Millay Repairs BaudelaireSabine I. Gölz*

Repetition

Literary and cultural spacetimes proliferate through repetitions. Every time we read, write, allude, cite, or translate, we contribute, however little, to that process. Those repetitions create patterns, favour one feature over another, amplify or mute, turn perspectives that way and not this. Massively repeated actions swarm, cluster, and give rise to profound asymmetries that grip our signifying spacetimes. We all conform to a host of habits, settle into an untold number of patterns that become ‘language’ and ‘reality’ for us. No individual reading, translation, or retranslation escapes these force fields. And yet, those processes also rely on each of us. We are the ones who, again and again, do the repeating. Of the decisions we thus make, of the subliminal signals we obey and that channel our readings into predictable traffic patterns, of all these we are overwhelmingly unaware. We ourselves are ‘the unknown […] right here in the very center of the known’.11Barbara Johnson, The Wake of Deconstruction (Oxford: Blackwell, 1994), p. 84. To resist this oblivion to our own agency, to the choices we constantly make, and ultimately to ourselves, we must begin to ask new questions. A choice can be recognized as such only if we can perceive the option to do otherwise. In this essay, therefore, I pair two texts in a contrastive reading: Charles Baudelaire’s poem ‘L’Invitation au voyage’ and Edna St Vincent Millay’s 1936 retranslation of that poem.22Charles Baudelaire, ‘L’Invitation au voyage’, in Oeuvres complètes, ed. by Claude Pichois, 2 vols (Paris: Gallimard, 1976), i, pp. 55–56; Charles Baudelaire, ‘Invitation to the Voyage’, in Flowers of Evil, trans. by George Dillon and Edna St Vincent Millay (New York: Harper and Brothers, 1936), pp. 74–77. Baudelaire’s poem can stand in as a particularly lucid metapoetic description of the gendered mechanisms by which Western literature has kept its readers docile. Millay helps us perceive these because she breaks with them. She finds a new vantage point and creates a divergent poetics. The systematic revisionary critique instantiated by her translation is not entirely unique, however. It has parallels in the works of many other women writers who tend to find similar solutions to a problem they all face.33E.g. Karoline von Günderrode and Ingeborg Bachmann. See respectively Sabine I. Gölz, ‘Günderrode Mines Novalis’, in The Spirit of Poesy, ed. by Peter Fenves and Richard Block (Evanston, IL: Northwestern University Press, 2000), pp. 89–130; Sabine I. Gölz, ‘Apostrophe’s Double’, Konturen, 10 (2019) <https://doi.org/10.5399/uo/konturen.10.0.4509>. While the profound gendering of this apparatus of inequality incentivizes women far more frequently to break with it than male writers, there are exceptions. Heinrich von Kleist is one of those exceptions, and an extremely lucid one at that. A brief introduction and conclusion situate the essay in relation to selected positions in retranslation theory.

Antoine Berman’s ‘Space of Translation’

In his essay ‘La Retraduction comme espace de la traduction’, Antoine Berman argues that while originals are forever young, translations are intrinsically marred by impermanence. We need to retranslate because translations ‘age’, they are ‘subject to time’, and none of them is ‘the one’. But, he continues, while translations are thus subject to aging and expiration, retranslations can sometimes escape that fate, overcome their ‘essential incompletion’, and become ‘great translations’.44Antoine Berman, ‘La Retraduction comme espace de la traduction’, Palimpsestes, 13.4 (1990), pp. 1–7: ‘aucune n’est la traduction’, ‘L’Histoire nous montre qu’il existe des traductions qui perdurent à l’égal des originaux et qui, parfois, gardent plus d’éclat que ceux-ci. Ces traductions sont ce qu’il est convenu d’appeler des grandes traductions’ (pp. 1–2; emphasis in original). Unless otherwise indicated, all translations are my own. First translations are ‘blind and hesitant’, but they can give rise, as an ‘après-coup’, to the possibility of a ‘traduction accomplie’ (accomplished/completed translation).55‘Toute action humaine, pour s’accomplir, a besoin de la répétition. Et cela vaut particulièrement pour la traduction, en tant qu’elle est déjà originairement une opération de redoublement, de duplication. La répétitivité première du traduire est comme redoublée dans la retraduction. C’est dans l’après-coup d’une première traduction aveugle et hésitante que surgit la possibilité d’une traduction accomplie’ (ibid., pp. 4–5). To become ‘great’, a retranslation must enter what he calls the ‘space of translation’,66‘Le thème de cette intervention est: la retraduction comme espace de la traduction. Par “espace”, il faut entendre ici espace d’accomplissement. Dans ce domaine d’essentiel inaccomplissement qui caractérise la traduction, c’est seulement aux retraductions qu’il incombe d’atteindre — de temps en temps — l’accompli’ (ibid., p. 1). a space of ‘l’accompli’ (accomplishment/completion) that can, ‘de temps en temps’ (from time to time), be reached by retranslations.77Ibid. Repetition over time also furnishes the ‘experience’ needed for translation to become ‘conscious of itself’.88‘Il faut tout le chemin de l’expérience pour parvenir à une traduction consciente d’elle-même’ (ibid., p. 4). Berman’s argument thus is predicated on the distinction between time — as the impermanence and transience of translation — and space — which he associates with the durability and permanence of ‘originals’ and ‘great’ translations.99For a lucid critique of this opposition, see Doreen Massey, ‘Space-Time, “Science” and the Relationship between Physical Geography and Human Geography’, Transactions of the Institute of British Geographers, 24.3 (1999), pp. 261–76. Curiously, however, the transition from one to the other, from incompletion (time) to completion (space), is to be achieved via the reduplication of the same fundamentally incompletable process — ‘from time to time’ — of translation itself. The secret to the stability of both ‘originals’ and ‘great’ translations will be found in the processes of repetition themselves — and thus in our own actions.

