A ‘historically informed’ stage performance of Horace Coignet and Jean-Jacques Rousseau's ‘scène lyrique’ Pygmalion (1770) is enough to whet the appetites of Rousseau scholars and theatre historians specializing in the eighteenth century. They have been dreaming about it for a long time, yet also dreading it: the score, written by the amateur musician Horace Coignet at Rousseau's request in 1770, is strictly functional and artistically mediocre at best. Rousseau's text (1763) is a long monologue by the sculptor Pygmalion lost in the contemplation of his creation, the statue of Galathée. Finally, the gods accede to his prayers and give life to the statue. The play ends with the animation of Galathée and the few words she pronounces as she realizes the marble she was made of has turned into flesh. The last words are left to Pygmalion, who wishes from then on to ‘live only through [her]’.

The conference ‘Rousseau and the Theatre: Political-Aesthetic Ideals and Practices’ (Stockholm, Royal Coin Cabinet, 24–26 August 2015) could not have chosen a better centrepiece than this ‘historically informed performance’ of Pygmalion. The event was organized by Performing Premodernity, an interdisciplinary research group formed to study the musical and theatrical practices and cultures of both Sweden and the rest of Europe from 1760 to 1815. Based at the Department of Culture and Aesthetics at Stockholms Universitet and led by Meike Wagner, Professor of Theatre Studies there, it is being funded for five years (2012–2017) by The Swedish Foundation for Humanities and Social Sciences.

The ten invited speakers were scholars in the fields of literature (Felicity Baker, University College London; Michael O’Dea, Université Lyon 2; Emmanuelle Plagnol-Diéval, Université Paris-Est; David Marshall, University of California Santa Barbara), theatre history (Patrick Primavesi, Universität Leipzig; David Wiles, University of Exeter; Maria Gullstam, Stockholms Universitet and herself the organizer of the conference) and music history (David Charlton, Royal Holloway, University of London; Jørgen Langdalen, Norges teknisk-naturvitenskapelige universitet, Trondheim; and myself (Jacqueline Waeber, Duke University)). In a larger context, this conference also testified to a welcome shift in Rousseau scholarship that had started by the late 1990s with a series of reappraisals of Rousseau as a man of the theatre, eroding a nearly three-centuries-old tradition that had indefatigably cast him as anti-theatrical on the sole basis of his Lettre à M. d’Alembert sur les Spectacles of 1758 – a piece of writing that has since its publication received as many ill-informed comments and interpretations as his equally polemical Lettre sur la musique françoise of 1753.

The Stockholm performance of Pygmalion took place on 25 August at the imposing Riddarhuset (House of the Nobility), a large and clear hall, but one which unfortunately generated an excessive level of reverberation that sometimes muffled the declamation of the actor. This production had previously been performed on stage by the same team at the castle of Český Krumlov in the Czech Republic on 15 June.

Two doctoral students from the Theatre Department of the University of Stockholm, both working on eighteenth-century theatre and members of the Performing Premodernity group, were responsible for the dramaturgy and the costumes – Maria Gullstam and Petra Dotlačilová respectively. The task of acting coach had been entrusted to Jed Wentz: never mind the complete anachronism, but Rousseau's Pygmalion seems to have been written for Wentz, whose research on historically informed performance during the long eighteenth century is well known. Indeed, Pygmalion is a goldmine for anyone working on eighteenth-century declamation and gesture and their relationship with music. In charge of the musical direction was the British-Swedish musical director Mark Tatlow, former artistic director of Drottningholms Slottstheater from 2007 to 2013 and currently research professor at Stockholms Universitet as well as one of the initiators of Performing Premodernity. Coignet's original score requires a string orchestra, two bassoons and two oboes: for the Stockholm performance, however, the score was performed by a string quartet only – not a bad call considering the acoustics of Riddarhuset.

The role of Pygmalion was played by João Luís Paixão, a Portuguese baritone already well experienced in eighteenth-century declamation and who had previously performed melodramatic monologues from the eighteenth and the nineteenth centuries under the supervision of Jed Wentz. Played by the Dutch soprano Laila Cathleen Neuman, the role of Galathée is certainly a most unrewarding one if performed à la lettre, as it was in Stockholm: that is to say, standing completely still for some thirty minutes on a pedestal, until being eventually able to move and walk a few steps on stage, all in order to pronounce no more than ten words.

