Narrative

Dance Narratives in Narratology

Dance¹ has received scarce attention from the field of narratology. In 1964, Claude Bremond (31) and Roland Barthes (49) explicitly acknowledged the narrative power of gestures, yet no later study has developed those initial references into an extensive and comprehensive critical enquiry of dance as a narrative art. Daniel Punday has stressed the importance of the body in shaping narrative concepts such as characters, plot, and setting, but he refers to the body as a textual object, not as a dancing, expressive agent on stage (11–14). The compendiums on transmedial narrative that have proliferated in the past two decades (see Ryan, Narrative across Media, “Narration in Various Media,” Meister, Grishakova, Thon, and Elleström) have generally overlooked dance, and only a few studies have even partially addressed some of its distinctive aspects. Astrid Bernkopf uses Vladimir Propp’s functions to flesh out the characteristics of a particular dance genre, the Romantic two-act ballet fantastique, which became popular on the European stages from 1830 to 1860. Jan Alber and Ingrid Pfandl-Buchegger and Gudrun Ruttensteiner also provide critical analysis of particular dances. While the latter focus on the seventeenthand eighteenth-century European minuet, the former examines a twentieth-century Australian aboriginal stage dance. In both cases, the instances selected for discussion are non-narrative and therefore their conclusions are difficult to extrapolate to narrative dance.

Among the sources that consider representational dance from a more general perspective, an often cited bibliographic reference is Susan Foster’s monograph Choreography and Narrative. She concentrates on the birth of the ballet d’action (the most direct antecedent of the present narrative ballets) in eighteenth-century France and describes how dance progressively developed the theatricality and vocabulary of movements and gestures adequate to convey a story. Her methodological approach is not, however, narratological, but historical. More precisely, she frames her study within cultural history, explicitly distancing herself from the hermeneutic tradition to which narrative theory belongs (Choreography and Narrative xv). Foster’s entry on dance for the Routledge Encyclopedia of Narrative Theory has a similar tenor. It has a strong emphasis on the political imbrications in dance and in narrative but makes scant reference to the narratological aspects of dance.

Marie-Laure Ryan (“Narration in Various Media”) ofers some theoretical reflections on dance as a narrative medium. Her brief overview, however, does not fully grasp its distinctive narrative potential, as she reduces the scope of the medium to illustrating a story known in advance by the audience and to representing the evolution of interpersonal relations and mental life. By contrast, in a previous work, I illustrate the broad range of narrative strategies deployed by the British choreographer Kenneth MacMillan (1929–1992) in his one-act ballet The Invitation (1960). I blend narrative and dance theory and analysis to dissect the ballet and to consider some of the distinctive characteristics of the medium. My argumentation is structured around six central narrative notions (story, plot, narration, time, space, and characters), each developed to illustrate the strong narrative power of that ballet.

On the basis of these antecedents, in this paper I seek to address some general theoretical questions posed by dance as a narrative art. In particular, I attempt to show how dance fits standard definitions of narrative, which are useful (1) to draw the [End Page 102] boundaries between theatre dance and other manifestations of dance (somatic, ritual, or social); (2) to mark the conceptual diference between narrative and non-narrative dance; and (3) to single out prototypical examples of narrative dance and less core examples. My main source here will be David Herman’s definition including plot and experientiality. Progressing from these ontological considerations to the traditional narratological distinction between story and discourse, Monika Fludernik’s assessment about how this dichotomy applies to plays will steer my proposal for a two-level analytical framework for dance, where discourse equals performance. The two last sections of the essay focus, respectively, on the semiotic diversity of dance and on the relevance of context for its interpretation.

I will elaborate on these points with particular attention to three examples of dance: two prototypical narratives, John Weaver’s ballet The Loves of Mars and Venus (1717) and Kenneth MacMillan’s The Invitation (1960); and one less-core example, Leonid Yakobson’s solo Vestris (1969). The former will be particularly useful, as it is considered to be the first dance narrative in the modern history of Western theatre dance.² I will begin by contextualizing these ballets within that history, and by highlighting the coexistence of narrative and non-narrative dance within that tradition.

Brief Note on the History of Western Theatre Dance

When John Weaver premiered Mars and Venus in London in 1717, dance was a well-established entertainment in European courts and theatres. The commonly called “first ballet,” Balthasar de Beaujoyeulx’s Ballet comique de la reine, inaugurated the tradition in 1581. It was a combination of dancing, music, poetry, spoken verses, and scenic efects seeking “to present something magnificent and splendid” (19). Although the intention was not to represent a story, some elements of narrativity could be detected in the performance. The dancers were characterised as gods and goddesses, and they performed a sequence of numbers very loosely connected by a thread of plot (19).