There is, however, another pattern to Berman’s list of ‘great translations’: his list contains exclusively names of male translators — often canonical authors in their own right. Even more intriguingly, the name ‘Baudelaire’ is particularly fortified by repetition: it appears twice, as both translator and translatee.1010‘La Vulgate de Saint Jérôme, la Bible de Luther, l’Authorized Version sont de grandes traductions. Mais aussi le Plutarque d’Amyot, les Mille et Une Nuits de Galland, le Shakespeare de Schlegel, l’Antigone de Hölderlin, le Don Quichotte de Tieck, le Paradis perdu de Milton par Chateaubriand, le Poe de Baudelaire, le Baudelaire de Stefan George’ (Berman, ‘Retraduction’, p. 2). The predominance of male authors suggests an additional criterion at work in the decision whether a translation joins the pantheon of the ‘greats’ — one that Berman does not mention, and may himself not be particularly conscious of. The suggestion that gender may be a factor in deciding which works become canonized is, of course, hardly news. The more interesting and more difficult question to answer is how: how does gender intervene in literary iterations? Edna St Vincent Millay’s translation of Baudelaire’s ‘L’Invitation au voyage’ can serve as a case in point. Millay is an established poet — yet she is a female one. So, how does her translation compare to Berman’s criteria of greatness? Preceded by two earlier translations into English (both published in 1909),1111Charles Baudelaire, ‘Invitation to a Journey’, in The Flowers of Evil, trans. by Cyril Scott (London: Elkin Mathews, 1909) <https://www.gutenberg.org/cache/epub/36098/pg36098-images.html#Invitation_to_a_Journey> [accessed 4 January 2024]; Charles Baudelaire, ‘The Invitation to the Voyage’, in Jack Collings Squire, Poems and Baudelaire Flowers (London: New Age Press, 1909), pp. 56–57. Squire’s translation is reprinted in Baudelaire in English, ed. by Carol Clark and Robert Sykes (Harmondsworth: Penguin, 1997), pp. 71–73. hers is, at least by that rather mechanical standard, a retranslation. We will see that it also fulfils the criterion of being conscious of itself as a translation. There is one way, however, in which Millay decisively departs from Berman’s criteria: rather than aspiring to the space of canonicity, she systematically prioritizes time, and thus impermanence and change.

Literary traditions have developed elaborate poetic strategies to prefigure and align our fleeting readerly actions, corralling them into patterns. Only by making reading predictable, by scripting our reading habits and thus ultimately our perceptions, can these strategies generate the canonical stability to which writers such as Berman and Baudelaire aspire. What Berman calls ‘space’, therefore, actually arises out of and exists in time: as a constantly renewed pattern of coordinated repetitions. But since those strategies are thoroughly predicated on the gender binarism, they do not work for a woman poet such as Millay; or rather, they work to opposite effect. For her, they produce not permanence but deletion from the canon, not acknowledgement but elision. There is a good reason, therefore, for why Millay revises Baudelaire’s strategies of control and replaces them with a different poetics that is ungendered and non-exclusionary.

