Version classiqueVersion mobile

Yeats's Mask

Margaret Mills Harper
Warwick Gould

Yeats’s mask

Playing with Voices and with Doubles in Two of Yeats’s Plays: The Words upon the Window-pane and A Full Moon in March

Pierre Longuenesse

Texte intégral

  • 1 In Per Amica Silentia Lunae Yeats develops the concept of mask, which puts a word on this sensatio (...)

1Despite the apparent allegiance of Yeats’s drama to classical rules, one of its key features, even from his early plays, lies in debunking one of the most standard of elements of mainstream drama in the western tradition: the notion of character. When the poet declares in Per Amica Silentia Lunae that ‘We make out of the quarrel with others, rhetoric, but of the quarrel with ourselves, poetry’ (Myth 331; CW5 8) the poet expresses an inner duality that has been energizing his thought and his poetic practice for some time.1 He may not even guess the dramatic stake stemming from such a statement. On the stage, however, this stake is large: questioning the character’s ability to express a whole and unified discourse cannot but change the form of the drama itself. In Yeats, character as traditionally expressed increasingly gives way to chorality and dialogic effects. The concept of the mask in Yeats’s drama is of paramount importance here as it highlights the disappearance of drama’s traditional mode of interpersonal relationship. Further clarification is needed, however: in a Yeatsian context, what a voice entails needs to be redefined, since voice acts as the cornestone of this issue. Whether in writing or onstage it is possible to focus on registers or on orality, thus raising the question of how the author, concealed or masked as he is, evolves through a unique voice – one which combines all the voices of the dramatic fiction. What I wish to discuss here is not this larger topic but the slightly more limited question of enunciation. In his dramatic texts, and even more so in his Plays for Dancers, Yeats often separates the enunciator from the speaker, and does so with virtuosity. At times, the same speaker may be in charge of several speeches, meaning acts of speaking that would usually be distinct from each other based upon the character from which they emanate. At other times a single discourse of speech may be shared by several speakers, hence the reverse of the first process. Finally, a speaker may even be in charge of someone else’s discourse or speech, thus speaking for, or right next to, someone else. What I wish to focus on is how these phenomena occur, in particular through two rather opposed examples: first, The Words upon the Window-pane, and then A Full Moon in March, a play I shall analyse in more detail.


  • 2 In Antiquity, the word ‘acousmatic’ was used to designate Pythagoras’ disciples who, for five year (...)

2From the early 1900s, Yeats’s drama stages characters in night-time or half-lit atmospheres – through different variations – and in which voices are more often than not separated from the bodies. Although this disassociation between sounds and images seems at first to be used particularly when characters enter and/or exit the stage, it appears to become more and more generalised and systematic. This phenomenon may be traced fairly easily by noting the great number of stage directions referring to sound, the link between sound information and visual details describing the light effects provided by these stages directions, or, more generally, the variety and contracts of voice effects. Starting from Plays for Dancers, Yeats’s theatre originally displays this obsessive acousmatic composition of its characters’ voices – acousmatic sounds being sounds heard but whose location cannot be identified.2 Yeats embeds voices in such a way that the (invisible) Dead are heard via the voices of the (visible) living. This phenomenon is similar to the well-known technique of reported speech, in which a character repeats something said by another. Yet it is a distinct phenomenon from reportage and also from acousmatic sounds, in which something heard is separated from its source. I would rather refer to this as ‘acousmatic enunciation’ than reportage or even acousmatic sound, because what is heard is not just the message (or utterance) but also how it has been voiced (the voice). This coined expression helps to pinpoint the split in enunciation between the speaker and the enunciator, especially when one voice is heard through another one.

3The most representative example of this phenomenon is undoubtedly The Words upon the Window-pane, whose story revolves around a medium-like game in which voices are transferred from one character to another. Written in 1930, the play premiered on 17 November 1930 at the Abbey Theatre but was not published until 1934. The play focuses on an episode of Jonathan Swift’s life, split as it/he was between the two women he called Stella and Vanessa, and on the dramatic triangle stemming from that situation. The originality of the play lies in its exploring this conflict in an indirect manner, through a spiritualist séance one hundred and fifty years after the death of the writer. Characters from the past burst into the present via Mrs Henderson, a medium surrounded by the host and five guests. ‘Who speaks?’ becomes the central question of the plot since Mrs Henderson changes her voice and alternatively ‘reports’ each of the three Dead characters’ voices.

Mrs Henderson [in a child’s voice]. That bad man, that bad old man in the corner, they have let him come back. Lulu is going to scream… O… O… [In a man’s voice] How dare you write to her?
Mrs Henderson [in Vanessa’s voice.]. I questioned her, Jonathan, because I love. Why have you let me spend hours in your company if you did not want me to love you? [In Swift’s voice.] When I rebuilt Rome in your mind it was as though I walked its Streets. [In Vanesssa’s voice.] Was that all, Jonathan? Was I nothing but a painter’s canvas? [In Swift’s voice.] My God, do you think it was easy? (VPl 948–49).