Half a century later, another ballet commentator, Monsieur de Saint Hubert, perceived similar incipient signs of narrativity in the dancing of his time and advocated for their development. His treatise on How to Compose Successful Ballets (1641) strongly recommended to choose a good subject and to develop it across the performance so that it could support and inform what was happening on stage (33).³

Efforts to narrativize dance finally crystalized in Mars and Venus (as I will show later), yet this success was neither universal nor sustained. It saw only two or three performances a month for about one year and was then dropped from the repertory (Ralph 25). Weaver would compose just two more narrative ballets: Orpheus and Eurydice, in 1718, which had no lasting success, and The Judgment of Paris, much later in 1733 (Ralph 26). The new dance genre would become cemented slightly later and elsewhere. Choreographers, such as Marie Sallé, Jean-Baptiste Dehesse and Jean Jacques Noverre in France; Franz Hilverding and his disciple Gasparo Angiolini in Vienna and Italy; and many others across Europe, developed and disseminated the ballet d’action in the second half of the eighteenth century. [End Page 103]

In the nineteenth century, the ballet d’action evolved into what is known as the “Romantic ballet,” which favoured stories packed with events and supernatural beings. Jules Perrot’s Giselle (1841) epitomizes the achievements of the period, with long passages where dance (not just mime) drives the narrative and delineates the psychological traits of the characters (Guest iii). The Russian Imperial ballet at the end of the century entailed a momentary setback in the evolution of dance narrativity, as it compromised dramatic integrity in favour of technical virtuosity.

In the twentieth century, following Michel Fokine’s innovative practices, ballet acquired a freer and richer narrative expressivity. Choreographers such as Anthony Tudor, Jerome Robbins, and Kenneth MacMillan, among many others, demonstrated that ballet can craft complex plots, use all types of subject matter, and create multifaceted characters that are able to evolve as the story progresses. Similarly, the new vocabularies of dance born at the beginning of the century (the so-called modern dance, with choreographers such as Martha Graham or José Limón) also embraced narrative expressivity from the onset.

Nowadays, the presence of narrative dance on the stage is still active, both through the revivals of past ballets and through new creations such as Matthew Bourne’s Swan Lake (1995), Christopher Wheeldon’s Alice’s Adventures in Wonderland (2011), or Cathy Marston’s The Cellist (2020). Twenty-first century ballet and contemporary dance (the evolved form of modern dance) have both continued producing narratives that build upon the levels of dramatic sophistication attained in the preceding century.⁴

It must be noted that when the ballet d’action was born in the eighteenth century it did not become a substitute for the dance divertissements that had populated the European stage since the advent of court ballet in the sixteenth century, but initiated a parallel development that still persists today. Thus, within Western theatre dance, two broad categories coexist. On the one hand, there is narrative dance (also known as “representational dance,” “story dance,” “imitative dance,” “storytelling dance,” etc.), which is dance enacting a story. And on the other hand, there is non-narrative dance (also known as “abstract dance,” “plotless dance,” “story-less dance,” “non-representational dance,” “non-imitative dance,” or “self-referential dance,” among other expressions); this is dance in which its value derives exclusively from technique and its aesthetic qualities (grace, beauty, dynamism, interplay with the music, etc.). I shall return to this segmentation of stage dance later in the following section.

Defining the Boundaries of Narrative Dance

When examined from a narratological perspective, the first conceptual problem posed by the long narrative tradition succinctly sketched above is that of the identification of narrativity. To this purpose, it seems sensible to start by providing a preliminary groundwork similar to that made by Ryan (“Toward a Definition”) and Herman about prose narratives. When these two scholars discuss and flesh out their respective definitions of narrative, they demonstrate how their notions are useful not only to identify prototypical narratives, but also to exclude or at least detect the very weak narrativity of other manifestations of language. Since I will frequently refer back [End Page 104] to these definitions when I propose a similar gradience of dance narratives, I find it worthwhile to quote both in full.

According to Ryan, the conditions for narrativity are:

1. 1. Spatial dimension
   Narrative must be about a world populated by individuated existents.

2. 2. Temporal dimension
   This world must be situated in time and undergo significant transformations.

3. 3. The transformations must be caused be non-habitual physical events.

4. 4. Mental dimension
   Some of the participants in the events must be intelligent agents who have a mental life and react emotionally to the states of the world.

5. 5. Some of the events must be purposeful actions by these agents.

6. 6. Formal and Pragmatic dimension
   The sequence of events must form a unified causal chain and lead to closure.

7. 7. The occurrence of at least some of the events must be asserted as fact for the storyworld.