Millay’s poem exemplifies a break with a pattern to which we all have been habituated, and to which we all conform. Therefore, it can help us learn from — and potentially repeat — her feat.

Lines 1–6: Baudelaire

The poem opens with an address to ‘mon enfant, ma sœur’ (my child, my sister), exhorting an addressee to consider going ‘là-bas’ (over there) to a land that is said to resemble her:

What resemblance is that? And where is that ‘là-bas’ to which the addressee is asked to cross over? To answer these questions, let us take a moment to consider the relationship between speakers and addressees in a poem.

Poetic address of the kind we see in Baudelaire, it has been argued, actually has not one but two distinct addressees. One is the ostensible addressee of the apostrophe — in this case, the ‘child’ or ‘sister’. The second group of addressees is implicit but actual: they are the readers for whose benefit the scenario of address is displayed. Jonathan Culler quotes Allen Grossman’s description:

The presence of a poem involves a complete triadic state of affairs, in which there is a self, and the beloved of that self, which always has a transcendental character ascribed to it, and a third — the third being the audience, the ratifier, the witness, and the inheritor of the drama of loving relationship to which the poem gives access.1616Grossman, quoted in Culler, Theory of the Lyric, p. 207.

‘Love poems’, Culler continues, ‘are the clearest instance of Grossman’s model, where the beloved acquires a transcendental, nonempirical character, less a person than a poetic function, addressed for poetic purposes’.1717Culler, Theory of the Lyric, p. 207. Apostrophic poetry, then, has a paradoxical effect for its ostensible addressee: caught in the spotlight of address, the ‘beloved’ fades from empirical existence, from person to mere poetic function. How can we explain that loss of personhood and (claim to) actuality?

To answer that question, we must look beyond the text and bring those readers into view for whose benefit the ‘drama of loving relationship’ is displayed, and who do have an actual existence outside of the frame of the poem.1818Stephen Usher, ‘Apostrophe in Greek Oratory’, Rhetorica, 28.4 (2010), pp. 351–62, traces the rhetorical figure of apostrophe back to ancient Greek oratory, where speakers arguing a case before a court would ‘turn away’ (apo-strophein) from the judges and with passionate rhetorical flourish address an absent third person. The actual audience in that scenario, the one that needed to be persuaded, was thus not that absent addressee but rather the judges of the court. In poetic address, the position of the judges is occupied by the actual listeners or readers of the poem. Those shadowy ‘other’ addressees are also the ones who are doing the repeating: they may just read the poem, they may recite it, they may interpret it. But every time someone re-cites the poem, they also again and again re-enact it. And in every such re-enactment, the ‘sister’ will be the addressee, not the speaker. Any repetition of the poem will orient its iterator-apprentices within the gendered apostrophic force field of European literature like iron filings in a magnetic field. Each repetition will assimilate yet another iterator to the role of speaker, who will again renew the call for her to go ‘là-bas’. This leads to a non-trivial realization: not only the addressees are doubled by the triangulation that Culler and Grossman describe; the speakers are, too! We all are those iterators. We all, merely by repeating, have an active role in perpetuating the models we find in literature, poetry, and in language more generally. Why, then, is our function as iterators so difficult to bring into focus? And why has it been so systematically neglected by literary theory?

A question that has probably exercised every single reader of Baudelaire’s poem is ‘where or what is that “là-bas”?’ In Indo-European languages, directions in space such as left, right, up, down, front, back are indicated not relative to an absolute coordinate system but relative to the speaker. Therefore, where ‘over there’ is depends on where the speaker is. If I imagine the speaker on the side of the text, ‘over there’ would be the side of the reader. If I assume the speaker to be the reader, ‘over there’ would be the text. There is no absolute answer to the question of where ‘là-bas’ is.

Baudelaire’s poem, however, does more than send the addressee ‘over there’ (là-bas). It also assimilates her to that other side: line 6 announces that the land to which the apostrophized ‘sister’ has been asked to go resembles her (‘Au pays qui te ressemble!’). The desired end of Baudelaire’s poem — and, we may add, of the larger apostrophic pattern it repeats — is ‘the perfect metaphorical union of its destinatrice with the destination’.1919Barbara Johnson, ‘Poetry and its Double: Two “Invitations au voyage”’, in The Critical Difference: Essays in the Contemporary Rhetoric of Reading (Baltimore, MD: Johns Hopkins University Press, 1980), pp. 23–52 (p. 26). In other words: the ‘other side’ comes to be gendered as ‘feminine’. Where ‘là-bas’ is no longer depends on where the speaker is: it is defined as wherever ‘she’ is!

