was later stressed in coverage in her declining years, that she was not pretty enough, and Barbra Streisand's stunning vocal skill is allied with remorseless coverage of her controlling and difficult character.37 None of this closes down the splendour of Smith, Merman, Garland, Streisand and the other women singers of the American musical, but it creates a complexity of handling 'when the woman sings' (Smith 2005: 54-116).38
The articles in this collection address such issues building on two arguments I have made elsewhere. The first of these (Dyer 1977) is that musical numbers (and other aspects of entertainment) are Utopian in the sense of presenting ideal, joyous feelings, above all happiness, but also broader, less specifiable affects. The second (Dyer 1995) explores what is only touched on there: Utopia for whom?39 This is not just a question of most of the films (before the final chapters) being made by and featuring almost exclusively white people (with white women largely figuring only as performers), but also of the histories that have made such Utopian expressivity possible and the feeling forms that this takes. The notion of expansion here is crucial, since it is both glorious and yet also involves claiming space and time, a way of relating to the world that takes the claims of personal and community expression as an absolute right, a feeling to which a person has an unquestionable, costless right, and which takes no heed of which persons are allowed to expand and which not, nor whose space and time (including nature's) can readily be encroached upon.
32 Introduction
Musicals are not 'about' these matters (and in any case, one should always try to avoid thinking in terms of works of art and entertainment being 'about' something) but they do derive from them. The eponymous Harvey Girls (USA 1946) arriving on the Atchison, Topeka and Santa Fe railroad get down from the train and sing of the delight in their journey, starting out close to the train, moving out to the platform and eventually encompassing the whole town and its folk. The imperialist implications of this joyous expansion, implications ineluctably present because of the history of the creation of the United States and, not least, westward expansion, remain so implicit as to be negligible (in the way that Westerns commonly forget the imperialism that made the genre possible). However, they are present, and as much in the pleasure of this spatially relatively modest though temporally considerably extended (and superbly realised) sense of expansion as in the mise-en-scene of the West. Racial co-ordinates are even more buried in Meet Me in St. Louis (as discussed in Chapter 2), where the role of African-Americans in producing the wealth, comfort and, to a degree, entertainment forms that made a Smith household possible is registered only in the most unconscious," subliminal shards of reference. The utopia-for-whom issue has to be unpicked from the 'Atchison, Topeka and the Santa Fe' number as it does from the whole confection of Meet Me in St. Louis, in ways that probably seem to violate the elan and fun of the former and the warmth and richness of the latter, but equivalent concerns are, I suggest in Chapter 5, a little nearer the surface of the 'Prehistoric Man' number in On the Town.
That piece and the one on Lena Home are pivotal to this collection (much of which was not specifically written for it). The former exemplifies many of the Utopian affects of the musical: the abundance of the space available to dance in, one filled with splendid artefacts, the energy of the music, dancing and improvisatory skill, the instant community of desire established between the five participants, the unalloyed transparency of libidinal enjoyment, and, exhilaratingly, a woman at the centre of it all. A white woman. The case of Home suggests that, those joys were harder to imagine with a black woman at their centre. Thus, whereas the sense of expansion is one of the defining glories of the bodily movement and cinematic construction of space in the classic (and white) musical, it is the sense' of containment that dominates Hollywood's handling of Lena Home. Within that narrow space, something could still be achieved, and, as I try to show, not only in terms of the record of Home's performances. Car Wash suggests one move beyond this, a continuous improvisatory relationship to music and space within a limited but not narrowly confined space and an interweaving of improvisations to create a kind of choral community feeling, all set against the awareness of the limitations of the work and the temporal boundedness, the not going anywhere, of the characters' lives. The moving-through-the-street-to-funk sequences in blaxploitation movies (discussed in the last chapter) suggest more: an ease with space (albeit one that some films see as problematic), a pleasure in a right to occupy space that is not about claiming other people's spaces but rather fully owning one's own. Provided you're a man.
Introduction 33
As discussed briefly towards the end of the chapter on Car Wash, pleasures and affects are seldom intrinsically, ineluctably male or female, black or white, straight or gay, or whatever. Human beings share fundamental qualities. If you prick us, do we not bleed? If you tickle us, do we not laugh? Yet such common humanity never exists separately from differences, culturally and historically constructed but profoundly lived, and characterised by differences in power as well as culture. All that necessarily inflects the senses of musical space and time discussed in this book.
The essays that follow deal with American cinema and, as this introduction so far makes evident, the musical is a key reference point. But the American musical is not the musical (cf. Martin 2001).t° The Hollywood instance should not be taken a priori as providing a paradigm for film-making in the rest of the world or in other periods (than the 1930s-1970s American musical); although nor should one assume that other film industries and filmmakers.either did not discover the same procedures independently or were not influenced by them. Although I have not given them systematic study, I think that much of what has been said above about Hollywood examples holds true to a considerable extent for, for instance, British musicals, German and Austrian operetta films, Finnish tango films and the use of fado in Portuguese cinema.41 However, I want to end by looking at two cinematic traditions that do handle song in many ways different from the way Hollywood does - Italy and India. Both differ from the Hollywood instance by virtue of their relation both to seeing singing and dubbing and to the functions song can have narrationally. This in turn makes possible different (but not always utterly different) senses of time and space.
No musicals: Italian cinema
Italians commonly claim that they don't do musicals. This is not because there are no Italian films with prominent songs, for there is a profusion of these. There is, though, an extreme fluidity in the way song figures in Italian cinema, a much more casual, pragmatic approach to the relation between voice and body, a much less clear sense of generic belonging or confinement and less inclination to focus on | the sight of singing, than in Hollywood cinema. This holds true even for the one film commonly held to come closest to the Hollywood model, Carosello napoletano (1954).
Italian cinema is not preoccupied with vocal belonging. Post-synchronisation of voice and body was ubiquitous in Italian cinema until very recently (and virtually all foreign films are dubbed).42 Actors dubbed themselves speaking and were also sometimes dubbed by others and the same was true of singing. In opera films -for several decades among the top box office successes in Italy - there was remarkable inconsistency.43 There were straightforward cases of a non-singer actor being dubbed by a singer (Sophia Loren by Renata Tebaldi in Aida, 1953, and Jacqueline Plessis by Tebaldi in Lohengrin, 1948; all the Japanese cast of Madama Butterfly, 1954, dubbed), though they might be dubbed by different singers in different films (Loren, while still under the name of Sofia Lazzari, by Palmira Vitali
34 Introduction
Marini in La Favorita, 1952, Franca Tamantini by Miriam Di Giove in the latter but by Vivien Caveglia in La Sonnambula in the same year). An established singer might be seen and dub themselves in some films, dub other people in other films and be seen but dubbed by someone else in other films (Tito Gobbi appeared and sang in Rigoletto, 1947, as he had in many non-opera films, but dubbed for Aldo Silvani in Rigoletto e le sue tragedie, 1954; 'Gino Sinimberghi sang for himself in L'elisir d'amove, 1946, and La Favorita, 1952, but was dubbed by Galliano Masini for La forza del destino, 1949, by Antonio Salvarezza for II trovatore, 1949, and by Licinio Francardi for La Sonnambula, 1952, while he himself dubbed Filippo Morucci in /pagliacci (Amore tragico), 1948). Nelly Corrado seems not to have sung for herself in any of her films (La signora dalle camelie, 1947,44 La leggenda di Faust, 1948, La forza del destino). There may have been contractual and timing issues in relation to some of these decisions, but the point is that it was felt not to matter, to be a routine practice. When the 1946 II barbiere di Siviglia used the voices of all its seen performers, some critics thought it made it seem like operetta and certainly not like cinema: Antonio Pietrangeli complained it was tantamount 'to using the cinematographic medium like any other means of mechanical reproduction',45 suggesting that a right fit between face and voice was more important than an ontologically secure one. Only later, when opera films became much rarer, did these habits disappear.
If there is fluidity in matching face and voice, so there is in terms of where and how songs occur in Italian cinema. There are many types of film based around song.46 The first Italian sound film, La canzone dell'amore (1930), was also the first in a long line of films based around (and in part selling) a current popular song; in the late 1950s this became what was subsequently dubbed the musicarello, based around the songs of one or more current pop stars (post-synchronising themselves). There are films based on revue, operetta and stage musicals. Opera, in addition to films directly based on specific works, provided plots, settings, performance styles and performers from the beginning of Italian cinema, and well into the 1960s continued to provide not only these but source music, biopic material and, as it were, occasions for films: sometimes much of the recitative was discarded, making the structure closer to that of a musical; Gli amori di Manon Lescaut (1954) has no singing in it but uses the music from Puccini's opera for its non-diegetic score; Ridi pagliaccio! (1941) tells the background story, about an opera singer, that was eventually made into the opera I pagliacci, with part of which the film ends.