8. 8. The story must communicate something meaningful to the audience. (“Toward a Definition” 29; emphasis original)

Ryan provides many examples of representations with no or low score in some of these properties. For instance, she claims that static descriptions struggle with condition (2), weather reports with (4), chronicles and diaries with (6), and recipes and instructions with (7) (“Toward a Definition” 30).

With a diferent set of defining features but with a similar possibility for gradation, Herman proposes that a narrative can be characterised as

1. i. a representation that is situated in—must be interpreted in light of—a specific discourse context or occasion for telling.

2. ii. The representation, furthermore, cues interpreters to draw inferences about a structured time-course of particularized events.

3. iii. In turn, these events are such that they introduce some sort of disruption or disequilibrium into the storyworld involving human or human-like agents, whether that world is presented as actual or fictional, realistic or fantastic, remembered or dreamed, etc.

4. iv. The representation also conveys the experience of living through this storyworld-in-flux, highlighting the pressure of events on real or imagined consciousnesses afected by the occurrences at issue. Thus—with one important proviso—it can be argued that narrative is centrally concerned with qualia, a term used by philosophers of mind to refer to the sense of ‘what it is like’ for someone or something to have a particular experience. The proviso is that recent research on narrative bears importantly on debates concerning the nature of consciousness itself. (14; emphasis original) [End Page 105]

For Herman, while an oral account of a past experience is strong in all four features, and an encyclopedia entry is weak in them, a narrativized description (Herman’s example is the nursery rhyme “This Little Piggy Went to Market”) might contain minimal amounts of (iii) and (iv) (21–22).

Prototypical Dance Narratives: John Weaver’s The Loves of Mars and Venus (1717)

Once the notion of narrative dance has been defined, firstly, as a type of theatre dance (not of somatic, ritual, or social dance) and, secondly, as a diferent form from non-narrative dance, the third step is to elucidate whether there can be prototypical examples of narrative or just less obvious instances. I have advanced several times in this paper that I consider prototypical two of my examples. I’ll now use Mars and Venus to illustrate that contention.

Click for larger view
View full resolution

Figure 1.

John Weaver. The Loves of Mars and Venus (1717), title page.

© British Library Board (General Reference Collection C.121.c.19, title page).

From the printed program of Weaver’s ballet (see Figure 1), we know that this work enacted the story of the romance between Mars and Venus, with a plot that progressed from the initial falling in love, through the rage and jealousy of Venus’ husband Vulcan, to the final happy ending (Weaver, Mars and Venus).⁵ The manager of London’s Drury Lane Theatre, where the ballet was premiered on March 2, 1717, perceived the novelty of this ballet. In his memoirs, Apology for the Life of Colley Cibber (1740), Cibber wrote: “To give even Dancing therefore some Improvement and to make it something more than Motion without Meaning, the Fable of Mars and Venus was form’d into a connected Presentation of Dances in [End Page 107] Character, wherein the Passions were so happily express’d, and the whole Story so intelligibly told, by a mute Narration of Gesture only, that even thinking Spectators allow’d it both a pleasing, and a rational Entertainment” (279).⁶

Cibber’s impressions provide a revealing insight into the key elements of the new dance genre. According to him, Weaver’s work was “more than motion without meaning.” It had a “fable” and conveyed a “whole story.” The diferent scenes or “dances” were “connected” (there was an underpinning plot). And, furthermore, the performers danced “in character,” expressing “passions” and performing actions that were “intelligibly” presented to the audience. Situated in context, this dramatically expressive dance was considered “an improvement.” In contrast with the usual dance divertissements of the time, whose aesthetic value stemmed from the grace and technical accomplishment of the dancers, Mars and Venus had the additional merit of suggesting narrative meaning.

If Herman’s definition is again brought to mind, it can be argued that a ballet with such qualities is a prototypical example of narrative. It is structured as a sequence of events (“connected dances”), introducing disruption or disequilibrium (Vulcan’s obstruction, among others), and involving human agents (there are twenty-six characters, including the three leading roles). Noticeably, it encodes human experientiality in Monika Fludernik’s sense of the term, since a fully and consistently characterized interpretation exposes the interiority (“passions”) and the actions of the human agents in the storyworld.

A similar conclusion might be drawn using Werner Wolf ’s analytic approach. In addition to experientiality, the other two basic narremes are also present. The ballet is meaningful (“intelligibly told”) and representational (with a “fable,” a “story”), as it presents “slices of life” (Wolf ’s expression) that resemble similar human experiences (falling in love, becoming jealous, etc.) (260). In addition, the core narremes locate the story in a particular spatial (the Olympus) and temporal setting (Roman times) and shape it through a chronological and causal sequence of events and actions. For instance, the last scene of the ballet (see Figure 2) starts with Vulcan’s interruption of Mars and Venus’ intimacy, which enrages him. Because he is jealous, he takes them as prisoners. Other Gods and Goddesses (Neptune, among them) then arrive, and Vulcan arrests them. Neptune intercedes with the teleological intention that Vulcan forgives the lovers and sets them free. When Vulcan concedes, all the characters join in the concluding celebratory dance.