This in effect reverses the usual speaker-centric functioning of how Indo-European languages code directions. No longer does the position of an actual speaker determine where the ‘other side’ is — especially since those actual speakers are, as iterators and ratifiers, overwhelmingly unaware of their own actions and relative location. Instead, according to the strange magnetism by which cultures and literatures twist our perspectives, the iterators will mechanically line up wherever the feminized ‘other’ is not. They will repeat after and merge with the perspective of the poem’s ‘speaker’. The feminized addressee thus comes to operate as a mere hinge around which readers and iterators keep getting folded again and again into the perspective of ‘not-she’, until they consider it the only possible one. Together with hers, any divergent perspective becomes unimaginable and fades from their world altogether.

And there we have the explanation for the strange fading of the apostrophized feminine other from empirical existence: by dint of that unconscious and subliminal alignment, the apparatus confines its iterators in a bubble within which any ‘she’ is automatically treated as a non-speaker. This has real-life consequences. Massively amplified through our cultural spacetimes, that obedient gendered alignment of ears and gazes creates a highly probable, mobile, flexible, and for all practical purposes absolute frame of reference that rests on the actual elision — mechanically reinforced and normalized by our everyday signifying business as usual — of ‘her’ as a person and agent.

The word là-bas registers that assignment of the feminine to the role as other. On the face of it, the word indeed means ‘over there’. But all it takes is to remove the accent, and the word là (there) turns into the French feminine definitive article, la. Bas means ‘down’ or ‘low’, and the imperative à bas …! means ‘down with …!’ The call for her to go ‘là-bas’, then, is a call not just to go across but also to submit to an always already hierarchized order of representation whose stability is founded on ‘her’ subjugation: ‘down girl’ — as Kate Manne puts it in the title of her recent book on ‘the logic of misogyny’.2020Kate Manne, Down Girl: The Logic of Misogyny (Oxford: Oxford University Press, 2019). The consequences of this are articulated in Baudelaire’s next two lines: for the ‘beloved’, to follow the invitation to ‘live together’ means not only to ‘love’ (aimer), but also to ‘die’ (mourir). She will be turned from a person into a blank signifier, a universal metaphor: ‘A particular kind of commodity acquires the character of universal equivalent, because all other commodities make it the material in which they uniformly express their value.’2121Marx, quoted in Johnson, ‘Poetry and its Double’, p. 36; emphasis in Johnson. For the selected commodity to function as a universal equivalent, it must lose the ability to express its own value.2222See e.g. Karl Marx, Capital: A Critique of Political Economy, trans. by Ben Fowkes (New York: Vintage Books, 1977), p. 161. Marx’s work, of course, is also part of that gendered discursive spacetime, which is why (as is well known) in the text of Das Kapital, commodities as objects of exchange are also insistently feminized and sexualized. That is what Ingeborg Bachmann calls the ‘murder’.2323The disappearance of the feminine narrating ‘I’ at the end of her novel Malina is sealed with the words: ‘Es war Mord’ (It was murder; Ingeborg Bachmann, Malina, in ‘Todesarten’-Projekt, ed. by Monika Albrecht and Dirk Göttsche, 4 vols (Munich: Piper, 1995), iii.1, p. 695). For its unselfconscious enforcers, by contrast, that scheme foresees the mere lazy mechanical repetition — ‘aimer à loisir’ — of the well-worn murderous pattern. Baudelaire’s poem is part and parcel of an entire literary episteme that has trained all of us to conform to and ratify, over and over, a vicious misogynist apparatus.

The misogyny of Western cultures has, of course, not gone unnoticed. Our results so far therefore may not seem overly surprising. But ‘she’ is not the only one who suffers consequences. The poetic apparatus modelled by Baudelaire’s poem sets out to regulate all of our perspectives. The systematic loss of readerly self-awareness it promotes affects all of the iterators who habitually and unwittingly fold themselves into the perspective of the ‘speaker’, and who therefore can no longer perceive texts in any other way than as an utterance. They can no longer wake up to and reclaim their difference from the apparatus of language.2424I would argue that this inability to establish a stable difference from the perspective of the speaker, and thus of language and representation, is also what has been called de Man’s ‘obstacle’; see Alexis C. Briley, ‘De Man’s Obstacles’, Diacritics, 43.3 (2015), pp. 40–65. To show this in detail will be a separate project. Baudelaire’s poem meticulously shows how this comes about.