The films do not always pause for a song or aria and focus their attention on it. When the characters start to sing in Rigoletto e le sue tragedie, which largely dispenses with recitative, they turn from the camera, as if not to show the sudden change in vocal delivery. In Quanto sei bella Roma (Rome How Beautiful You Are, 1959), a vehicle for the very popular light bel canto tenor Claudio Villa (perhaps the last of his kind to have this level of mass success), there are several scenes in which he starts to sing, sometimes having been begged to do so by the characters, only for the film to cut to other characters, sometimes even speaking over his singing. The credits and publicity for Napoli eterna canzone (Naples: Eternal Song, 1949) announce
I
Introduction 35
songs sung by the tenor Gianni Avolanti, a tenor of the San Carlo opera house in Naples. He is seen twice, briefly, so much to the right of the frame as not to be entirely within it, in a segment in which the heroine, Maria, listens to a radio broadcast of the title song and we see it being performed. The place of the performance is unclear: Avolanti and the orchestra are in front of a cyclorama of clouds, which might suggest a stage performance (which would fit with the narrative: Maria is listening to a song composed and conducted by the man from whom she is estranged, Roberto), but the conductor, Roberto, facing the orchestra is, in reverse shot, also standing before the cloud cyclorama, which at most suggests a (rather fancy) recording studio, but even more a non-space of both the airwaves and longing. In the course of the song, scenes from Maria and Roberto's relationship are superimposed over the performers, complicating the time scheme as well as further taking our attention off Avolanti. Elsewhere in the film, Avolanti is mosdy only heard non-diegetically over scenes of lovers (including Maria and Roberto, but also unnamed couples) in suitably romantic settings: the seashore, the park. In one scene in a restaurant, Avolanti is heard, unseen but presumably diegetic (live, broadcast or recorded), singing a particularly turbulent, strongly delivered song, but the characters ignore him and the sound of their speech is louder than Avolanti's voice, in effect the development of the narrative competing cacophonously for our attention with the attraction of the singing.
Such casualness relates to an assumption of the pervasiveness of song. This may derive in part from the model of through-sung opera and also that of the Neapolitan sceneggiata, in which music accompanied short acted dramas that took their cue from a song that was then sung as part of the performance (the term means a song that has been turned into a scene) (cf. Marlow-Mann 2011). Much Italian film melodrama, especially when set in or around Naples, makes extensive use of song as part of the dramatic development, such that the two genres, musicals and melodrama, seen in Hollywood as distinct and yet both defined by their relation to music, become near indistinguishable in Italian cinema.47 It is no accident that Naples should especially be the locus of this work, in part because of the rich tradition of Neapolitan song - occupying in the Italian national musical repertoire something like the place German Lieder, French chanson and Broadway show tunes do in theirs - and in part because of the image of Naples as a city in which there is always singing.48 It underpins Carosello napoletano,49 considered by Simone Arcagni (2006: 98), in a near universally agreed judgement, to be 'the only true Italian musical in a position to compete with the US model' (Tunico vero musicale italiano in grado di competere con il modello statunitense').50
The fluidity of voice-body and song-narrative arrangements is matched by fluidity in the handling of space and time in Carosello napoletano. It consists of a series of scenes, sketches (tantamount to sceneggiate), performances and ballets, all built around Neapolitan song. Neapolitan place is secured not only - though above all - by the songs (nearly all the most beloved titles are present) but also by a well established iconography of Naples (the seafront, the gardens high above the bay, Caffe Gambrinus, the tightly packed dwellings, alleyways and piazzas), the archetypal
36 Introduction
Introduction 37
Neapolitan name of the intermittent narrator figure Salvatore Esposito and the casting of famously Neapolitan performers: Vittorio Caprioli, Beniamino Gigli,5' Sophia Loren, Clelia Matania, Dolores Palumbo and Giacomo Rondinella. Yet the'casting is not in this respect-uniform: Esposito is played by the well-known but not Neapolitan actor Pablo Stogpa, the ballets are performed by the Grand Ballet du Marquis De Cuevas, including their choreographer Leonide Massine, and many of the other well-known performers are not Neapolitans. Neapolitan song is both insistently regional and an Italian national treasure, permitting this spatial slippage at the level of casting as well as song and mise-en-scene.
There is also a rough chronology, moments in Naples' history, starting in the present (a post-war shanty town) then moving back to the seventeenth century and thence forwards through the Turkish invasion, the emergence of the figure of Pulcinella (especially in his nineteenth century incarnation, Antonio Petito), the inauguration of the Vesuvius funicular, emigration to the USA, modernity (photography, electricity, telephone), the First and Second World Wars and back to the present, each stage featuring songs more or less composed at the time (or claimed to have been). Yet what is emphasised as much is the sense of continuity, of plus ca change, plus c'est la meme chose. This is achieved in part by the (rightly) assumed familiarity of songs long past the era of their first appearance and by the recurrent presence of Salvatore Esposito (who, when the film has reached 1900, says to camera: 'I'm presenting [sto illustrando] my city - probably you saw my face and heard my voice in earlier centuries, because as long as Naples has existed, so have F, and who often appears as other characters in different scenelets). When Pulcinella dies on stage and it turns out the actor Petito has really died, Esposito's voice tells us that Pulcinella, earlier described as a figure in whom 'for centuries Neapolitans laughed at their own faults and misadventures', of course never died for he had already passed to 'the stage of the streets of Naples'. The sequence of the invasion of the Turks is followed by a kind of riff on the idea of invasion, including that of tourism, so that the unique historical event becomes seen merely as one of recurrence, embodied in a mimed and danced number in which male 'invaders' exchange gifts with pretty female Neapolitans. Each epoch, though it indicates specific historical events and sometimes merely showcases the songs as part of the time, also has one or two key songs that sing of and also form part of a sad love story (e.g. the fisherman's beloved who, raped by the Turks, kills herself in disgrace; a girl who buys a love potion from a quack (Stoppa/Esposito) to make a boy fall in love with her on waking, only for him to wake up and see someone else first; the story of Sisina and Luigi, separated by her mother when she becomes a successful singing star and then, after their reconciliation, by his death in the First World War): the songs, which themselves provide the keynote to each sequence, suggest that love and its sorrows are the same and are the enduring heart of Neapolitan song and experience.
Song and narrative are inseparable because, as Esposito says in his first appearance, playing a barrel organ (as he does recurrently throughout the film), the songs are 'the songs of our life'. Yet although this may be in some sense purveyed as a truth,
the perception of the interpenetration of the artifice of song and the reality of life allows for a continuous ontological fluidity. The film opens with a panning shot across a Neapolitan shanty town and ruins, in which the sound of a woman singing mingles with everyday shouts and. exchanges and a church bell. Bar the fact that it is in colour, this could be the opening of a neo-realist film: the iconography of ruins and poverty, the establishment of an area from which particular but representative stories and characters will be taken, location shooting, post-synchronised sound made to appear to be emanating from the location (here by sounding distant from the camera). The singing voice is part of the realism, a statement that there is always singing in Naples, especially among the common people. The transition to sets is gradual: the wind carries sheet music into the sea, leading to the (seventeenth century) story of the fisherman and his beloved, initially still shot on location but much posed, the characters in period clothes and with a much less naturalistic sound level for the singing. It is very reminiscent of the Neapolitan film melodramas mentioned above, where songs are part of the means by which the loves, deceits and tragedies of the story are told. The news of the invasion of the Turks occurs first without music and then with dramatic non-diegetic music. The young woman runs into caves and her confrontation with the Turks involves stylised gestures that morph into dance on the flat floor of what is now obviously a set. We move, in other words, with a fair degree of seamlessness, from almost pure neo-realist naturalism in the opening shot, via semi-naturalism (period costumes, location shooting, onscreen occasions for song, off-screen singing), just about plausible visible singing (the young woman sings out to the fishermen), classical cinema (continuity editing, a mix of location and verisimilitudinous sets, no music and then non-diegetic music), to the stylisation of dance on a dance floor. The transition is also one from the (albeit factitious) authenticity of Neapolitan culture and song to a shrieking post-Rite of Spring orchestra and the less-than-Turkish Ballet Africain di Keita Fodeba, shimmying, yelping and screeching, some playing African drums. ">
The Turks routed, the film returns to cinematically classical, melodramatic norms for the fishermen returning and the young woman, disgraced by the Turks, killing herself rather than face the young man. At which point, a dissolve to a painting of the young man on the side of the barrel organ returns us to an obviously painted, theatre-style set and to Esposito announcing that 'this is how the story ended' and going on to make general points about invasions, looking pointedly, as do the children gathered round, at a black man (presumably a GI) passing by. This in turn leads to the number riffing on invasion.