4On the one hand, the confusion of identities is unsettling and the fantastic world bursts in an otherwise naturalist setting. The voices of the Dead are not technically ‘reported’, since Mrs Henderson suddenly alters her voice without a hint of who is next to speak. She is not so much the (indirect) subject of the speech or discourse as the speech channel of others who turn out to be ‘possessing’ her. We may even wonder if this multiplicity of voices uttered by a single body is not the sign of a split self.

5On the other hand, two illusions remain: first, the words spoken are reproduced exactly and, if I may use the term, objectively; second, ᆳist dimension of the play. For if the séance seems to go briefly over the edge when Mrs Henderson abruptly changes her voice in her last cue, dropping and breaking a cup as she is possessed again even though the séance is finished, she nonetheless shifts simply from one tres-passing the limits between past and present or between the dead and the living.

Mrs Henderson. How tired I am! I’d be the better of a cup of tea. [She finds the tea-pot and put the kettle on fire, and then, as she crouches down by the hearth, suddenly lifts up her hands, and counts her fingers, speaking in Swift’s voice.] Five great ministers that were my friends are gone, ten great ministers that were my friends are gone. I have not fingers enough to count the great Ministers that were my friends and that are gone.
[She wakes with a start and speaks in her own voice.]
Where did I put that tea-caddy? Ah! There it is. And there should be a cup and saucer. [She finds the saucer.]
But where’s the cup? [She moves aimlessly about the stage and then letting the saucer fall and break, speaks in Swifts’s voice.] Perish the day on which I was born! (VPl 956).

  • 3 Jacqueline Genet suggests that by studying Swift’s personal crisis in the 18th century, Yeats has, (...)

6In this regard, her change of voices, however unusual it might be, guarantees these limits. Paradoxically, in a play whose major focus is on listening to spectral voices, the classical form of drama – and its underpinning notion of character – is far from being threatened. For not only is the setting framed by naturalism but so are the organisation of speech and the progression of the dialogue. For this reason, I believe that this story of mediumship is in fact quite anecdotal: it strongly suggests that the audience is experiencing an enactment of a previously secret history. Indeed, what one mainly remembers is first the historical episode relating to Swift, however brilliantly and unusually it may be expressed, and, second, the metaphorical echoes Yeats sees with the current affairs of twentieth century Ireland.3 The question of the authenticity of mediumship itself takes a back seat to the other issues of the play.


7In A Full Moon in March (1935), a play almost contemporaneous with The Words upon the Window-pane, two ‘Attendants’ play as arrangers of a unique scene freely adapted from the story of Salomé. As they open the curtain from us, they reveal an encounter between a Queen – offering herself and her queendom to whoever might seduce her by singing her praise – and a Swineherd whose claims lead to beheading before he is given a chance to sing. The plays ends with a seduction dance in front of (or with) the dead man’s head. In the staged version of the story, both the acting composition and the order in which the actions occur defer from the original storyline. The beheading comes before – instead of after – the love parade. What is more, the latter is not meant for a third party (such as King Herod) but for a corpse; moreover, it is staged as a danced sequence alongside a dual soliloquy which is both poetic and fantastic.

  • 4 The play A Full Moon in March, though it was written long after the Plays for Dancers, is inspired (...)

8Influenced as this play is by Noh theatre,4 it is broken down into six successive sequences: the first, fourth and sixth ones are a prologue, an interlude, and an epilogue during which the two Attendants – also musicians – open, comment on, and close the story. The three remaining sequences reveal the first encounter between the two protagonists, the Queen’s dance of death in front of the Swineherd’s severed head (this third sequence is made possible thanks to the time ellipsis created by the closing of the inner curtain), and, finally, each of the two protagonists’ fantastic soliloquy. Two sequences symptomatically revolve around the breaking down and reconstitution of speech interactions, namely the first, the sung prologue of the musicians, and the last but one, or post-mortem conversation – if it may be called as such – between the Queen and the Swineherd.

9Let us start by taking a closer look at the first sequence. In the prologue, the two musicians wonder about the protocol that needs to be followed in order to start the lay. In a very Pirandello-like manner, they refer to orders given by an unknown ‘he’, very likely the author or stage director, and eventually open up the curtain as they sing.

First Attendant. What do we do?
What part do we take?
What did he say?
Second Attendant. Join when we like,
Singing or speaking.
First Attendant. Before the curtain rises on the play?
Second Attendant. Before it rises
First Attendant. What do we sing?
Second Attendant. ‘Sing anything, sing any old thing’, said he.
First Attendant. Come then, and sing about the dung of swine.
[They slowly part the inner curtain. The Second Attendant sings –
The first Attendant may join in the singing at the end of the first or second verse... (VPl 978–79)

10Their song is comprised of three successive stanzas, each one of them being composed of two tercets (similarly to medieval ballads); the last line of each tercet being an italicized refrain.