Borderline Cases: Leonid Yakobson’s Vestris (1969)

If stage dance is able to produce prototypical narratives, instances located further from the core of narrativity illustrate how it can also generate less central examples. I argued earlier that a continuum between prototypical instances of narrative dance at one end and non-narrative dance at the other allows each particular example to be located at some point on the scale, depending on its degree of narrativity. I suggested that Herman’s definition was an apt tool to this end. I will show now how it might [End Page 108] allocate narrative intensity with more subtlety than just the notion of experientiality, as Fludernik proposes for plays.

Click for larger view
View full resolution

Figure 2.

John Weaver. The Loves of Mars and Venus (1717), page 27.

© British Library Board (General Reference Collection C.121.c.19, page 27).

In the context of her definition of narrativity as the evocation of human experientiality, Fludernik argues that the minimal requirement of narrativity of drama does not rely on the existence of a plot, a narrator figure or fictional world, but on the presence of characters on stage (360). She contends that they guarantee consciousness and experientiality, for they speak, think, and perform actions, and they are perceived as being located in a specific space and time that resembles the human experience of space and time (378). For these reasons, these embodied figures become the indispensable constituents of the narrativity of plays.

Given that dance is also a generally mimetic art, it might seem that Fludernik’s proposal neatly applies to dance too. Her view coincides with the emphasis that the choreographers of the Enlightenment placed on the representation of human experience [End Page 109] as a characteristic value of the ballet d’action (see Weaver’s Essay, Noverre, and Cibber’s appreciation of Mars and Venus, quoted earlier), and even concurs with the conclusions drawn by recent appraisals of the origins of representational dance (see the stress on characters by Cohen, “John Weaver”; Nye).⁷ Yet the coexistence of narrative and non-narrative forms on the dance stage, but not on drama’s (where a non-narrative form is an extraordinarily rare possibility), requires a more exact gradience of narrativity. Definitions of narrative encompassing both plot and experientiality perform this task more satisfactorily. In support of this argument, I provide the following short analysis of a dance with a low degree of narrativity.

Leonid Yakobson’s seven-minute solo Vestris (created in 1969 for dancer Mikhail Barishnikov) is a borderline case of dance in terms of its narrative power.⁸ It delineates the most salient features of some characters (an old woman, a drunkard, a dying man, etc.), but these characters do not interact with each other, inhabit a particular storyworld, produce any change of state or perceive any salient or lasting alteration in their circumstances. No events happen or surround their existence. The performer just evokes some states of mind or characteristic movements of each figure. Since this solo was created in memory of the famous eighteenth-century dancer Gaétan Vestris, one possible interpretation is that the choreography recreates Vestris’ experience of performing those roles onstage. His stage practice is thus the human experience filtered through a consciousness that Fludernik requires for the solo to be considered a narrative. When the notion of plot is introduced into the equation, a robust narrative nature is more difficult to find. Certain narrativity might be perceived (for there is some experientiality) but, in absence of plot, its intensity is far less powerful than that of the prototypical narratives. With this criterion, Vestris would be placed at a midpoint between typical examples of dance narratives and typical instances of non-narrative dance.

Vestris illustrates how the combination of plot and experientiality allows for a more exact gradation of narrative weight in dance works, helping to address the wide variety of stage dance from a classificatory perspective with greater precision. Additionally, it helps to avoid the danger of considering all non-narrative dance as always possessing some degree of narrativity. In dance, there are always human bodies onstage, moving and interacting with each other. The mere presence of dancers might misleadingly steer an automatic identification of a fictional consciousness in them. The requirement of a minimum plot adds a supplementary layer in conceptual terms, which facilitates a more nuanced conclusion about the exact nature of each dance.

Discourse Level as Performance Level

If Fludernik’s conclusions on the narrativity of drama are significantly, albeit partially, useful for examining the narrativity of dance, her model for the narratological analysis of drama provides important clues for driving the discussion on dance forward. She considers two alternate models for dissecting plays, with two and three levels of analysis respectively. The two-level model comprises the level of the story (where issues of plot, characters’ conception, and temporal arrangement of events might be [End Page 110] analyzed) and the level of the discourse (which absorbs the narration process and is equivalent to performance) (362). The three-level proposal keeps the story level, but diferentiates between a discourse and a performance level, with the former referring to the text of the play that might host a narrator/narratee relationship and the latter referring to the enactment on stage (363). When Fludernik extrapolates this analytical frame to all types of narratives, she suggests that the textual and performative levels should be treated as optional (365). She provides two examples: whereas in drama, the narratorial level is optional and the performative level is constitutive, in epic narrative, it is the performance level that is optional (365).