The transition via a painting, and one specifically allied to music (in this case on the side of Esposito's barrel organ), is characteristic of the way Carosello napoletano slips continuously between modes of representation. The Pulcinella sequence grows out of the invaders sequence, when a white GI picks up and asks about a mask lying on a stall and is told it is a carnival mask of Pulcinella. A close-up of the mask cuts to a puppet show featuring Pulcinella and Colombina. As the camera tracks back, disclosing a seaside piazza, people dressed in carnival costumes
38 Introduction
come on and a caller invites people into the Teatro San Carlino.52 The puppet Pulcinella makes fun of the troupe, and the puppet-master is revealed to be Salvatore Esposito. After the latter is carted off by the police for his cheek, the puppet Pulcinella comes alive and sneaks into the San Carlino. Teased and teasing backstage, he is knocked on the head and falls onto a swing, at which point there is a cut to him in a ballet that starts out as Adam and Eve and the expulsion from Eden but then merges into an escapades-of-Pulcinella scenario. This is ballet clearly danced on sets and a dance floor, but, even if in fact transposed from the successful stage original of Carosello napoktano,53 it is nonetheless conceptualised cinematically in terms of editing and transitions much in the style of the Red Shoes ballet (1948), itself ambiguously on stage and choreographed for the camera (and by Massine). At a certain point, a ballerina with a veil over her head comes in on pointe; Pulcinella dances with her briefly before throwing back her veil, to reveal a ghastly face that he recognises as Death; he collapses, but a voice urges him to carry on - it is the voice of the prompter. The film cuts back to a shot from the back of the auditorium and people laughing at what they take to be all part of the act — but 'Pulcinella', that is Antonio Petito, has really collapsed. The sequence thus moves from a sign of theatre and carnival (the mask) to a puppet show to an ontologically unclassifiable figure (a puppet come to life as the character Pulcinella danced by a famous dancer) who plays a part (Adam) in a cinematically presented ballet that is then shown to take place on a theatre stage in which coincide the death of the character and the death of the performer (though not of course the actual performer, Massine). Finally, dying, Pulcinella/Petito/Massine utters 'La maschera! La maschera!', returning us simultaneously to the sign of theatre (the mask) and Esposito's words quoted above to the effect that Pulcinella is still alive in the real streets of Naples.
The Pulcinella sequence starts then in (studio created, heightened) realism and moves through different types of representation (iconic sign, puppets, characters, ballet, cinema) that are not always distinguishable. Yet more fluid and prolific shifts occur in the World War One sequence that has the story of Sisina (Sophia Loren) and Luigi (Giacomo Rondinella) at its heart. It starts from something more or less offering itself as cinematic realism. The camera roams over emigrants in the hold of a ship; an officer closes window shutters in front of the camera, which tracks back to disclose the window of a customs office with people standing and sitting in front of it. A sad song accompanies all this but when it finishes, a flash and a man's voice reveal that they are people posing for a studio photo of emigrants and the backdrop of the custom's house is hoisted up, not revealing the scene that we have just seen and took to be behind the backdrop but rather another photography studio backdrop, itself clearly hanging before others. Other parts of the studio are revealed, including three young men trying to write a song. Sisina and her mother arrive. Sisina poses for a photograph, Luigi walks towards her, the lights dim and an oval spotlight encircles her, they gaze at one another (ex-lovers still in love), the camera tracks slowly in on her and, when he is no longer in view, Rondinella's
Introduction 39
voice comes in singing 'Reginella' (a song about ex-lovers still in love).54 Cut to Luigi/Rondinella singing, looking in the same direction as in the previous shot. Cut back to Sisina as before, but the camera tracks back to reveal it is her image, based on the photograph for which she was posing, on a poster outside the Caffe Gambrinus, and it is this poster that Luigi/Rondinella is now serenading (Figures 1.5, 1.6, 1.7).
There is a brief dance number of people arriving and entering the cafe. Luigi sings a jolly song with his co-writers from earlier, but the mood changes again when Sisina and her mother arrive. Ragtime blares out from a phonograph in the street and people dance. A man looks at a poster for Lily Kangy; the poster tears open to reveal a singer, Vera Nandi, singing 'Lily Kangy' and doing 'la mossa' ('the movement', namely shaking her breasts): in other words, an interpolated sequence having nothing to do with the ongoing narrative but chronologically and spatially accurate enough.55 Backstage, Sisina is crying, looking at a photograph of her and Luigi, with 'L'addio' ('The farewell')56 written across its frame. She goes onstage and does a light-hearted duet. As soon as it is finished, we cut back to the street outside, where it is raining, and Luigi/Rondinella is heard singing 'Quanto chiove' ('How it rains'),57 one of the most heart-rending of Neapolitan songs; as Sisina, her mother and a rich admirer drive off in a carriage, Luigi/Rondinella is revealed, singing in the rain, until he is framed, making the image look like a postcard featuring the song he is singing (a once popular Neapolitan postcard genre) (Figure 1.8).
Cut to another song in Gambrinus, with singers poked through cut-out holes in a mock postcard. Cut to a location shot in the public gardens high above the Bay of Naples, with Luigi/Rondinella singing a disconsolate song ('Catari Core 'ngrato' ('Ungrateful heart'),58 to which a group of girls (one of whom has a postcard of him in her book) provide the chorus. Towards the end they and he leave the garden location and, still singing the same song, enter the studio set of the exterior of Gambrinus. Luigi sits with his fellow composers and they start trying out a song again, but an army recruitment poster summons them to join up. At the train station, Luigi and Sisina, first separately, then together, sing a sad duet (Loren, here as elsewhere, dubbed). There is a short tracking shot of the three friends/composers at the front: Luigi freezes and is held in an oval spotlight, a stylised representation of his death. Cut to Gambrinus and Sisina performing a cheery number that turns into a cancan routine (another interpolated element, performed by 'II French Can-Can di Miss Joan Baron'). Backstage Luigi's comrades arrive and Sisina realises that Luigi is dead. There is an extreme close-up of Sisina — but here also surely we should say of Sophia — her eyes glistening, with a heavenly choir over (in other words, a pure movie moment (Figure 1.9)). She is called back on stage with the other girls, but as she takes a bow, the heavenly choir music continues over, her mood replacing musically that of the stage show. Then as the music fades a voice calls out 'Hold it!' and we are back in the photographer's studio and this is a pose for a patriotic photo.
FIGURE 1.7
FIGURE 1.6
FIGURE 1.8 'Quanta chiove': Luigi (Giacomo Rondinella) (Carosello napoletano Lux 1954) ■ '"
42 Introduction
FIGURE 1.9 Sisina/Sophia (Carosello napoletano, Lux 1954)
The WWI/Sisma-Luigi sequence runs the gamut from location shooting to movie photography, via any number of modes of photographic and non-photographic representations:
location shooting (the public gardens) with post-synchronised, non-naturalistic59 singing;
film sets, with varying degrees of evident artifice, and both acted and sung and danced performance (the emigrants, in front of and backstage at Gambrinus, the train station, the front); backdrops in a photographic studio; photographic poses;
posters (Sisina, Lily Kangy, war recruitment);
spotlighting (diegetically motivated (Sisina posing) and not (Luigi shot)); on stage performance (in Gambrinus); postcard representation (of a song);
the movies (Sophia in close-up, subjective non-diegetic music taking precedence over diegetic stage music).
Introduction 43
The last transition in the film begins with one of the most celebrated items in Neapolitan tradition, a presepio. These are huge Christmas cribs, commonly (as in Carosello napoletano) constructed as on a rock-like promontory, with multiple dwellings and artisanal workplaces centred on the family in the manger and attendant shepherds and Magi. In the film Esposito and his son visit one, and when the latter touches a model musician, the film cuts to the musician as a living person and the camera tracks back to show people frozen in the positions just seen on the presepio, who come to life, the camera gradually revealing not the rocky structures of the presepio but the steps, balcony and multi-levels of a Naples courtyard - the continuity between the Neapolitan imagination and its reality. This develops into a tarantella for the whole neighbourhood, a big production carnival finale.
It is song, enhanced, heightened vocal delivery pitched between the real and the theatrical and seen as constitutive of Neapolitan life, that permits the fluidity of Carosello napoletano: song in 'Naples' is everywhere, exceeding the historical moment of its production, integrating speech and musicality in everyday life. While the theatricality of the film, what Simone Arcagni (2006: 104) calls 'the filmic creation of a theatrical world', makes possible this non-realist presentation, it does so on the basis of a cherished idea of Naples, clung to even if not necessarily literally believed in. The film and the idea condense a wider notion of Italy and song, which in less elaborated form underpins the characteristic sense in Italian cinema of song's ubiquity.
All musicals: Hindi cinema
Hindi cinema seems to offer still another different approach to the use of song and dance in film. I say 'seems' because I am well aware of my lack of expertise and the comments that follow are based on a tiny handful of canonical films from the largest industry in film history. The observations I make on this basis indicate that Hindi movies can be significantly different from Hollywood and European movies in their deployment of song and dance. This does not mean always and utterly different. In particular, it is well to remember that songs in many Hindi movies can also function as, for instance, escape or interlude, as they do in Western cinema, or as 'often . . . little more than digressions gratuitously inserted into the plot' (Manuel 1988: 175). Yet the commonest term for what in the Hollywood context would be called musical numbers is song picturisation, suggesting, whatever the narrative occasion, the primacy of the song, the way it sets the agenda for the visuals (Booth 2000: 143; Majumdar 2009: 180).
One index of the difference is the use of dubbing. In Hollywood cinema this is largely deceitful, in Italian casual or fluid; in Hindi it is transformative. The voice we hear is not only not that of the person we see singing but is also usually an extremely well-known voice. Westerners unfamiliar with Hindi cinema may need to imagine Brad Pitt making a film in which his character sings and the voice you hear is Bruce Springsteen's, or — given the long career of many Hindi movie singers - Doris Day providing the unmistakeable voice for Lana Turner, Diane Keaton,
44 Introduction
Introduction 45
Julia Roberts and Angelina Jolie. Lata Mangeshkar alone, in a career spanning from 1942 to almost the present, sang for Nargis in Mother India (1957) and Meena Kumari in Pakee.zdh (1972), the two films discussed at greater length below, as well as for two of the biggest hits of Hindi cinema in the last decades, Dilwale Dulhania he Jayenge (1995) and Lagaan — Once Upon a Time in India (2001), and along the way for, among many (some would say practically all) others, Zeenat Aman, Geeta Bali, Dimple, Kajol, Nanda, Nimmi, Nutan and Sadhana.