Since the late 1990s, I have myself attended four stage performances of Pygmalion (two in Geneva, one in Montréal, one in Paris), and none of these had given much, if any, attention to the gestural requirements of this work. To my knowledge, the 2015 performance of Pygmalion is the very first one to have been explicitly historically informed, which means emphasizing its pantomimic dimension, an essential parameter for the intelligibility of the play. Rousseau not only wrote Pygmalion's monologue, but he also gave detailed didascalies (stage directions) that describe the actor's pantomime. The reasons previous performances have consistently neglected the pantomime are as simple as mundane: most actors, who lack any specific knowledge of the eighteenth-century art of gesture, are reluctant to realize those conventional gestures that were part of the eighteenth-century vocabulary of eloquentia corporis, fearing to appear ‘silly’ and redundant on stage. It is true that nowadays, a work like Pygmalion surprises us less by its boldness than its apparent naivety, and by its ‘melodramatic’ display of passions. (I use quotation marks here since Rousseau himself never referred to his work as a ‘mélodrame’: the term only became popularly used to refer to that kind of play, mixing theatrical declamation, pantomime and music, during the early nineteenth century.) Staged renditions of late eighteenth-century melodramas (be it Rousseau's Pygmalion or Benda's Medea), or nineteenth-century melodramatic ballads such as Liszt's Der traurige Mönch, are sometimes on the verge of making us smile as their perceived dramatic excess looms dangerously close to the ridiculous – the same way we watch, incredulous, the expressionist, ‘over-the-top’ acting of film stars from the silent era. Yet this expressive excess is the essential trademark of melodrama, one that would be well pursued in the nineteenth century: redundancy and expressive pleonasm are consubstantial with melodrama, and they are already epitomized in Pygmalion. In the previous productions I have attended, the tendency was consistently to tone down the melodramatic excess, thus denaturing the very nature of the play.

Such a melodramatic quality is evinced through the rhetorical style of Pygmalion's monologue, written in the typical ‘style entrecoupé’ popularized during the late eighteenth century in which the declamation of the actor is frequently interrupted, interspersed with interjections, not to mention pantomimes. This style was widely featured in the genre of the ‘théâtre larmoyant’, and also affected the writing style of the sentimental novel, in which the text becomes saturated with dashes, ellipses and exclamation marks – no surprise that one of the earliest proponents of this style was Rousseau himself, through his own Nouvelle Héloïse.

‘Style entrecoupé’ is precisely all about those moments when ‘the actor, agitated, transported by a passion that prevents him from saying everything, interrupts himself, stops, hesitates, during which moments the orchestra speaks on his behalf; and these silences filled in this manner move the listener infinitely more than if the actor had said with words everything that the music gives to hear.’ (‘L’Acteur agité, transporté d’une passion qui ne lui permet pas de tout dire, s’interrompt, s’arrête, fait des réticences, durant lesquelles l’Orchestre parle pour lui; et ces silences, ainsi remplis, affectent infiniment plus l’Auditeur qui si l’Acteur disoit lui-même tout ce que la Musique fait entendre.’ Rousseau, Dictionnaire de musique, entry ‘Récitatif obligé’, in Œuvres complètes, volume 5: Écrits sur la musique, la langue et le théâtre, ed. Bernard Gagnebin, Marcel Raymond and others (Paris: Gallimard, 1995), 1012–1013; my italics and my translation.) The Stockholm production rendered this just perfectly. João Luís Paixão's French pronunciation may certainly not be perfect, and how to know whether it was correct according to eighteenth-century standards is another story anyway. But this became a rather moot point, for there were truly astonishing moments during which the heightened declamation of the actor felt thoroughly ‘contaminated’ by the musical surroundings that the instrumental ritournelles provided: crescendo effects in the declamation, and ascending lines in the vocal register, gave to the declamation a decidedly musical aspect.