Every loutish lad in love
Thinks his wisdom great enough,
What cares love for this and that?
To make all his parish stare,
As though Pythagoras wandered there.
Crown of gold or dung of swine.
Should old Pythagoras fall in love
Little may he boast thereof.
What cares love for this and that?
Days go by in foolishness.
O how great their sweetness is!
Crown of gold or dung of swine.
Open wide those gleaming eyes,
That can make the loutish wise.
What cares love for this and that?
Make a leader of the schools
Thank the Lord, all men are fools.
Crown of gold or dung of swine. (VPl 979)

  • 5 Ibid.

11The text indicates that this song is sung entirely by the second Attendant. However, one of the stage directions also mentions that the first Attendant ‘may join in the singing at the end of the first or second verse’.5 Therefore, throughout the refrain a second voice clearly intervenes in the second servant’s cue: the very syntax of the first stanza indicates that one of the sentences is interrupted by an aside: ‘Every loutish lad in love | Thinks his wisdom great enough, | What cares love for this and that?’ The voice keeps asking the same question, and answers it with a symbolical image: ‘Crown of gold or dung of swine’. It is therefore not surprising to hear the closing song of the last sequence – when the curtain is drawn – being clearly shared between the two Attendants’ voices.

Second Attendant. Why must those holy, haughty feet descend
From emblematic niches, and what hand
Ran that delicate raddle through their white?
My heart is broken, yet must understand.
What do they seek for? Why must they descend?
First Attendant. For desecration and the lover’s night.
Second Attendant. I cannot face that emblem of the moon
Nor eyelids that the unmixed heavens dart,
Or stand upon my feet, so great a fright
Descends upon my savage, sunlit heart.
What can she lack whose emblem is the moon?
First Attendant. But desecration and the lover’s night.
Second Attendant. Delight my heart with sound; speak yet again,
But look and look with understanding eyes
Upon the pitchers that they carry; tight
Therein all time’s completed treasure is.
What do they lack? O cry it out again.
First Attendant. Their desecration and the lover’s night. (VPl 989)

12As the second Attendant expresses doubts and questionings in the first person, the first one replies with an obstinate refrain. This is precisely when the italics disappear. Yet, one question remains: if this sharing of voices were clear, why did it not occur from the first song? The confusion obviously derives from the very nature of the character, who is both one and more, being at times two musicians into one. The musicians are the very representation of dialogue heard in a monologue, if not a quarrel with oneself (Myth 331) then a thought in movement, made up of several voices, asides, and episodes. The musicians are basically neither two at the end of the play nor one at the beginning. The uncertainty of the text as to who is speaking shows the emergence of a half-way zone in which the traditional conception of the character fades away. The simple notion of the voice is in fact used to express the cumulative effect of the same and the other and to stage the inner dialogism of discourse. Indeed, the one and only element Yeats chose to distinguish one musician from the other is a vocal sign: the first is a soprano while the second one is a bass.


13The confusion in the organisation of enunication also occurs in another passage in A Full Moon in March, namely in the fifth sequence when the Queen and the Swineherd meet for the last time behind the inner curtain that has been just opened by the musicians. Although the encounter is brief (the scene is only 28 lines long), this is when we hear the two songs, that of the Queen and that of the severed head.

  • 6 VPl 987. The allusion to Mallarmé’s Hérodiade is here limpid, especially the verses: ‘J’aime l’hor (...)

They begin to part the inner curtain. […] The Queen is discovered standing exactly as before, the dropped veil at her side, but she holds above her head the severed head of the Swineherd. Her hands are red. There are red blotches upon her dress, not realistically represented: red gloves, some patterns of red cloth.
First Attendant. Her lips are moving.
Second Attendant. She has begun to sing.
First Attendant. I cannot hear what she is singing. Ah, now I can hear.
(singing as Queen)
Child and darling, hear my song,
Never cry I did you wrong;
Cry that wrong came not from me
But my virgin cruelty.
Great my love before you came,
Greater when I loved in shame,
Greatest when there broke from me
Storm of virgin cruelty.
[The Queen dances to drum-taps and in the dance lays the head upon the throne.]
Second Attendant. She is waiting.
First Attendant. She is waiting for his song.
The song he has come so many miles to sing.
She has forgotten that no dead man sings.
Second Attendant [laughs softly as Head].
He has begun to laugh.
First Attendant: No; he has begun to sing.
Second Attendant (singing as Head):
I sing a song of Jack and Jill […]6

14In the first song, an ‘I’ addresses a ‘you’ and both interlocutors seem to be properly identified: the Queen on the one hand and the severed head on the other. What is more, the tone is one of command, thus confirming that there are two entities, one addressing the other. The scene therefore seems to be a dialogue between two perfectly distinct identities. Yeat’s two games of masks, or duality, are quicky revealed as these two identities come across each other or accumulate with one another.