One of Lata's earliest hits was 'La-ra-lappa la-ra-lappa lai rakhada',60 from Ek Thi Ladki (There Was a Girl, 1949); however, it was the star seen singing it, Meena Shorey, who became known thereafter as the 'La-ra-lappa Girl' throughout her life. There maybe many explanations for this, not least Lata's extraordinarily long and productive career,61 but it is suggestive of the fusion of the embodied star/character and the disembodied voice, that in some sense the voice in a memorable song in a given film can still be felt to belong to the seen star even while it is known to belong to the unseen one. Similarly one can buy a CD such as Hits of Hema Malini, even though everyone knows that Malini is not the singer (Majumdar 2009: 186).
The voice need not belong to one actor/character in a film. In Mother India, Mohammed Rafi sings for both Shyamu (in duet with Radha/Lata) and his son Birju; while this might suggest male familial continuity, it is also the case that Birju is the second son and less like Shyamu than the first son, Ramu. Lata sings nearly all of Radha's songs but also Champa's flirtatious song to Ramu, perhaps here indeed suggesting the continuity of the role of fiancee/wife/mother from one generation to another rather than the specificity of Radha or Champa. We may not see singing but may make a connection with the person seen on screen: Manna Dey (not Rafi) sings the song that begins 'Life goes by as we cut the crop', which accompanies a sequence centred on Shyamu, covering working the fields, selling the crop to the swindler Sukhilala, the growth of Shyamu's family and he and Radha gradually ground down by hardship; the sequence is principally, though by no means exclusively, about Shyamu, one might say that this is his number in the film, but at no point do we see him singing. This permits a kind of abstraction, whereby he is as much the Husband and the Father as he is specifically Shyamu.
The Bombay film industry did not set out with such procedures in mind. The early films used singer stars and even in the 1940s, when playback singing had become common, there was still a conception of the right particular voice for the right particular onscreen star. It is in the late 1940s that the practice takes hold of the singer as a star in his or her own right, able to provide the voice for many onscreen stars, often less famous than them (Sundar 2008: 146—7; Majumdar 2009: 179-87). There is an economic dimension to the importance of playback singers (a term that suggests an inappropriate secondariness). Hindi film music is — or, for most of the last century since the mid-1930s, was — Indian pop music; although widely available on records and radio, going to hear the tracks, and the voices, in the context of attractive bodies and interesting stories was a major reason for going to the movies. But there are further resonances, beyond pulchritude and economics.
As discussed above, in Western movies, in hearing the voice that comes out of the body that we see, a special sense of the individual and revelatory intimacy of the voice is suggested, whether in terms of the body or the inner person. Something else is at play when the voice we hear not only does not belong to the body we see, but belongs to a familiar voice, so that we can always know what Hollywood habits of dubbing insistently sought to conceal: that we are not seeing that body's voice. By having a voice other than that of the actor, musical numbers shift the film onto a different ontological plane.62
This can be augmented by formal qualities of the songs. The language is often more elevated, more poetic, often "drawing on Urdu and Persian as opposed to Hindi, and thus contrasting with the spoken speech before and after it (and without the value placed on everyday language in Western popular and pop song). Despite the heterogeneous character of Indian film music (Arnold 1988), there is a relation, albeit often strained and now diminishing, to classical Indian music (Lata trained in classical singing and continues to'practice it; cf. Sundar 2008: 148; Srivastava 2004: 2020): as noted above, the drone establishes an unmeasured time that may suggest the eternal, not least because it is something that cannot be produced by the human voice and the aesthetic structure of the central form, the rag, tends towards stasis (Morcom 2007: 159), of suspended time. There is a fondness for the particularly high pitch of the women singers, in contrast to the range of timbres in female singing in Indian music as a whole (Sundar 2008: 174). Gangadhar Gadgil wrote symptomatically of Lata:
[hers] is a voice that is ageless, pure, vibrantly alive, untrammelled in its range and flexibility, hauntingly expressive and enchanting in its sweetness. Above all, it has a certain ethereal quality, an indefinable something, with a unique appeal for us Indians.63
The pitch of the voice is perceived as unsullied, untouched not only by sexuality but by earthly (that is, spatial and temporal) concerns. Lata's voice has not changed in the course of her career, suggestive of a realm of unchanging womanhood, unchanging at the level of the soul (the voice) but not at the level of the body (the onscreen stars) (Majumdar 2009: 198); moreover, her public image is one of austerity and virginity.64 By having a voice not rooted in a specific human body,
| the voice can be taken as disembodied, allowing it association with the eternal.
1 Even when the ostensible feeling of the numbers in Hindi movies is straightforward enough - love, hate, happiness, loss, desperation — casting them in this decorporealised way lifts them onto a higher affective plane.
The use especially of women's voices in this connection, and a fortiori Lata's, has been much debated. Drawing on a formulation of Partha Chatterjee (1993), Nilanjana Bhattacharjya (2009: 54) relates song to the 'inner' domain of Indian society, which is 'entrusted with preserving an "essential" cultural identity - those traditions related to religion and mythology, literature, dance, music, and art', as opposed to the realm of business, economics and politics (something especially
46 Introduction
significant in the context of the diasporic films that are the focus of Bhattacharjya's article). Women's place in the inner domain is complex: they can be seen as its guardians and embodiment, in their role in cultural nurturing as well as their ready association with nature and timelessness, but in an era of ideas about female emancipation and, not least, of one of the world's first elected female national leaders (Indira Ghandi, in power 1966—77, 1980-84), women's identification with this realm is not secure.
Sanjay Srivastava (2004) argues that the coming to prominence of Lata's voice — and the kind of voice of which she is emblematic — was linked to regressive policies in the 1950s that sought to promote a reactionary (and Hindu-ised) timeless representation of woman: 'at the same time that women's bodies became visible in public spaces via films, their presence was "thinned" through the expressive timbre granted them' (Srivastava 2004: 2021). This view is disputed by Ashwini Deshpande (2004). On the other hand, while playback singing allows women who are seen not to be literally heard, it also allows an unseen woman to be heard. This provides an opportunity for female expression, perhaps an empowering one. Pavitra Sundar (2008) discusses the ways that the voices in Lagaan empower — give a voice to — women characters, though short of being transgressive. Priya Jha (2003: 44) argues that in films of the 1970s
the reiteration of a narrative template that appears 'authentically Indian', insofar as it goes back to well-rehearsed cultural forms [notably mythic male friendship], in fact imposes a generalizing stereotype that works in relation to women in order to recolonize them, to rob them of a place, a voice . . .
but that nonetheless, the song space, because apprehended as one of feminine excess, means that even a song sung by men in a spirit of male friendship may in some sense be thought to be feminised.
Playback singing lends itself to an archetypal approach to character rather than a novelistic one. A cornerstone of the unified personality is dislodged with the separation of body and voice, enabling a perception of aspects of character in generalised gender and caste terms. This may be shored up on the one hand by the notion of dharma, combining a sense of both fate and duty, of demands on behaviour and disposition beyond individual will and desire,65 and on the other by the sense (on which not all scholars would agree) of Hindi movies being constant reworkings of the foundational Hindu epics, The Mahabharata and The Ramayana.66 Rosie Thomas, for instance (1989: 17), points out the way that Radha in Mother India resonates with a number of figures from Hindu mythology: Sita, Savitri, Radha, Lakshmi, Durga, Kali, Surabai and Mother Earth.67 Starting with its title, this film especially lends itself to archetypal characterisations, here inflected with Communist ideas, but it is a potential in the fundamental organisation of the ubiquitous and indispensable song sequences of Hindi cinema.
Because all Hindi movies are musicals, all involving foregrounded singing and dancing, thus none of them are: there is no genre of the musical as such. This also
Introduction 47
ZZ (Th ^ n T bE Perf°rmed mUS1C^ The « of the masala
movie (Thomas 1985) emphasises the presence of a huge range of rasas (we might
zi::st s) r ^the importance °f the ^ « l
balanced. To present this schematically, from Mother India:
Song (opening words)6s
Emotion/subject
'The beloved bride leaves for the house of her beloved'
'My foolish heart sings with the breeze' 'Life goes by as we cut the crop'
'I look for my beloved in every city, every threshold'
'If we have come into this world, we must go on living' 'Oh you departing ones!'