In dance, it is impossible to dissociate the discourse and the performance levels, for no text, script, or blueprint for the performance is ever-present. There might exist some documents planning aspects of the dance performance, such as a notated score of the movements, a libretto or programme note with a synopsis of the story, a musical score, a text with the stage directions for lighting, sheets with set and costume designs, etc., but none of them are exactly equivalent to the text of a play. For instance, the notated score, perhaps the most obvious candidate to qualify for this equivalence, lacks essential narrative content. It represents movements, gestures, and floor patterns at diferent levels of descriptive detail but has little or no reference to the events in the story or the characters.⁹ It is additionally useless for assigning the narrator/narratee roles characteristic of the narrative discourse, not only because of the absence of information about the story level but, crucially, because neither the choreographer nor the audience generally has access to it.

In addition to those intrinsic limitations of dance notation, some other contingent shortcomings might be considered. Most choreographers are illiterate in notation. They create their works in the studio, with the dancers, not at home on a piece of paper. The translation of what happens in the studio into the symbols of a system of notation is the responsibility of another person, the notator: a highly skilled expert who is required not only to write the score but also to decode it later. Furthermore, this professional intervention has costs, preventing most dances from being notated.¹⁰ Consequently, if most dances lack notation, and when it does exist, it usually provides no reference to the story, the notated score cannot be considered analogous to the text of a play.

Librettos are more frequent and provide much more narrative information than notation, but still cannot be considered the text of a dance performance. The libretto of Mars and Venus alludes to aspects of the plot, characters, and fictional time and space, but contains little reference to how the story was enacted. In comparison to a notated score, which might be partially useful for the analysis of the performance level, libretti ofer many clues about the story level. Yet they have their limitations too. In Weaver’s and Noverre’s times, libretti were very popular and, as dance historian Judith Chazin-Bennahum elucidates, were of considerable size (up to forty pages). They described the plot in minute detail. Very often, the choreographer wrote them as an additional tool to help the audience understand the story. This auxiliary function was, however, soon questioned. The Italian choreographer Gasparo Angiolini opposed Noverre’s defence of lengthy libretti and advocated for the power of dance to convey the story through its own means and without the assistance of a printed text.¹¹ [End Page 111]

Although Noverre’s view initially prevailed and many libretti were published in the eighteenth and nineteenth centuries, the opposing view imposed itself. Dance scholar Susan Au fleshes out their role in twentieth-century dance and notes a change in their nature (“Libretti”). When reformers such as Michel Fokine insisted that dance on its own possesses the expressive ability to convey a narrative intelligibly, the publication of libretti soon declined. They progressively evolved into programme notes, which could provide some advance information to the audience, including a plot synopsis, but not exclusively. Nowadays, programs might also contain epigraphs, extracts of poetry; musical, choreographic, or historical analysis; or reflections on the creative process. In addition, it is no longer the exclusive responsibility of the choreographer to write them. Other people usually participate, such as the marketing team, a scholar, or a dance reviewer.¹² To account for this variety in content, Edward Nye borrows Werner Wolf ’s theory of framing to suggest that the best way to describe the controversial nature of programs is to treat them as paratexts, with a framing function similar to that of the cover of a novel or of the trailer of a film (216–18). Following this line of thought, it seems unsuitable to claim that they constitute the discourse level of dance.

To summarize, the particularities of dance narratives suggest that, rather than the three-level model proposed by Fludernik for drama, a two-level model is better suited to the narratological analysis of dance narratives. They comprise a story level (plot, characters, temporal arrangement, etc.) and a discourse as performance level (where the narration process is located and therefore the features of mediation might be detected). Similarly, if Gérard Genette’s notions of story, discourse, and narration are to be applied to dance (despite his resistance to non-prose narratives), the most obvious approach would be to argue that the process of narration (that is, the producing of the narrative) is embedded within the discourse.