15The first layer of this game of doubles is found in the inner split occuring in both the Queen and the Swineherd, between each one’s public and private (if not spectral) sides. This split stems from the association between reported speech and singing. The Queen is seen on the other side of the inner curtain. She dances silently, holding the Swineherd’s severed head in her hands covered with blood. Rhythmed as it is by the percussions, this slow dance is getting more and more frenetic: its ecstatic (and even orgasmic) nature is evident, and this love-and-death ecstasy is possible only on the ‘Other Stage’ of Death or Unconsciousness. The dance expresses this ‘second state’ of the Queen, neither living nor dead. Moreover, the relatedness of the dancing Queen’s verbal silence and her speech reported by the female musician turn her words into the expression of a thought, of the character’s inner vision. In a very subtle manner the musician does not say ‘she says’, or ‘she speaks’, but ‘her lips are moving’. Both women are therefore sharing the speech process as one holds the internal intention while the other is in charge of the act of utterance. Then the male musician speaks and sings his turn for the severed head; and his speech helps solve the problem of the impossibility of a song by a dead man, thus contradicting the female musician’s line, ‘She has forgotten that no dead man sings’ (VPl 988). The dialogue between the two songs sung by the two Attendants is therefore a dialogue of the mute, as vocal communication is imposed over a non-vocal engagement: while the Queen is dancing with the Head on the inner stage bordered by the inner curtain, with blood on her hands, the two Attendants speak near us, looking at them. This process of two concurrent duos enables the poet to stage the communication between the living and the dead, in order to highlight their spectral characteristics and to gather them in a zone half way between reality and fantasy, half way, in other words, between the world of the living and that of the dead.

  • 7 In dramatic arts the song opens frequently to a way out of the dramatic action line or diegesis: t (...)

16The singing form of the two texts is also worthy of closer study. By being turned into essentially song lyrics, speeches become enclosed in themselves. The singing creates a fiction within the fiction into which both characters split their own self and, in addition, that of their interlocutor.7 The maternal and endearing terms – child, Darling – used by the Queen in her song as well as the sudden intimacy of a speech bordering on confession and guilt clearly contrast with the character’s initial contempt for the Swineherd. The sung speech is not addressed to a direct interlocutor but rather derives from another self addressing ‘the other in oneself ’, one’s spectral other rather than a real other. Although of a different form, this phenomenon is identical in the head’s song. The latter is inspred from a well-known nursery rhyme and leads to a burlesque mise en abyme of the play’s dramatic situation: Jill, the Queen’s double, has killed Jack, the beggar’s double.

Second Attendant (singing as Head):
I sing a song of Jack and Jill
Jill had murdered Jack;
The moon shone brightly;
Ran up the hill, and round the hill,
Round the hill and back.
A full moon in March.
Jack had a hollow heart, for Jill
Had hung his heart on high;
The moon shone brightly;
Had hung his heart beyond the hill,
A-twinkle in the sky,
A full moon in March.
[The Queen in her dance moves away from the head, alluring and refusing](VPl 988).

  • 8 See n. 9. The choice of the two songs, in their form and their substance, nourishes the battle bet (...)
  • 9 The word ‘ghost’ is worthy of further commentary, even if I won’t analyse here thoroughly Yeats’s (...)

17The two songs are thus two soliloquies answering one another in counterpoint and whose succession builds up a poetic battle. The first is a lament, or self-elegy, while the second sounds more like a popular ballad. The dead Swineherd mocks the Queen he has won in death while she cries over the Mallarmean tragic dimension of ‘the horror of her virginity’.8 In other words, the disassociation between the body and the voice alongside the singing dimension all lead to a speech of intimacy; the emergence of a dream-like speech or parole within the fiction is the sign of subjective splitting. The figures appearing suddenly in this sequence are the inner doubles, if not the ghosts of the two characters seen in the previous sequence.9