'Sorrow-filled days are over and happy days are here, Oh brother!' 'I will not remove the veil covering my face'
'Neither am I a God, nor a Satan'
'Our carriage rolls along merrily'
'The Festival of Colours' 'My son come back'69
marriage, happiness, apprehension (Radha journeys as a new bride to her husband Shyamu's family home and they prepare for their first night)
sowing, happiness, nature, birth harvesting, happiness, labouring, selling, hardship
desperation, loss (Radha searches, in vain, for Shyamu who, unable to work, has in shame left the family) stoicism, hardship
exhortation, inspiration, achievement (Radha persuades the villagers not to abandon the village after the devastating floods and together they reclaim the land) rejoicing
flirtation, young love, amorous pursuit
(Champa with Ramu)
male youthfulness (Birju enjoying, and
justifying, himself, contrasted with older
brother Ramu's toiling)
festivity, love, marriage, flirtation (Ramu
and Champa's wedding procession, Birju and
Rupa flirting)
festivity, remembering
desperation (Radha calls after Birju who has become a bandit)
The second and third numbers sing of the joys and hardships of labour, in ways probably echoing Soviet musicals such as Tsirk (Circus, 1936), Volga-Volga (1938), Traktoristy (The Tractor, 1938) and Svetlyi put (The Shining Path, 1940)70 but hard to imagine in a classic Hollywood film. (The single exception I can think of is the sequence, done to music but not sung, towards the end of High, Wide and Handsome, 1937, where a circus comes and helps Westerners laying an oil pipe in a hurry to beat a landowners' deadline;71 musicals such as The Pajama Game, 1957, or Finian's Rainbow, 1968, celebrate the community of labour, but not in numbers featuring, respectively, factory work and share-cropping; even Hallelujah! eschews
48 Introduction
extended sequences of blacks singing joyfully as they pick cotton.) The keenness of the..despair, suffering and desperation of 'I look for my beloved in every city, every threshold', 'If we have come into this world, we must go on living' (notably the line 'Even if life is a poison, we must drink it') and 'My son come back' could be matched in opera and also in those musicals in many ways aspiring to the condition of opera (sombre subject matters, largely or wholly through-sung, vocally taxing): Jesus Christ Superstar (1971), Les Miserables (1980), The Phantom of the Opera (1986), Rent (1994); it is hard to think of their equivalent in a conventional Hollywood musical.72
The body-voice relation and range of possible emotions in Hindi cinema go with other ways of handling space and time. I look here first at Mother India, one of the most beloved of all Hindi movies, which makes considerable use of temporal and spatial expansion, but with different implications from those discussed above in relation to The Harvey Girls, On the Town, The Sound of Music and Funny Girl, or indeed blaxploitation (as discussed in Chapter 8). Following this, I look at Pakeezah, something of a cult film, where the handling of space is much more bound up with the situation of the singer and yet is still not mostly about the outpouring of feeling.
The first number in Mother India shows the journey of Radha to Shyamu's family home. Although the journey does involve crossing a river and, as the film shows it, starts in broad daylight and ends at sunset, there is no attempt to suggest how far this is (the next village? a far distant one?) or thus how long roughly one might assume this would take (as one can vaguely construct Maria's or Fanny's journeys). Cuts rather than dissolves (the usual Western convention for indicating a passage of time) are used. No sense of overall direction is maintained through the editing (unlike the right-to-left movement across the screen in 'Confidence' and 'Don't Rain on My Parade', itself pressing against the eye's tendency to look left to right): the procession moves first away from the camera but turning right, then it seems to be curving left; in the river shot it maintains this left-to-right trajectory, but in the final, sunset shot, it is moving right to left (Figures 1.10, 1.11). Radha is not named by the song and, sung by Shamshad Begum (not, as elsewhere, Lata), the song shifts from third person to first when Radha arrives at the family home ('My father and brother gave me all the happiness I needed'). The lack of clear temporal and spatial co-ordinates, and consistent personal ones, is appropriate, for this is not so much Radha going to Shyamu's home as 'The beloved bride [going to] to the house of her beloved . . . This is her destiny'.
'I look for my beloved in every city, every threshold' is likewise, in conventional Western narrative terms, spatially and temporally incoherent. It shows Radha and the boys searching in vain for Shyamu in a range of locations: first posed together in a temple by a river, then wandering fields (in an image that combines growing and harvested crops), then past camels (a laterally tracking shot with she and the boys walking right to left and the line of camels behind them moving left to right, (Figure 1.12)), ending in silhouette against a sunset: temple, countryside, desert, nowhere particular. The loss of any sense of the time and space covered, together
FIGURES 1.10-1.11 'The beloved bride leaves for the house of her beloved' (Shamshad Begum (over)) {Mother India, Mehboob Productions 1957)
SO Introduction
FIGURE 1.12 'I look for my beloved in every city, every threshold': Radha (Nargis/ Lata) and her sons (Mother India, Mehboob Productions 1957)
with the first person lyrics, does permit a psychological understanding, but it also, along with the usual register that the song in another voice (especially this voice) brings with it, allows for this to be seen as mythic (the abandonment of Sita with two sons, as noted in Thomas 1989: 17), dharmic (woman's destiny and duty), archetypal.
The numbers 'Oh you departing ones!' and 'Sorrow-filled days are over and happy days are here, Oh brother!', which follow immediately upon one another, cover an enormous amount of time, both that taken to persuade the villagers to stay and start reclaiming the land, eventually triumphantly, and that for the two little boys to grow into young men, both processes encapsulated in a dissolve from a shot of the boys holding their exhausted mother up to one of them as young men lifting her up as all three laugh together. The ten years, at least, that the reclamation of the village and the boys' reaching adulthood take is achieved in the space of the song, nearly in that of the dissolve. This sequence has a sweeping but clear overall chronological thrust, one given further historical resonance by the final tableau very near the end, the villagers shot from above forming the shape of India: the film was made on the tenth anniversary of Indian independence and released on 15 August, Independence Day (Chatterjee 2002: 60). Yet the sequence also does much to stymie that onward thrust. One element of this is the tableau
Introduction 51
of India, which notably does not register the partition with Pakistan (Figure 1.13). It is not specified when Mother India takes place but, on the one hand, there are no British nor references to them in the film, which puts it without the colonial era, and, on the other, the whole film is framed by Radha as a very old woman opening a new dam for the village, encouraged by young men wearing the style of cap associated with the (post-independence) Congress party, which puts the main events logically well within the period of the Raj. The still unpartitioned India of the tableau is part of an idealised image of India that is historically atemporal, something made possible (or at any rate easy) by the separate time-space of the song. Secondly, although there is a forward trajectory, what is seen is much more static, a long sequence of Radha appealing to her fellow villagers and then toiling in the mud when the sons are children, another shorter one when they are older, a brief montage of ploughing and reaping (now that the 'happy days are here') and then a sequence focused on a young woman on a swing with Ramu in attendance. In other words, there is no sense of onward temporal pressure, rather a series of set pieces. This relates to a third aspect of the sequence. As Radha and the grown Ramu and Birju ride along, a close-up of a wagon wheel dissolves into a series of scenes from Radha's past: marriage, harvesting, a fair, Radha and the boys threshing, ending with a dissolve back via the wheel to Radha with her arms round her grown sons. Such a flashback is entirely credible within the vocabulary of film, but, in
FIGURE 1.13 'Sorrow-filled days are over and happy days are here, Oh brother!' (Mother India, Mehboob Productions 1957)
52 Introduction
addition to its looking backwards against the implicit forward look of the sequence, the wheel imagery emphasises the idea of cyclicality, of events that have no one-off historical specificity, not even Radha's marriage (since this is part of the cycle of life, a notion reinforced by the segue in the immediately following sequence to Ramu and his beloved). Even the shots of little Birju dancing at the fair (shots not seen before in the film) (Chatterjee "2002: 59) suggest unchanging-ness, for this merely anticipates the laddish young man that Birju is to become (or rather that he already is, only smaller). The time, space and place are ahistorical, ideal, archetypal.73
The space-time of the songs in Mother India is especially appropriate for a film advertising its general concerns - and notions of timeless space - in its title. It is Hindi cinema's habit of using space and time this way that makes it so easy for a film like Mother India to push towards the ahistorical. However, even when a film seems pitched at a more particular level, the handling of space and time in the songs does not privilege a highly personalised or private space and time.
Most of the songs in Pakeezah are performed" to order by the courtesan, and main character, Sahibjaan (Meena Kumari). Her first song74 takes place in a pavilion, through which are visible other pavilions with women dancing solo in them as well as a bazaar, linking the women to merchandise; in addition to musicians, a few select clients sit dotted around the pavilion watching Sahibjaan (Figure 1.14). She is there for display and to incite erotic and even perhaps romantic desires, and her dancing provides this but perhaps not the words and the voice. The former are specifically about the situation in which she finds herself and the disgrace of it, symbolised by reference to the dupatta, the scarf that is the sign of female modesty and purity.75 She sings that 'these people' are the ones who have taken her scarf away, gave her money for it, dyed it pink; and the voice is Lata's:
As the husky elaborations of the thumri give way to Lata's sweet lilting melodious tenor, the very quality of her voice purifies the dadra she sings of the disrepute associated with the kotha and suggests the purity of Sahibjaan herself even in the disreputable quarters of the kotha.
(Allen and Bhaskar 2009: 27)76
Two further songs take place in a secluded, more exclusive pavilion, the Pink Palace, which, 'seems to exist in a timeless space. Its architectural idioms are of uncertain vintage, and its physical space appears as an entirely self-enclosed and self-consciously fabricated world' (25; my emphasis).