This narratological model for the exploration of dance narratives, in which discourse equals performance, also coincides with the well-established view of dance as discourse or text among dance scholars. Susan Foster’s Reading Dancing and Mark Franko’s Dance as Text paved the way for many later studies, but it is perhaps Janet Adshead-Lansdale’s insight that is closer to narratological thinking.¹³ Drawing from semioticians of drama such as Keir Elam, Susan Melrose, and Marco de Marinis, she argues that dance is a performance text (as distinct from a written/notated text) and thus suitable for “reading” or interpretation (Adshead-Lansdale xii–xiii). AdsheadLansdale finds that this view of dance is extremely productive for developing dance studies, for it allows for direct borrowings from other theorists, such as Foucault’s argument that discourse practices have an actively constructed nature, Roland Barthes’ claim that a text is open, or Umberto Eco’s notion of “model reader” (8–14). If dance can be thought of as text, then those and similar theoretical constructions can also be applied to dance. Lansdale’s construction of a theory of intertextuality for dance is based on the same premise (3). [End Page 112]

Semiotic Composition of Dance: Gestural, Kinetic, Visual and Acoustic Cues

Reprising the sketch of a two-level framework for the narratological analysis of dance, the intricacies of the semiotic composition of the discourse deserve lengthier discussion. By comparison to other media, the distinctive texture of dance is composed of a mixture of gestural, kinetic, visual, and acoustic cues, to use Ryan’s terminology. Gestures, motions, and movements are the most characteristic elements in the compound. Typically, they receive most of the attention in analysis and commentary. In the quote about the premiere of Mars and Venus, Cibber stressed the fact that the story was conveyed “by the mute narration of gesture only” (279). Weaver himself emphasized the role of movements when, in the libretto, he claimed that the narrative was transmitted “by gesture and the action of the hands, fingers, legs and feet, without making use of the tongue” (Mars and Venus 740). Interestingly, Weaver included a repertory of expressive gestures in the libretto. For instance, joy was expressed by leaping, grief by casting down the head, and anger by lifting up the hands (see Figure 3).

Click for larger view
View full resolution

Figure 3.

John Weaver. The Loves of Mars and Venus (1717), page 28.

© British Library Board (General Reference Collection C.121.c.19, page 28).

[End Page 113]

It is important to note, however, that despite their significance, motions and gestures do not exhaust the semiotic composition of dance. Acoustic cues are usually present in dance performances, most of the times in the form of music (I will comment on it in the next paragraph). But they might also have the form of linguistic cues, uttered by a voice-over performer (as in in Frederick Ashton’s A Wedding Bouquet [1937], where a voice-over narrator, or a voice-over chorus in some stagings, names the scenes, describes actions and characters, and evaluates their decisions [Morris 83]) or by a dancer/actor onstage (as in Kenneth MacMillan’s Isadora [1981], where the leading role is split between a dancer and an actress on stage who narrates her story to the audience [Parry 534]).¹⁴ This practice of combining language and movement is nowadays characteristic of the hybrid genre of dance theatre, whose most salient exponent is perhaps the late German choreographer Pina Bausch. Additionally, a linguistic cue can also come from a visual sign, written on a prop or on the set. In any case, the presence of spoken or written language in a dance performance, while possible, is generally incidental.

By contrast, music is a typical ingredient of dance. Mars and Venus did not make use of “the tongue,” but it was not mute either. It had a musical score (composed by Henry Symonds and Charles Fairbank), which according to Weaver was “well suited to the subject [it] introduced” (Anatomical 143). Since it has not survived¹⁵ and the choreography is similarly lost, it is impossible to know how the combination of aural and kinetic cues shaped the narrative in this case. A much later ballet, Kenneth McMillan’s The Invitation (1960) with a commissioned score by Mátyás Seiber, is more helpful to illustrate the crucial role that the music plays in the configuration and understanding of dance narratives. This role, which narratological incursions on dance have thus far overlooked, has typically no lesser relevance than that of the rest of the semiotic signs in dance, including the kinetic and visual components.

As is generally the case in dance narratives (Jordan, Moving Music 71), the music of The Invitation is mainly non-diegetic. With the exception of two diegetic moments (a dance lesson and a ball party), it is not part the storyworld. The characters do not hear it, but the audience does. It encodes narrative information, as part of the mediation shaping the story. In this ballet, the aural cues are in fact the first semiotic signs presented to the audience, for the orchestra plays a few bars before the curtains go up.¹⁶ In this opening fragment, the music introduces the story, announcing its tragic tenor. The full orchestra plays a disturbing tune very loudly, with the horns stressing dissonant chords. The efect is unsettling and violente, as the notation in the score reads (Seiber 1). This sense of impending tragedy appears repeatedly in the ballet, aurally tingeing the key events in the plot that will lead to the harrowing end: the rape of the main female character (the Girl) by an adult visiting her parents (the Husband). Distressing sounds introduce this character when he first appears onstage, completing his characterization with a dark hidden side that is not evident from his attractive appearance and polite manners. His actions toward the seduction of the Girl and, particularly, their waltz together during the ball are insistently punctuated with dissonant music, thus warning the audience about the atrocious future consequences of that behaviour (see Figure 4). The role of the music in these dramatic passages is therefore essential to shaping the narrative. It contributes to the characterization [End Page 114] of a leading character and regularly sends foreshadowing cues that anticipate the ending (thus playing with the temporal dimension of the discourse). It also provides information about the type of narration, which can be considered reliable since the denouement of the story, albeit shocking, has been insistently announced.