18The second layer of the play on doubles in A Full Moon in March occurs not as splits within the characters of the Queen and the Dead Swineherd but between the two characters and the two musicians. In order to explain, let us go back to reported speech. The communication between the two musicians is similar to a two-voiced tale, expressed in the présent, almost a dialgue of stage directions. ‘Her lips are moving’, the first one says; ‘She has begun to sing’, the second one says. ‘She is waiting’, the second one adds (VPl 987). Within this dialogue is reported speech. The Queen herself does not speak physically but the first Attendant (a female musician, in fact) does. The severed head does not speak, but the second Attendant does. In other words, two narrators report the speeches of two characters to which they are physically close. So far, this is not too different from what happens in The Words upon the Window-pane: one character takes on another’s speech, yet without disrupting or blurring the boundaries between them. In A Full Moon in March, however, the reading is not this straightforward: there is a disturbing shift between the stage directions given in speech by the musicians and the related mute and danced scene – the pantomime – of the characters. For instance, nothing indicates that the Queen does indeed move her lips, nor that she is expecting the other’s song. The only stage directions referring to her are descriptions of her dance. The general impression is that the musicians are merely telling each other the story of a Queen and a Swineherd, and that the latter, who actually are beside their tale-tellers, are nonetheless onstage to do something else than illustrate this story. In fact, in speaking as the Queen or the Head, the musicians do not change their voice or their behaviour. The female musician doesn’t actually say ‘she says’, or ‘she speaks’, but ‘I hear’. If the supposed voice she says she ‘hears’ is external to her, it may very well be – also or exclusively – her own inner voice. If this were the case, her speech would only be hypothetically reported. (This is in fact a grammatical paradox: a reported speech in direct speech, with neither punctuation nor an introductory verb, fuels the confusion in the identification of speakers. Actually, the notion of reported speech is not devoid of any ambiguity even in ordinary usage, since the term usually refers both to ‘indirect speech’, that is, reported speech in the strict sense of the expression, and to ‘direct speech’ as long as the latter is reported to, or by, a narrator: in other words, if I say ‘she says that she is here’, or I say ‘she says ‘I am here’’, I am using reported speech in both cases.)

19What is happening onstage is a clear suggestion that the identity lying at the origin of this speech is in fact impossible to pin-point. There is not an illusion of reported speeches. The boundaries between the supposed reporter and the supposed reported seem to disappear. A very subtle play on doubles (two in one, one in two) is gradually taking place between the two Attendants on the one hand and the Queen and Swineherd on the other. In a way, everything comes from the Queen or the Swineherd while the servants only act as their spokespersons. Yet, in another way, the pantomime or danced sequence is a product of the Attendants’ imagination. The two characters may very well be the mere mental projection of the two Attendants, their ghosts, or their puppets. Staging the play in such a way may not be totally nonsensical after all. When the second musician laughs (as the Head) and then says ‘he has laughed’, who is ‘he’? When later the female musicians laughs (as the Queen), but this time it is the male musician who speaks again: ‘she laughs’, who is ‘she’? Words flow back and forth as one comments on oneself or on others, and one becomes both the subject and the object of the discourse. The boundaries between speeches, speakers, and actual bodies do not match anymore. They are sometimes one, two, and even three at times: the one who laughs, the one who says she laughs, and a sillent body next to them. The same may be two or three while two or three may just be the same one.

20We are therefore faced with a paradox: in The Words upon the Window-pane the Dead were not here, but the medium did not really embody them. She only acted as their spokesperson during the séance. Here, the ghosts are onstage and have spokespersons to boot. If the Queen is onstage, the point is not to replace her (there is not a Queen, and another person who speaks for her), but to express, by duality, a different state of hers, a spectral or ghostly state. This is the same for the Swineherd. The split between two halves emphasises their fragility, their respective crack, voiceless bodies or bodiless voices as they are: dancing bodies in between worlds, fleeting voices visiting others.

21It is important to bear in mind that the first half of the play was already preparing the audience for such a perception: although there were four characters onstage, the second musician (male) was already mentioning the Swineherd in his songs while the first musician (female) mentioned the Queen. In the second half of the text, there are still four of them onstage, physically speaking, but there are only two left. Everyone is ghostly, everyone bears a mark – to be there and not to be there.


  • 10 W. R. Veeder, W. B. Yeats, the Rhetoric of Repetition (Berkeley and Los Angeles: University of Cal (...)
  • 11 For instance, ‘The Mask’ published in 1910 in The Green Helmet and Other Poems (VP 263) is partly (...)

22One question remains: whom do the two musicians serve? Do they serve the Queen? Or the mysterious ‘he’ from the prologue? They most likely serve both insofar as they stand in between the time of writing and the time of the story. On the one hand, they encompass the inner movements of a metatheatrical discourse, hidden as it is behind their fictitious identity, within their sung prologues. Yet on the other hand, they act as the spokespersons of the story’s two main protagonists. The play is founded on two embedded elements: an undidentified narrative authority whose voice is entrusted to the two musicians, while the latter project themselves onto two fictitious figures. What is more, a parole does sometimes interrupt the Queen and the Swineherd’s discourses; the said parole actually seems to refer to the initial narrative authority. For instance, in the Swineherd’s song (whether a nursery rhyme or not), the same phenomenon as the one identified in the Attendants’ initial song is found: every third and sixth verse of each stanza is in italics, thus confirming the existence of a refrain. Do the italics refer to a second voice within the song? And is this the case even when the song is already voiced by a voice’s double (the musician’s) which belongs to the creation of an unknown ‘he’? If so, whose voice is indicated in italics? William Veeder suggests that ‘the refrain can be the narrator’s comment on the action’.10 This is a rather tempting hypothesis, and it would help us loop the loop since we would find the writer’s ‘initial voice’, or the voice of his narrative double, at the two ends of the writing process, ‘end’ considered here not as the beginning and end of the play, but as the two ends of the enunciation process. If we consider the text as a setting of several discourses in between each other, organized as concentric circles, we can find: in the external circle, the narrator or ‘he’ mentioned by the two musicians in their first cues; then, the musicians themselves; then, the characters; and at the end, hidden in the core of the latter’s speech (i.e.: in the refrain of the songs), the narrator again. This hypothesis can be indeed confirmed by the very fact that Yeats often used self-quotations and mixed genres.11 In A Full Moon in March, it is in fact rather hard to decide on the level of reality for each of these narrative authorities: it seems each one acts as the other one’s dream, literaly speaking, the product of his/her imagination.