By this point in the film, Sahibjaan has already met Salim Ahmed Khan (Raaj Kumar): by chance he shared a train compartment with her, was captivated by the sight of her feet as she turned over in her sleep and left a note of appreciation between her toes. Now in the Pink Palace, which is beside a railway line, the high pitched whistle of a passing train (heard along with the tinkle of her anklet bells when she turned over in her sleep before Salim) brings back the magic of the encounter (all the more magical for being dream-like yet real). Both the songs she j
Introduction S3
L j       ;l I* I
FIGURE 1.14 'These people have taken away my scarf': Sahibjaan (Meena Kumari/ Lata) (Pakeezah, Mahal Pictures 1972)
sings here are, on the one hand, performed to delight the clients and, on the other, seem also to be about her encounter with Salim. The latter is so despite the fact that in both cases the words of the song have nothing to do with him. In the case of the first song,77 one might say that she is singing what she would like to be the case, since 'he' clearly is not present.
On the last line she gestures towards the garden and fountains and beyond them the train line; the clients all turn to look in the direction of her gesture and are then puzzled by there being no one there; Sahibjaan smiles a little at the joke she has played on them with her gesture, but in effect she has, for us who know about Salim, referred by this gesture to both the desire for his presence and the fact of his absence. Throughout the song, her dancing and expressions are seductive, courtesan-like, although she sings much of the song looking straight ahead, only occasionally acknowledging the men, and at one point singing to her own reflection in the pool. The song in itself, lyrics, music and delivery, as well as some of the way it is addressed (straight ahead, to her reflection) suggest an expression of her desire, yet her bodily performance and coy looks also indicate how well she is doing her job.
A second song78 in the Pink Palace is performed to a single rich admirer, ZafFarali Khan (Kamal Kapoor). The words evoke a romantic encounter. The sense that, despite the specifics (walking on the road as opposed to asleep on a train) the song is about the encounter with Salim is emphasised by mise-en-scene. Even before the number starts, as she walks towards the pavilion from the living area, the tinkling of her anklet bells mingles with the plashing of the fountains: although the former are a standard item of feminine allure in the film, Sahibjaan's are by this point ineluctably associated with the encounter on the train with Salim. The number is performed on a magnificent red and gold carpet that Khan has earlier given her
54 Introduction
and Sahibjaan's clothes for the number combine the same colours: she appears to be his. At the start of the number, Khan sits at one end of his carpet and Sahibjaan in the middle of it, facing him. Much of the time she looks down or to the side, as if singing to herself, but she .does look sometimes straight in his direction and holds out her hand and twists7 it seductively, towards him, yet giving strongly the sense of not making eye contact with him. About halfway through the number she stands up and moves away from him, turning her back on him and leaving the space of his carpet. She moves outside, her back still to Khan; after a while she turns round and moves very slowly back towards him, onto the carpet; but then again turns and leaves the space of the carpet and the pavilion. On the words 'Finally, the night of waiting/Will have its reward', she raises her hand across the fountains and courtyard towards of the railway track, and the camera follows the direction of her gesture. The significance of her movement on and off the space of Kahn's carpet becomes explicit: she is performing to Khan, but she is thinking of Salim; she is not really Khan's. Again she returns to the carpet, facing him, but first her gaze at the candles and then her closing her eyes (crosscut with an extreme close-up of Khan's eyes looking at her) emphasise that she is not really addressing herself to him. Finally she looks upwards, and the camera rises and a pink-red chandelier fills the screen, its candles going out, as the sound of the train whistle is heard, the sound of her encounter with Salim. So loud and long does it sound that it brings the number to an end and Sahibjaan rushes out into the garden to gaze after the train. She does not return to Khan.
The last number in the film takes place at Salim's wedding. He and Sahibjaan meet again some time after their first encounter on the train and fall in love, their love celebrated in 'Chalo Dildar Chalo' ('Let us go darling'),79 a song not only not part of Sahibjaan's courtesan performance, but sung as a duet in a much more Western musical style80 over a montage of the lovers who are not shown to be singing. Despite opposition from his family, he determines to wed her but she runs away, unable to bring shame to Salim by his marrying her. Later he invites her, with some bitterness, to dance at his wedding to someone else. Present, unbeknownst to her but not to us, are Sahabuddin, who is her father and Salim's uncle, as well as her aunt Nawabjaan. She performs a song81 that speaks directly of the situation, of how this day will see the effects of her prayers and shame her listener (implicitly Salim). She and Salim both wear white and she performs on ■ white carpet: white is the colour of mourning in Indian culture, and of purity in Islam (the film has a Muslim setting). She appears veiled but before she even starts to sing, and then again during the song (on 'glances like arrows'), she lifts her veil and looks straight at Salim, who is clearly deeply upset, so much so that he eventually leaves, bringing her singing to a temporary halt. Distressed at his departure, Sahibjaan, through a despairing swinging movement, crashes a chandelier to the ground, shattering the glass. She resumes the number, no longer singing but dancing on the shards of glass, causing her feet to bleed, staining the white carpet, until she collapses into the arms of her aunt. It is a concatenation of ironies: a courtesan, the bridegroom's ex-fiancee, performing at his wedding, in front of her own father;
Introduction 55
the clashing multiple symbolisms of white and blood; the public self-destruction of the centre of Sahibjaan's attraction as courtesan and beloved, her feet; the song that says all this and yet can only mean anything to the singer, Salim and her aunt. The earlier numbers do express Sahibjaan's point-of-view, but to varying degrees displaced, from her erotic performance, from the precise content of the lyrics; only the last song is wholly — and shatteringly - an expression of her feelings.
The emotional displacements of Pakeezah in no way run counter to its powerful melodramatic tone - they do not make the songs any less moving. Both the displacement and the maintenance of emotional intensity are achieved in part by the fact that the voice we hear singing is Lata Mangeshkar's. The sense of different levels of address in Sahibjaan's songs is facilitated by the combination of different aural and visual sources of performance, which also enable the numbers to function in terms of a shift away from narrative investment in an individual character to a sense of the general, the shared, the mythic — in other words, if the numbers are expressive, it is not necessarily in terms of what the character is feeling (though it includes this to some degree) but in terms of general, shared values and tones or rasas. However much the film may be about Sahibjaan, it is also about the name Salim gives her, Pakeezah, meaning pure of heart. This has resonance in the context of a film about a courtesan, but it also encapsulates the notion of the true dharmic self. All songs that work, work in terms of a shared appeal, but this is made explicit in classic Hindi cinema. If the power of the end of Funny Girl is the exposure of the private self in public, in Pakeezah it is the recognition of the wider public resonance of a private destiny.
Gainsbourg, Carosello napoletano, Mother India and Pakeezah, like the Hollywood films discussed here and in the rest of this book, all combine spoken and acted narrative with foregrounded songs and, often, dances. The songs and dances in these films are not interludes, added attractions or commentaries external to the narrative, although they are this in thousands of European, Hindi and Hollywood films. In the films discussed here, there is a relation between the narrative and the numbers, the characters (but not necessarily the actors) in the one sing in the other, the songs have to do with the narrative situation. There are exceptions to parts of this: in Gainsbourg, the relation is oblique; Carosello napoletano sprinkles songs across its sequences, 'Lily Kangy' and the can-can, for instance, shoe-horned into the story of Sisina and Luigi; some of Lena Home's numbers are variety spots in the course of narratives that do not concern her; the songs in Car Wash and the blaxploitation movies are not performed by the characters, although they are often aware of and relate to them. Such procedures are not less typical or worthy of cinema but they are not the focus of this essay and most of those that follow.
Characteristically in the films discussed here, the sense of time and space in the song and dance sequences is different from that in the spoken and acted parts. In the white American films, the particularised individual is at the centre of the song
56 Introduction
and its time-space: the voice comes out of the body (or is taken to) and the time and space amplify the feelings of the song. This does not exclude group and collective song, but it does'embody the figure and ground structure of Western music, usually centred on a (star singer) individual or passed along ('Meet Me in St. Louis', 'Isn't It Romantic?', 'Sunday Clothes') in a chain^of individuals, with chorus functioning to foreground the individuals. There is a much more fluid sense of song and space in both Italian cinema and the African-American films considered here. In the former, there is individuality, and a prized voice may be recognised as belonging to a seen person (Giacomo Rondinella, Claudio Villa), but this is not necessary and voice-body belonging is as unfixed as the location and the representational register (stage, photography, radio, movie) for song. White Hollywood films may recognise that a song is not individual property, in the sense that it perforce draws on conventions and familiarities, and personnel (composers, accompanists), which mean that it cannot be unique to the person singing; nonetheless, this is less often recognised and emphasised compared to the significance for the individual of giving out with a song. Song in Italian cinema — in a manner especially condensed and * elaborated in Carosello napoletano — is more clearly recognised as shared, traditional, pervasive and unconfmed (to persons seen singing, to particular locations or genres), loosely choral, contrapuntal or even acceptably cacophonous. In the case of the African-American movies discussed in Chapters 7 and 8, the songs obviously do not in a literal sense belong to the characters and the people playing them, since they are produced by others elsewhere (radio, records, stages), yet they are understood as part of the community that is portrayed and thus available for individual use. Some of the tension in the films derives from protagonists who wish to distinguish themselves from the community and those who see themselves as integrated into it or acting on behalf of it, and the music addresses both the individuation and the community belonging. Individual and community also characterise song and dance in Hindi movies, but, at any rate in the films considered here, in ways that move towards the archetypal: the time and space of the song, especially by virtue of the use of familiar playback singers but also changes in mise-en-scene, camerawork and editing, constitute a shift in affective register that, to varying degrees, is also ontological, partaking of the mythic or spiritual.