To be able to generate a sense of discursive continuity between music, declamation and pantomime in Pygmalion, so as to create a sense of cohesion between these elements, is another difficulty that previous staged performances have failed to solve. These versions have lacked such cohesion: typically, ‘dead moments’ accumulate between the portions of declamation, a sequence of gestures and the beginning of a ritournelle. The Stockholm performance managed to avoid this pitfall, thanks to Tatlow's careful attention to these transitions. As a melodramatic text, Pygmalion is a patchwork-like score gathering on the same discursive level music, declamation and pantomime. It requires that the musical director be more than just that: indeed, Tatlow had to conduct not so much the string quartet (they could readily play Coignet's bare music by themselves) as these bridging moments, ensuring the smoothest transition possible between the three elements. In so doing the Stockholm performance managed to avoid the anachronistic yet frequent solution of having the actor's declamation being superimposed over the instrumental ritournelle – something that Rousseau's score never specified.

The stage set was minimal yet effective, simply outlining the workshop of Pygmalion, with a half-finished statue, a bust, and, in the centre of the stage, a large fabric shell mounted on a pedestal showcasing the statue of Galathée. It was Mark Tatlow's brilliant and last-minute idea (a few hours before the first performance of the work at Český Krumlov) to have the string quartet hidden behind Galatea's pedestal, using the shell as a backcloth – a solution that uncannily evokes the black backcloth wanted by Arnold Schoenberg in his Pierrot Lunaire: in the first performances of the work in 1912, the musical ensemble was hidden so as to leave the entirety of the scenic space to the sole reciter, Albertine Zehme.

The costumes of the Stockholm Pygmalion also followed historically informed principles: both Pygmalion and Galathée were wearing outfits that perfectly corresponded to the engravings and other written testimonies of the first public performances of the work at the Parisian Comédie-Française in October 1775. This meant maintaining the mismatch of attires between the two characters: the actor Larive (Jean Mauduit), who created the role at the Comédie, had caused a sensation by appearing on stage wearing a Greek tunic with cothurns. Larive's attention to historical detail was, however, not followed by Mlle Raucourt, the actress who played Galathée, as she displayed her statuesque immobility in a contemporary dress with gown hoops. João Luís Paixão filled the stage with many attitudes (all of them following what Rousseau had requested in his stage directions), allowing ‘frozen’ moments to happen in which the actor literally stops his pantomime during a culminating gesture – for instance, when Pygmalion acts out his expression of fear (one arm above the head, another one as if pushing away the statue) immediately after dropping his mallet, having just felt under the marble of Galathée a living pulse. Many of these attitudes resulted in visually striking tableaux vivants that had been immortalized in a few engravings we know of the work (notably some representing Larive himself, and a series of 1775 engravings by Moreau le Jeune).

It was unfortunate that Jed Wentz could not be present in Stockholm on the day of the performance, since the discussion with the audience that followed the performance was much focused on the justification for a historically informed rendition of the work. This discussion also allowed us to revisit issues and enrich the debates that had been previously emphasized during the conference talks.

To sum up the discussions raised by Pygmalion (after the performance and during the conference), the work certainly remains as vexing a topic as it must have been since its inception, as it continues to provide multiple and irreconcilable readings. It is as much a reflection on the creative act, on the moral dimension of theatre, as a reflection of Rousseau's own self. The play expands on another of Rousseau's own theatrical preoccupations, the figure of Narcissus – losing himself in the contemplation of his own image, Narcissus is not far from the sculptor Pygmalion who wishes only to become one with the creature he has created. Pygmalion can also be understood within Rousseau's own conception of a moral and public theatre, devoid of the excesses of theatricality and its artificiality; yet nothing appears more theatrical and artificial than his Pygmalion. And because of the musical component, essential for the intelligibility of the play, we can also read Pygmalion as a work about the limits of musical expression, an issue Rousseau had visited indefatigably since his early writings on music in the 1740s. Another much debated stumbling-block was that if Rousseau had indeed ever had a theory of the actor, and thus of theatre – one in which naturalness, an absence of artificiality, would have prevailed – then how should we understand a historically informed performance of Pygmalion, a work which magnifies theatricality ad absurdum? How can we reconcile Rousseau's search for naturalness with a plot in which the marble artefact becomes a sentient being? How to align Rousseau's ideal of a collective, participatory theatre with a work in which the plot is a pure moment of contemplation that turns its back to the presence of an audience? As so often with Rousseau, the remedy must come with its poison.