As this short example shows, musical cues play a significant role in modelling dance narratives. For the purposes of their analysis, it is worth noting briefly that there exist tools for breaking down that contribution. The most comprehensive proposal is perhaps that of dance musicologist Stephanie Jordan, who proposes a wide range of structural categories and expressive devices that can be detected in the choreomusical relationship¹⁷ (Moving Music 63–89).

Click for larger view
View full resolution

Figure 4.

Kenneth MacMillan. The Invitation (1960), with Yasmine Naghdi and Thomas Whitehead as the Girl and the Husband. Design by Nicholas Georgiadis. Music by Mátyás Seiber. Royal Ballet, at the Royal Opera House, London, UK, May 2016.

© Bill Cooper/Royal Opera House/ArenaPAL.

In addition to kinetic and aural cues, visual signs also play a role in shaping the narrative. They might communicate narrative information through written linguistic signs, as suggested above, but also through the qualities of the lighting, through the shapes, colours, and properties of the set, through the costumes and props, etc. For instance, the passage of time in The Invitation is suggested by, among other markers, the colors in the set design. The bright green in the set for the opening scene suggests morning or early afternoon in a garden, the orange coming through the windows [End Page 115] during the dance lesson evokes sunset, and the moon over a deep blue sky in the backcloth presiding over the last scene frames the events at nighttime.

Since the narrative information might come from such a variety of semiotic components in dance narratives, specific methodological tools are needed to analyze them. Jordan’s choreomusical framework is apt for concentrating on the combination of gestural, kinetic, and musical signs, but models of dance analysis provide a wider perspective. Janet Adshead’s seminal framework proposes that the elements to consider are movement, aural cues, visual setting, and the dancer. She provides further details to guide the gaze through the most basic features of each component, with particular emphasis on movement and on the dancer, whose training, dancing style, interpretative skills, and diferent degrees of involvement in the creation of the dance work are of great significance for the final outcome (Adshead 21–32). A well-known alternative for the analysis of dance is Susan Foster’s proposal, which she outlines in her monograph Reading Dancing, which focuses almost exclusively on movement. If these references are succinctly mentioned here, it is only to point out that any narratological discussion on the semiotic components of dance narratives requires appropriate tools to address their distinctive qualities.

Dance and Context

The way in which the combination of semiotic cues in dance triggers the mental representation of a narrative (to use Ryan’s conceptualization) also possesses certain peculiarities. As dance philosopher Henrietta Bannerman reflects, dance, like poetry, “is rich in associative or connotative content but weaker in terms of everyday referential function” (69). The openness of texts that Barthes defended, and the arbitrary relationship between the signifier and the signified that Saussure claimed for language are heightened in the case of dance. In general terms, there is not a systematized method of dance representation, in which a correspondence between certain body movements and their meaning is established. Most dance material is allusive and less literal than human language. Only conventional, ordinary gestures or sophisticated codes of mime inserted in dance sequences might convey ideas on a similar basis to the words of a language (Bannerman 68).¹⁸ Other than that, dance is characterized by a “semanticity” that draws associative ties with things and situations in a less literal and more evocative way than language (Bannerman 69, who borrows Charles F. Hockett’s term “semanticity”). This implies that in order to attach narrative meaning to those floating signifiers not tied by habit or custom to referents, the innate ability to narrativize and the cultural conventions that contextualize every narrative possess a special relevance in the case of dance. Awareness of and exposure to dance practices might considerably increase the chances that the script developed in the mind of the viewer is richer in narrative detail.

The choreographic choices frequently present in the works of a particular choreographer are, for instance, an aspect to consider when contextualizing a dance narrative. When John Weaver premiered Mars and Venus, the accompanying libretto recorded some of the intended meanings of gestures and motions in the choreography. [End Page 116] The libretti that he wrote for later ballets, Orpheus and Eurydice and The Judgement of Paris, were not as rich in the description of motions as Mars and Venus, but still contain enough detail to show that Weaver reworked and reused some of those gestures (Ralph 68–70, 79). As he produced new ballets, he developed a personal way of crafting narratives. It might be assumed that the spectators aware of this evolution might have reconstructed a narrative script more easily and in more detail than those unfamiliar with Weaver’s creations.