In short, there is something about singing in all these films, be it amplifying, elating, elevating, intensifying, letting go, with the flow, against the grain, but I differently understood and felt in different film traditions. None of the variations sketched above wholly excludes the others: there is individuality, chorality and transcendence across all film cultures, voices that emanate from the mouths we see and voices that don't, songs that concern themselves with inner feelings and songs that don't, performances that occupy space and time naturalistically and those for which space and time contract or expand or become unclear. It is a question of emphasis, of what something happens for whom, when and where, and what's at stake in that, in the space of a song.
I   THE PERFECTION OF MEET ME IN ST. LOUIS
Meet Me in St. Louis has often been considered the perfect musical. It blends the fractious elements of the genre - narrative and spectacle, dialogue and song - into a seamless whole that itself embodies the harmony of the Smith family whose life it depicts. And yet it has also been seen as a film of repression and unhappiness, a critique of the family ideal beneath an overt celebration of it. I have myself had the experience on one occasion of finding what I had always thought of as the most blissful of films the most bleak and miserable. This essay is an exploration of this apparent contradiction.1
★   * *
.   A ginger peachy show.
(Bosley Crowther)2
Has there ever been a more joyous, more warming movie musical than Meet Me in St. Louis'?
(Springer 1966: 135)
. . . immaculate in its perfection.
(Kaufman 1994: 56)
Although the term is perhaps used with less insistence now, for a long period the ideal for the musical was seen to be integration.3 The genre can depict people in more or less realistically presented situations who from time to time unfealistically burst into song and dance. This defies not only verisimilitude but also aesthetic unity: numbers replace the pitch of talking and walking with that of singing and dancing and hold up the narrative drive in favour of lyric expressivity and display. The task
174  Music and presence in blaxploitation cinema
encapsulates the cycle's narrative dilemma: the forward thrust of action cinema against the sense of community it also wishes to celebrate. The sequences of the protagonist in the street to the sound of funk are caught on the cusp of this paradox, before or without the pressure for narrative resolution and community integration: they are purposeful sequences,: going somewhere, yet they fade or are brought to a halt with an intrusive chord,18 not really coming to an end, leaving always the elating sense of ongoing forward movement though time and space.19
mm
I Introduction
1 Quoted here from the Kingjames Bible (1611).
2 The play is by William Shakespeare and others and the song based on Ovid in The Metamorphoses (ad 8); the earliest known settings seem to be by Maurice Greene (1745) and Thomas Chilcot (pre-1760), and it has subsequently been set by, inter alia, Arthur Sullivan (1863/4), Ralph Vaughan Williams (1925), Marc Blitzstein (1937) and Roger Quilter (1938).
3 Poem by Heinrich Heine; the earliest and still the best known setting is by Friedrich Silcher (1837); it has acquired the status of a folksong in Germany.
4 'The March of the Women' by Ethel Smyth.
5 The argument, suggestive of both Henri Bergson and Marcel Proust, is drawn explicitly from Victor Zuckerkandl (1973).
6 The song was written by Hugh Martin and Ralph Blane and performed by Lena Home in Ziegfeld Follies (1946). See Chapter 6.
7 By Al Stillman and Henry Mancini (1963).
8 See, for example, Tagg (1982, 1987) and Tagg and Clarinda (2003).
9 I have seen productions that supplant or supplement it by contemporary songs. In the case of a Declan Donnellan Cheek By Jowl production (1986), the choice of'My Way' captured the mood of the scene, like 'O Mistress Mine', irrelevant in terms of literal meaning and brilliantly right in terms of feeling.
10 See Frith 1998: 183fffor a discussion of the ambiguities of whose voice is heard in a song.
II 'Mon homme' (1916) by Maurice Yvain and Albert Willemetz; English lyrics by Channing Pollock,
12 See Moore (1989) and Frith (1998: 200).
13 See Smith (2005: 106-16).
14 For further discussion of Gainsbourg, see Vincendeau (2010).
15 As quoted in WoUen (1992: 56). See also Green (1981: 261) and Fordin (1984: 235) and, as well as the discussions in Wollen op. cit., Clover (1995: 722-5) and Cohan (2005: 235-7).
16 See also Feuer (1982a: 46) and Altman (1987: 254-8).
17 See also Rogm (1996: 205-8).
18 It is widely held that Hepburn did not get the Oscar for My Fair Lady because she was dubbed (Siefert 1995: 57); Smith (2005: 69) notes the irony of a film that shows a woman
176 Notes
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
finding a voice, especially in some of the more exultant or defiant musical numbers, undercutting the star when she opens her mouth to sing.
Much of what follows draws upon more general discussions of the relationship between music and moving image in film (and other media). This field is now too large to provide a full set of references, so I will particularly just mention here Gorbman (1987) and Cook (1998); there are several excellent collections on the topic. On the musical analogy, see for instance Bordwell (1980, 1993).
Cf. the jumps in time within the limited space of the rooftop for the aspirational 'There's Gotta be Something Better Than This' number and the temporally incoherent, rhythmically secure editing of the delusional 'I'm a Brass Band', the delusory quality of which is emphasised by being filmed in a deserted lower Manhattan with Charity dressed in a band-leader's top and black tights, leading a group of dancing bandsmen. On this reading of Sweet Charity, see Dyer (1992b). On Easter Parade, see Babington and Evans (1985: 12-44).
Because she believes — erroneously — that Jerry is the husband of her best friend, Madge. 'Non-time' because what is being realised is dialectical thought, and how long does a thought take? I am grateful to Michael Piggott and his PhD on time and film at Warwick University for opening up these considerations for me.
By means of a shift from minor to major and slighdy arpeggiated piano chords (see Smith 2002).
Brackett (2000: 24), for instance, indicates some of the reasons musicologists, especially those working in popular music, have been critical of spatial understandings of music, especially for the way these have tended to lead to an emphasis on structure at the expense of experience and impact.
This is discussed by Davies (1994: 231—2), who concludes: 'That music is heardin spatial terms would appear to be more or less universal' (even if the terms of that spatiality may differ culturally, notions of small and big, fat and thin, for instance, being deployed rather than high and low).
Van Leeuwen draws on Tagg (1990: 108).
Leonard G. Ratner (1980) discusses the way that many of the conventions of modern Western musical affective expression derive from the development of song, dance and the theatre in the eighteenth century. By Charles Dumont and Michel Vaucaire.
Cf. Stefani's observation on the relation between movement, flying and song quoted above. He references an observation of Ivan Fonagy: 'On se sent "emporte dans les airs" par une melodie. Le verbe anglais to soar s'applique surtout aux oiseaux et a la musique ("a melody soaring up to the Heavens")' ('One feels "carried away up into the air" by a melody. The English verb to soar applies above all to birds and music') (Fonagy 1983: 146; quoted in Stefani 1987: 34); 'air', commonly in French but archaically in English, suggests a tune as well as the stuff we breathe.
On dance and film, see, inter alia, Delamater (1981), Dodds (2001), McLean (2008), Mitoma (2002), Odom (1977). Cf. Walker (1977).
Note too Potter's argument, signalled in the title of his book, Vocal Authority, that particular
kinds of singing in given periods acquire the cultural capital that gives them status and
thus seem to sound authoritative (Potter 1998).
Dubrow draws here on Bloch (1974).
On St. Louis Blues, see Cripps (1978: 75—85).
On Streisand, see Aaron (2000).
See also Wolf (2002) on such singers as lesbian icons.
For discussion of this argument see Griffin (2002), MacKinnon (2000) and Smith (2005: 5-53, 117-19).
Note also Creekmur and Mokdad's forthcoming collection on the international film musical. Fado, often considered the Portuguese song form par excellence, occurs in a number of different kinds of film in Portuguese cinema: the comedia a Portuguesa (revue based films with a narrative thread, featuring many different kinds of song), isolated numbers in
Notes 177
otherwise unsung dramas and occasional films built around fado or a fado star (e.g. A Severa, 1931; Fado: História d'utna Cantadeira (Fado: Story of a Singer), 1947). What I know about fado is entirely based on the work of Anthony De Melo, currently completing a PhD on fado in Portuguese cinema at King's College London.
42 On dubbing in Italian cinema, see Giraldi, Lancia and Melelli (2010).
43 My account draws especially on Casadio (1995).
44 This version of La traviata was given the title of the Dumas play on which it is based.
45 Star 19, 11 May 1946, p. 6, quoted in Casadio (1995: 25).
46 For detailed overviews see Venturelli (1988) and Arcagni (2006).
47 For further discussion of this, see Dyer (forthcoming). Neumeyer (2004) argues that the formal difference between the musical and 'the Dramatic Feature Film' in Hollywood is less secure than critical habits assume.
48 I discuss this perception of Naples, especially as it relates to the work of Nino Rota and Eduardo De Filippo, in Dyer (2010: 73-80, 96-7). See also Arcagni (2006: 115-18), Iaccio (2000).
49 Neapolitan Merry-Go-Round or Carousel (the latter the title of the film in US distribution; in the UK it was titled Neapolitan Fantasy).
50 Italian criticism tends to treat the Hollywood musical as the Platonic form of the musical, such that any film that does not conform to it is not even deemed to be a musical. This is a point Martin (2001: 68—74) makes in relation to writing on the musical more generally.