  • 12 The influence of Villiers de l’Isle Adam’s play Axël is quite visible here. In his Autobiographies(...)
  • 13 See J. P. Sarrazac, Poétique du drame moderne (Paris: Seuil, 2012), ch. 2, ‘Drame-de-la-vie: le no (...)
  • 14 In A Vision, Yeats explains that souls cross successive incarnations which coincide with the twent (...)
  • 15 i.e. une ‘théâtre de l’hallucination et de la peur, où ce que l’acteur danseur accomplit semble co (...)

23One last reading of the play highlights its most heightened moment and major claimant to meaning: the Dead Swineherd’s song. It keeps being mentioned until the female musician’s cue: ‘She has forgotten that no Dead man sings’ (VPl 988). He must sing, he wants to sing, the Queen has him beheaded, and yet he sings nonetheless. All leads to the miracle of the dead man singing. We may even add that he will never sing until his head is cut off, and that she won’t dance and sing unless death had come. The song has no other dimension than a spectral one. The song and the voice we hear can be assimilated to the beginnings of intensive love, expressed as it is by the dance.12 As Jean-Pierre Sarrazac suggests, the play is about the drama of life rather than the drama in life.13 It deals with the initiation to mortality, with birth through loss and death, andwith the ecstatic experiences it provokes. This might be ‘the Full Moon in March’, the Resurrection of the God, and the fifteenth moon phase described by Yeats in A Vision: a tie for the unity of the self, a time in which the living and the dead are reunited, a time when man and the divine have become one.14 If the poet resorts to the ritual demanded by the two musicians to stage this story, the ritual is not only formal but necessary. It is required, first, because the onstage presence of the musicians is intrinsically necessary for the ghosts to appear as such, and for the dead man’s voice to be heard: ghosts are merely figments of sight, visions, eidolon maybe, but not characters. Second, ritual is crucial if the logic of haunting is to be followed to the end. Indeed, if the musicians introduce a fiction by ‘inhabiting’ the fictional characters’ voices, they are also quite unsure as to who they are and what they’re expected to say. As they are constantly listening, they are being visited as well as they, themselves, are visiting others. Their speech is haunted by both their fictional characters and their very own patron. This is reminiscent of what Artaud wrote about Balinese theatre, the latter being a ‘théâtre of hallucination and fear, in which the dancing actor’s performance seems to be dictated from an other world’.15

  • 16 Jean-Luc Nancy, Le Partage des voix (Paris: Galilée, 1982), 55.
  • 17 Ibid., 61 et seq., ‘L’oraculaire, le divinatoire, le poétique et l’herméneutique entretiennent des (...)
  • 18 Sigmund Freud, ‘The Uncanny’ (Das Unheimlich), in Writings on Art and Literature, trans. Angela Ha (...)

24We do face here a rather original interpretation of hermeneutics or, as Jean-Luc Nancy describes it, of ‘voice sharing’. Nancy evokes a magical relationship between the poet and the rhapsodic actor, asserting that ‘rhapsodes are poem proclaimers or rather, more accurately, poets’ proclaimers16‘. The rhapsode (or rhapsodist) is not a mere performer but a kind of hermeneus in that he stages the poet’s logos. This situation creates yet another magical play on doubles, between a present voice and an absent voice. ‘The oracular, divinatory, poetic, and hermeuneutic worlds are closely-knit’, Nancy writes17. We may extend this observation a little and bear it in mind when analysing A Full Moon in March. The play focuses indeed first and foremost on inner controversies, solitary voices plagued by inner turmoil, faced with intra- rather inter-subjective conflicts between oneself and the other in oneself. In this regard, all monologues are made of a mental dialogue of a type. Similarly, all dialogues are a disguised form of monologue, as if a unique parole were broken down into serveral opposed and complementary voices. Through its plays on split and duality, A Full Moon in March also displays the confrontation with one’s own death. In his famous article on The Uncanny (‘Das Unheimlich’), Freud explains that ‘the algebraic sign of doubling (doubling of the self, splitting of the self, substitution of the self) ... becomes the uncanny harbinger of death’.18 In A Full Moon in March, the dance enables the character to escape from a duality between incarnate and non incarnate, between the living and the dead. At the same time, the mise en abyme of voices – neither of them being totally reductible to a single entity – never allows for an incarnation to fully take place. The notion of double is here both a psychic and metaphysical category. It embodies the wandering spirit in search of its impossible definitive incarnation.