A similar progressive consolidation of body expressivity might be detected at a more global level. In her historical revision of the ballet d’action, Susan Foster suggests that as the eighteenth century progressed and the new dance genre spread across Europe, similar efective ways of moving the story forward could be detected in diferent ballet productions (Choreography and Narrative xvii). One of Foster’s examples illustrating this tendency is the action of throwing oneself in the arms of another character, which was regularly used to suggest the idea of feelings being shared or of being in search of empathy and comfort (Choreography and Narrative 183). As in poetry (where words induce resonances beyond their literal meaning), the “semanticity” of that embrace is thus larger and more powerful when contextualized within the ballet tradition.¹⁹

Currently, narratology is highly sensitive to context. Both Herman and Ryan (“Toward a Definition”) include it in their respective characterizations of the basic features that help identifying narratives, as do Wolf in his analytic approach to narratives through narremes, James Phelan in his rhetorical perspective of narrative, and Marco Caracciolo in his enactivist proposal. The importance of the choreographer’s style and of ballet conventions are just two examples that illustrate how necessary is to pay attention to artistic contextualization when assessing dance narratives.

Conclusions

The relevance of context for the understanding of dance, as well as the other general considerations of its nature presented here, ofer a first broad approach to framing its study within the field of transmedial narratology. The general guidelines suggested here systematize the complexity of dance storytelling by pinpointing Western narrative theatre dance as the breeding ground for prototypical Western dance narratives and by aligning dance with drama in the narratological debates on its status and essence. Additionally, the analytical reflections on the semiotic configuration of dance and on the tools appropriate for dissecting it provide some methodological options for examining the mixed fabric of this type of narrative.

Building upon this theoretical basis, the most urgent issue demanding research attention is the study of the full potential of dance as a narrative medium. The two prototypical dance narratives illustrating the argumentation here demonstrate how an analytical approach, alert to their idiosyncrasy and based on narratological concepts, can provide revealing insights into the narrative aspects of dance. Further research in the same direction might expose its rich narrativity, hence refuting claims about [End Page 117] the limited narrative power of dance, which seems to be the dominant view in today’s narratology.

If transmedial narratology intends to explore common grounds and shared practices in a plurality of media, the conclusion of this preliminary study is that it should consider dance too. Theoretically, it can ofer perspectives that may refine established narratological theories and practices, as suggested here on some methodological issues. Historically, dance has been present in the constellation of media from an early stage, forging its ability to enact stories in parallel with other media such as novels and plays. The intersections between dance and other media expose affinities and exchanges that enrich the historical overview. For instance, at the same time that John Weaver premiered Mars and Venus in London, writers such as Daniel Dafoe, Samuel Richardson, Henry Fielding, and Laurence Sterne were establishing novels as a new genre of prose narrative in England. The advent of the ballet d’action also coincided with important shifts in drama. The British actor David Garrick (1717–1779), who regularly performed in London a few years after Weaver and who was a decisive influence on Noverre, fostered a shift from a verbal to a more physical and visual theatre.²⁰ Dance was therefore in tune with the atmosphere of renovation of the time when it inaugurated its own narrative genre. Moreover, this new dance form advanced practices that contributed to the general development of storytelling. According to Nye, the ballet d’action was thematically more adventurous than drama, for it controversially presented depictions of earthy peasant life and kingship bound together in the same scenes in a period when drama tended to focus on the everyday life of ordinary people only (56).

Furthermore, dance historically preceded cinema and musical theatre and decisively contributed to shaping their narrative force. Dancer and choreographer George Balanchine (1904–1983), who had trained as a ballet dancer in the prestigious Imperial Ballet School in Saint Petersburg and who would later found the New York City Ballet, worked on Broadway in the 1930s. He developed and popularized musical theatre as a form of conveying a story entirely through dance and music. Dance historian Beth Genné claims that Balanchine’s musicals were just a modern American form of the narrative ballet tradition that he had learnt in Russia (6). Shortly after his experience in Broadway, Balanchine brought his innovations in musical theatre to Hollywood, thus bringing storytelling practices from dance to the screen (6). He was also a direct influence on film-music director Vincente Minnelli, who had worked with him in Broadway and who in turn would work with Fred Astaire and Gene Kelly (7). Through this and possibly other routes, the narrative strategies that dance had been developing for two centuries in Europe penetrated Broadway and Hollywood. If the agenda of transmedial narratology aspires to account for the cross-fertilization among narrative media, exchanges like this should be taken into consideration. They allow for a fuller view of the history of narrative, its theory and practice. [End Page 118]