51 Only on the soundtrack, though he appeared in a number of films in the 1930s.
52 This theatre was named as a diminutive of the San Carlo.
53 Ettore Giannini originally conceived Carosello napoletano for the cinema, but, Italian cinema being caught up in neo-realism, he produced it first as a stage show (at the Teatro La Pergola, Florence, in 1950) (Cäprara 1998: 19).
54 1917, by Libero Bovio and Gaetano Lama.
55 Just enough perhaps. The poster gives the impression that Lily Kangy was a person who is the singer revealed on stage rather than the title of a song; by Giovanni Capurro and Salvatore Gambardella, it was first performed in 1905; 'la mossa' was associated with it, and it almost certainly was given at Gambrinus.
56 Tide of a 1923 song by Libero Bovio and Nicola Valente.
57 1923, by Libero Bovio and Evemero Nardella.
58 1911, by Salvatore Cardillo and Ricardo Cordiferro.
59 That is, no concern with sound perspective, no ambient sound and the girls singing in smooth, close harmony.
60 By Aziz Kashmiri and Vinod.
61 See Majumdar (2009: 232) for a discussion of the disputed claim in the Guinness Book of Records that Lata Mangeshkar is the most recorded artist in history.
62 My understanding of playback singing was especially stimulated by a paper, 'New Voice, New Body: Female Playback Singers in Early South Indian Cinema', given by Kiranmayi Indraganti at the conference 'Sound, Music and the Moving Image', University of London, 10—12 September 2007.
63 Quoted in Deshpande (2004: 5179), Gadgil apparendy himself quoting from an interview with Lata Mangeshkar in the Illustrated Weekly cf India, April 1967.
64 Cf. inter alia Srivastava (2004: 2022) and Majumdar (2009: 192-8).
65 In the song in Mother India beginning 'If we have come into this world, we must go on living', there is a line '[a true woman]'s modesty is her dharma' and Chatterjee (2002: 53) glosses the latter as 'her rule of conduct, life's imperative'.
66 For discussion of the relation to the epics, see Booth (1995), Gokulsing and Dissanayake (1998: 19-20, 43-4), Mishra (1985, 2002).
67 See Chatterjee (2002: 28—9 and passim) for a more discussion of the links between Mother India and Indian mythology.
68 Translations of song lyrics fromMother India are taken principally from Chatterjee (2002), supplemented by those on the 35mm print circulating in the UK. The lyrics are by Shakeel Badayuni, music by Naushad.
178 Notes
Notes 179
69 For a detailed musical discussion of this song, see Morcom (2007: 139-41, 158-9,161-2).
70 See Enzensberger (1992), Taylor (1999).
71 On High, Wide and Handsome, see Altman (1987: 300-5).
72 Fo.r a discussion of mainly non-Hollywood musicals that do use song and dance •v beyond love and happiness (notably Dong (The Hole), 1998, and Dancer in the Dark,
2000), see Martin (2001); nqtealso Kelly Kessler (2010) on the 'destabilizing' of the Hollywood musical and Kevin Dorinelly's discussion (2000) of the punk musical as an anti-musical.
73 For a more detailed account of this sequence, see Shakila (2008).
74 By Ghulam Mohammed and Majrooh Sultanpuri.
75 The translations of the songs from Pakeezah are an amalgam of the subtitles on the version shown on Channel 4 television in the 1980s and those provided on websites such as www.bollyfm.net/bollyfin/mid/604/songs.html and www.bollywhat.com/lyrics/pak_ lyr.html.
76 'The thumri is a form of light classical music central to the courtesan tradition that is rooted in the romantic and erotic love-play and the conditions of viraha (separation) that define the love of Radha and Krishna. The dadra is similar to the thumri but faster in tempo with a beat cycle of six' (Allen and Bhaskar 2009: 36). The kotha is the courtesan salon.
77 By Ghulam Mohammed and Majrooh Sultanpuri.
78 By Ghulam Mohammed and Kaifi Azmi.
79 By Ghulam Mohammed and KaifBhopali.
80 The male voice is Mohammed Rafi. Allen and Bhaskar (2009: 31) argue that the shift in musical register indicates that 'a complete transformation of idiom is required to articulate Sahibjaan's desire for fulfillment that suddenly seems so possible', related to their overall argument that in Pakeezah 'the discovery of love is cast as a leap into modernity, the experience of lived time' (22). Morcom argues that more generally Western music in Hindi cinema is associated with 'action and plot progression' or 'disturbance', where Indian music is associated with narrative stasis (2007: 159).
81 By Ghulam Mohammed and KaifBhopali.
2 The perfection of Meet Me in St. Louis
1 I gave a very early version of this paper at the University of Stockholm.
2 Quoted in Morella and Epstein (1969: 121).
3 See Griffin (2002: 22-8).
4 Another song, 'Boys and Girls like You and Me', by Richard Rodgers and Oscar Hammerstein, itself dropped from Oklahoma!, was recorded (as a solo for Garland) and filmed as a song for Esther and John as they visit the site of the World's Fair. Like 'The Boy Next Door, it was almost certainly not grounded in any real-life song mode, being an instance of what Raymond Knapp terms the MERM, the musically enhanced reality mode (2006: 67-70). Naremore (1993: 71) notes that Meet Me in St. Louis was inspired by Oklahoma! and discusses their approach to Americana.
5 Hence, strictly speaking, a little anachronistic.
6 Knapp (2006: 395) notes a slight anachronism in that the song was introduced at the Fair and therefore could not have been sung before it, as would be the case in this scene. He also observes (394) that the form of its introduction in the film resembles the 'trading verses' format that had developed in opera and operetta and become known as the 'vaudeville finale', thus in another manner evoking the period.
7 I have been unable to ascertain whether it is a known or specially composed piece.
8 In fact written by the African-American entertainers Robert Cole and J. Rosamund Johnson and featured in their musical Sally in Our Alley (1902). Johnson would become one the key musicologists of the Harlem Renaissance. See Knapp (2006: 98 and 394-5 fn6).
9 Though written for the film by Conrad Salinger and Roger Edens.
10 Though again specially composed, by Nacio Herb Brown and Arthur Freed.
11 See Genne (1983) for more discussion of this shot and the formal aural and visual harmony of the opening sequence and the film as a whole.
12 The sense of depth and also breadth, of the disposition of people and objects (such as musical instruments) in space, is also enhanced throughout the film by developments in sound engineering in the period; it 'allowed technicians to adjust the volume and balance of sounds, masking any contradiction between an intimate dramatic setting and a large performance area'. See Naremore (1993: 77).
13 Minnelli draws on Renoir (among others) in the final ballet of An American in Paris (1951).
14 James Naremore (1993: 76) observes that Mr. Smith's proposal to leave town 'threatens the sense of repetition and continuity', one of the comfortable pleasures of the film.
15 Tom Drake appeared in musicals but not as a singing and dancing star. Van Johnson, a good singer and dancer, was originally slated for the role and presumably would have been given numbers, very much altering the balance of the film.
16 McElhaney supports his endorsement of this reading by a comparison with an ostensibly similar film, Summer Holiday (Rouben Mamoulian, 1948).
17 Cohan (2005: 36) discusses this unclear interjection, rejecting the 'Howdy, Judy!' interpretation.
18 On this sequence and music, see Neumeyer (2004: 124—5).
19 Including 'Have yourself a merry little Christmas, it may be your last/Next year we may all be living in the past. [. . .] No good times like the olden days, happy golden days of yore/Faithful friends who were dear to us will be near to us no more'; see Kaufman (1994: 9-10), Knapp (2006: 101-2).
20 Quoted in the (uncredited) liner notes to the CD of the complete soundtrack of Meet Me in St. Louis.
3 A Star Is Born and the construction of authenticity
1 In his discussion of this number (and much in the MGM and Garland musical), Cohan (2005: 27) brings out the camp aesthetic at work in it, saying that 'The Man That Got Away' '. . . is performed, staged, and filmed so as to allow Garland/Esther's singing to be legible as authentic and theatrical at the same time'. In my account I come close to arguing that you would choose between these: you'd either accept the performance as authentic, an untrammelled outpouring of emotion, or deconstruct it, seeing it as a fabrication of emotion. I still think, indeed it is proper that I am true to my experience and to the testimony of others, that with performers you love, you may well enter with them (that is, by virtue of them) into the experience of unmediated - uncontrolled, unpremeditated, private - feeling, you can be caught up in the Utopia of authenticity. However, Cohan's discussion gets past the idea that such an experience, once recognised as a performance, indeed as Utopian, would thus be seen 'only' as performance. Rather, his argument suggests that seeing a performance of authenticity as performance is not to see it as inauthentic but rather to recognise the inescapable fact of performance. It is 'evidence of the performer's intense involvement in the song's emotional content and, simultaneously, evidence of her equally intense detachment from it' (26).
4 'I seem to find the happiness I seek': heterosexuality and
dance in the musical
1 Subsequent research (e.g., Griffin 2002, Kessler 2010) suggests that this is most 'integrated' musicals before the 1960s.
2 See Croce (1972: 60-6), Gallafent (2000: 33-43), Hyam (2007: 180-2, 202-7) and Mueller (1985: 76-87) for further discussion of these dances.
3 On this dance, see De Kuyper (1983) and Mueller (1985: 356-9).