1 In Per Amica Silentia Lunae Yeats develops the concept of mask, which puts a word on this sensation of inner duality. In the proem, ‘Ego Dominus Tuus’ (see also VP 367–371), the poet stages two opposite voices, Hic and Ille, respectively the mask of the author and the author himself, the former walking alone under the moon, in quest of his ‘anti-self ’, and the latter, who projects himself in action. On earlier uses of the term ‘Mask’ and the paradox of the MSS of the poem, see Warwick Gould, essay, above pp. 32–37, and elsewhere in this volume, passim.

2 In Antiquity, the word ‘acousmatic’ was used to designate Pythagoras’ disciples who, for five years, listened to his lessons hidden behind a curtain, not seeing him and being scrupulously silent. Ten centuries later, the Church fathers gave a religious meaning to the term, thus naming ‘acousmats’ the manifestations of angels’ speaking or singing voices; a whole mystical thinking contributed to the development of this aspect of the concept, based on the listening to celestial voices. Finally today, and since the contemporary acoustic experiments conducted by people such as John Cage or Pierre Schaeffer, the concept of ‘acousmat’ is at the heart of the experimental mechanism of concrete music, as modern acouticians focus on placing the listener in a listening situation where the source of the sound is hidden.

3 Jacqueline Genet suggests that by studying Swift’s personal crisis in the 18th century, Yeats has, in a same movement, exalted the splendour of a past century he took as an example, and lamented its foreseen decadence. Of the noble souls in Mrs Henderson’s living room, of which the quiet Stella is the ghostly figure, is only left a small group of talkative and capricious petit-bourgeois. See J. Genet, Le Théâtre de W. B. Yeats (Lille: Presses universitaires du Septentrion, 1995), 429–40.

4 The play A Full Moon in March, though it was written long after the Plays for Dancers, is inspired, as are the latter, by Japanese Noh, a form to which Yeats was introduced by Ezra Pound, his secretary for several years. The Noh model helped him find a new form more appropriate for the dramatic themes or stories he wanted to write, all dealing with the encounter between reality and the world of spirits, between a living character and some spectral figures coming from the past or out of the imagination. See Mazaru Sekine and Christopher Murray, Yeats and the Noh: a Comparative Study, Irish Literary Studies 38 (Gerrards Cross: Colin Smythe, 1990) and Richard Taylor, The Drama of W. B. Yeats, Irish Myth and the Japanese Noh (New Haven and London: Yale University Press, 1976).

5 Ibid.

6 VPl 987. The allusion to Mallarmé’s Hérodiade is here limpid, especially the verses: ‘J’aime l’horreur d’être vierge et je veux | Vivre parmi l’effroi que font mes cheveux | Pour, le soir, retirée en ma couche, reptile | Inviolé, sentir en la chair inutile | Le froid scintillement de ta pâle clarté, | Toi qui te meurs […]’. See Stéphane Mallarmé, ‘Hérodiade – Scène’, in Poésies, Paris, Gallimard (coll. ‘Poésie’), 1992, 27–34. Yeats remembered Arthur Symons’s translation, and quoted it: ‘Symons first read me Herodiade’s address to some Sibyl who is her nurse and, it may be, the moon also: ‘The horror of my virginity | Delights me, and I would envelop me | In the terror of my tresses, that, by night, | Inviolate reptile, I might feel the white | And glimmering radiance of thy frozen fire, | Thou that art chaste and diest of desire, | White night of ice and of the cruel snow!’ See Stéphane Mallarmé, Herodiade (1864–67, published 1893) II.ii, 103 et seq. tr. Arthur Symons, The Savoy December 1896; Images of Good and Evil (London: William Heinemann, 1899), 77. The interest of such an association is not so much in the ‘quotation effect’ itself as it is in the analogy of enunciation modes, from the swing of a seemingly dialogued voice into the solitude of a mental theatre. The poetical play is not dialogued poetry but imaginary theatre. In the speech, the choice of indirect speech allows to express this intermediary status of a subject vacillating between physical and psychic world.