We tend to take texts for representations. But they are in fact dispositions for actions:

Every perception is prolonged into a nascent action […]. Thus is gradually formed an experience of an entirely different order, which accumulates within the body, a series of mechanisms wound up and ready with reactions to external stimuli ever more numerous and more varied, and answers ready prepared to an ever growing number of possible solicitations.

This memory ‘no longer represents our past to us, it acts it’.2525Henri Bergson, Matter and Memory, authorized translation by Nancy Margaret Paul and W. Scott Palmer (London: Allen and Unwin; New York: Macmillan, 1911), pp. 92–93; emphasis in original. Henri Bergson wrote this over a century ago. Yet a systematic consideration of how language programs our actions by handing us scripts that we then live by, and what we do to each other as a result, is largely absent from scholarly discussions of language and rhetoric. The rare exceptions to this general oblivion are most likely to be found among those of us who have abundant and painful motivation to wake up to both the fundamental dysfunctionality of the representational apparatus for ourselves and the utter inability of those surrounding us to wake up to themselves and their own actions. An army of living enforcers mechanically perpetuates the apparatus of oppression. What do we do? How can we even begin to change this intolerable state of affairs? Before anything else, we need to unglue our eyes from the page and make discernible the actions by which we all ceaselessly turn those representations into our lived reality. Only then can we first understand and then rewrite that apparatus itself, one poem at a time. And that is exactly what Millay does.

Lines 1–6: Millay

Millay’s translation is no mechanical repetition. It begins by rewriting the starting point and foundation of Baudelaire’s poem: Millay deletes the conventional gendered apostrophe and replaces it with an invitation to ‘think’. She starts by drawing attention to our own living, breathing, and embodied actuality:

This initial change triggers a cascade of others that reverse, one by one, all the hierarchies Baudelaire’s poem relies on, systematically rewriting his entire poetic apparatus. The rhymes change. Millay replaces the lure of Baudelaire’s ‘sœur’/‘douceur’ (sister/sweetness) with a rhyme that in his poem (which systematically excludes first-person pronouns) would be unthinkable: ‘be’/‘me’. Baudelaire’s poem suggests that we begin by going ‘over there’ to live together (‘D’aller là-bas vivre ensemble!’). Millay’s begins with the singularity of each of us: ‘all alone’. Notably, however, each of these actually results in its opposite. Baudelaire’s promise of togetherness leads to the feminized other being singled out and turned into a blank signifier. Millay’s invitation to live ‘all alone’, by contrast, invites us to embrace our singular and actual existence as separate from the text — the conditio sine qua non for making reading a reflexive and interactive process, and for opening a critical conversation about how language programs us. Rather than luring us ‘over there’, she asks us to realize that we are ‘here’. With the disappearance of the gendered apostrophe, the word ‘love’ is de-eroticized and degendered, suggesting instead nurturing support for the (always belated) actual reader — ‘my child, my love’. The threat of murder and death has simply disappeared. Where Baudelaire’s lines are dominated by imperatives and exclamation marks (‘Songe à […]!’, ‘Aimer à loisir, | Aimer et mourir […] !’), Millay has questions and hypotheticals (‘would it not be […]?’, ‘Sleep together, share […]?’). This shift in rhetoric follows directly from her strong beginning: if readers bring their thinking — their unrepresentable, self-reflexive, and embodied presence — to bear on the text, signification becomes interactive and unpredictable. Such readers no longer take orders from texts. Whether or not I can ‘live with’ a given poem is not a foregone conclusion but an open question.

One deceptively small change, finally, is a dash not found in Baudelaire that intervenes between lines three and four. What is that dash?

The first word after the dash is ‘sleep’. To agree to ‘live with’ a poem is to give up some of one’s conscious distance and difference — to engage what it models for us, to temporarily go from ‘all alone’ to ‘sleep together’. By marking it with the dash, however, Millay wakes us up to this transition itself. We can thus read the small horizontal line of the dash as a figure for the reflexive textual surface, a signal that we are crossing the line of inversion around which text and reader trade places, taking turns in who takes the lead. The first-person pronoun ‘me’ appears on both sides of the dash, figuring the indexical symmetry and reversibility that allows for this exchange of places. Which I is speaking here? Is the text inviting me to live with it, or am I inviting the text into my life? By placing open indexicals on both sides of that line of inversion, Millay strategically enables a reflexivity and reversibility that supports both readings. The relation between ‘you’ and ‘me’, between reader and text, is figured as a reflexive event: ‘that fair | Country you remind me of?’ Her translation invites our repetition to be reflexive, mind-ful, and self-conscious.