7 In dramatic arts the song opens frequently to a way out of the dramatic action line or diegesis: the character breaks off the temporal linearity, interrupts the dialogue, and stages him/herself in an other space-time frame, which can be epic or, on the contrary, more lyrical and introspective. In the latter case, we frequently find this phenomenon of ‘closure’, the discourse being less addressed to the other protagonists, and more reflexive. Moreover, the intertextuality often appearing in Yeats’s drama when he uses songs (since he often borrows ballads from other poets, and uses the collage technique), troubles the enunciative system: the singing character sometimes holds a speech which overtakes him, an oracular speech on the edge of madness or hallucination. Another invisible character, coming from Memory or Past – the author of the ballad? Yeats himself? – seems to double the real and visible character. This invisible character trespasses the borders of the double enunciation of drama: in the song, the poet talks to the public, as much as the characters talk to each other.

8 See n. 9. The choice of the two songs, in their form and their substance, nourishes the battle between the two characters. In the first song, the female figure uses the first person (‘... my song, ... I did you wrong’) to speak to her ‘Child and darling’ (VPl 987, emphasis added). By using the imperative mood, she is asking for forgiveness and repudiating her own cruelty. In the other one, the male figure tells a burlesque tale and sets himself remotely, speaking in the third person. For him Death becomes a game, since Jill dances around the stake on which Jack’s heart is stuck, this heart becoming ‘A-twinkle in the sky’ (VPl 988). In other words the two characters give with their respective song a version of their story different from reality, and different from each other’s version. Another game of double emerges here, between the character and his/her fictional double (in the song), or between the two couples.

9 The word ‘ghost’ is worthy of further commentary, even if I won’t analyse here thoroughly Yeats’s vision of this concept. In A Full Moon in March it is not a question of a Greek ‘eidolon’, i.e. the image (or ‘fantasma’) of a Dead or a God coming back to dialogue with the living. Yeats believes in ghosts as far as he believes art or magic are able to give appearence to the character’s double, his/her Daimon: ‘Every voice that speaks, every form that appears ... is first of all a secondary personality or dramatization created by, in or through the medium’ (Introduction to The Words upon the Window-pane, Ex 364). Moreover, for Yeats, this mission of Art creates the conditions of appearance of this Epiphany, the incarnation of one’s double or ‘antiself ’.

10 W. R. Veeder, W. B. Yeats, the Rhetoric of Repetition (Berkeley and Los Angeles: University of California Press, 1968), 21.

11 For instance, ‘The Mask’ published in 1910 in The Green Helmet and Other Poems (VP 263) is partly quoted by Yeats in scene ii of The Player Queen, performed in 1919 and published in 1922 (VPl 738). Later, the introductory lines of The Resurrection (1927, VPl 902), are re-written as ‘Two Songs from a Play’ for The Tower (1928, VP 437). On this occasion, Yeats, consciously or not, does not mention any speaker for the opening lines of the play (‘Song for the Folding and Unfolding of the Cloth’): it seems as if it is the Poet himself who pronounces those words, thus trespassing all ordinary borders of speech organisation in drama.

12 The influence of Villiers de l’Isle Adam’s play Axël is quite visible here. In his Autobiographies Yeats explains his discovery of this particular play, which he saw at its première in Paris in 1894. Later, this influence re-appears in the preface he writes for the first edition of the play in English, translated by H. P. R. Finberg (London: Jarrolds, 1925). In both plays (as in The Shadowy Waters, written and re-written when Axël was created), Love and Hate are mixed together to express a quest for an absolute which finds its outcome in Death.

13 See J. P. Sarrazac, Poétique du drame moderne (Paris: Seuil, 2012), ch. 2, ‘Drame-de-la-vie: le nouveau paradigme’.

14 In A Vision, Yeats explains that souls cross successive incarnations which coincide with the twenty-eight Phases of the Moon. Each of them is the product of a specific conflict happening between different faculties combining with each other. Only two of them escape the rule: the first phase, because during it all the existence is submitted to the material world; and the fifteenth (that of the Full Moon), which symbolises a perfect ‘Unity of Being’. A Full Moon in March is bathed in a full moon atmosphere, and this image, developed by the musicians in their song, seems to be the horizon of the characters’ quest. Thus, when the Queen dances with the head, the Divine (the Queen) and the Human (the Swineherd) seem to join themselves, and reach their quest.

15 i.e. une ‘théâtre de l’hallucination et de la peur, où ce que l’acteur danseur accomplit semble comme dicté d’ailleurs’: see Antonin Artaud, Le Théâtre et son double, in Œuvres (Paris: Quarto Gallimard, 2004), 535–36.

16 Jean-Luc Nancy, Le Partage des voix (Paris: Galilée, 1982), 55.

17 Ibid., 61 et seq., ‘L’oraculaire, le divinatoire, le poétique et l’herméneutique entretiennent des liens étroits’.

18 Sigmund Freud, ‘The Uncanny’ (Das Unheimlich), in Writings on Art and Literature, trans. Angela Harris (Stanford: Stanford University Press: 1997), 211.


Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search