It is no coincidence that Millay begins by foregrounding indexicals. The latter play a critical role in the interplay between texts and readers. Let us therefore take a moment to consider how they work, and why it is so crucial that we pay attention to their placement on the matrix of the text. The Stanford Encyclopedia of Philosophy defines them like this:

An indexical is, roughly speaking, a linguistic expression whose reference can shift from context to context. For example, the indexical ‘you’ may refer to one person in one context and to another person in another context. Other paradigmatic examples of indexicals are ‘I’, ‘here’, ‘today’, ‘yesterday’, ‘he’, ‘she’, and ‘that’. Two speakers who utter a single sentence that contains an indexical may say different things. For instance, when both John and Mary utter ‘I am hungry’, Mary says that she is hungry, whereas John says that he is hungry.2626David Braun, ‘Indexicals’, in The Stanford Encyclopedia of Philosophy, ed. by Edward N. Zalta (Summer 2017) <https://plato.stanford.edu/archives/sum2017/entries/indexicals/> [accessed 4 November 2021].

Whenever I repeat an indexical in a new context — and the context of the iteration is always new — what it refers to changes: time, place, speaker, addressee, and so on. The word I (as the example given in the definition shows) stands out, since speakers use it to refer to themselves. That is unproblematic if we are just telling a friend that we are hungry and ready to have lunch. But how does this work in literary texts? If I, in repeating a poem, assume that the one who is really speaking is still ‘the poet’ — that I am ‘just quoting Baudelaire’ — then I efface my own role as iterator, and thus my own agency. Millay’s invitation for us to re-mind the text envisions a different use. She reinserts indexicals such as ‘me’ and ‘you’ in a way that makes them actualizable. Rather than effacing our own agency and presence, we can use indexicals to shift the entire frame of reference performatively into our own present. We then no longer read the text as an utterance by someone else but use it reflexively to ‘come to language’ ourselves.2727Cf. Ingeborg Bachmann, Kritische Schriften, ed. by Monika Albrecht and Dirk Göttsche (Munich: Piper, 2005), p. 348. It is significant that Baudelaire’s poem, by contrast, includes no ‘here’ and no ‘I’. Instead, in its first six lines, the first person is only marked twice by possessive adjectives (‘mon’ or ‘ma’) — each time relative to the addressee. The iterators of this text can say ‘my’ and ‘mine’, but they cannot say ‘I’. This withholding of open, reversible indexicals tightens the text’s prefigurative power over the minds of readers. Together with the gendered address, the withholding of open ungendered indexicals is the second part of a two-pronged approach to aligning readers with the perspective of the ‘speaker’, and to having readers snap to the grid the poem provides. Indexical consciousness, self-reflexive actualization, by contrast, allows for us to recover and reassert our difference from the text, and thus to absolve ourselves from the authority of the script. That is why Baudelaire withholds open indexicals that readers could use to wake up to their own presence — and why Millay begins with them.

Lines 7–12: Baudelaire

So, let now us see how the next six lines of Baudelaire’s poem use indexicals to align readers’ perspectives with one place on the prefigurative map of the poem — and train them to avoid another. They unfold an entire canopy — metaphoric skies, complete with suns, eyes, and tears — for us to repeat:

The ‘suns’ appear first. They are plural, as if subject to some sort of astigmatic doubling that puts reading out of focus. They are also ‘mouillés’ — wet or drenched — as if the fire of actuality had gone out in them. The skies (‘ciels’), plural like the suns, are ‘brouillés’ — i.e. cloudy, overcast, but also uneven, mixed, non-homogenous, scrambled, or blurred (of vision). Precisely as such however — as unfocused, divided, and overcast — those suns and skies appeal to the ‘esprit’ that appears in the third line of this section and that is marked with the possessive adjective ‘mon’. Barbara Johnson notes the perspectivity of this desire when she writes:

The important common denominator between land and lady, between suns and eyes, is less their shared shining roundness than a common effect produced on the ‘spirit’ of the beholder. The rhetorical meeting point between the two terms (eyes and suns) is not simply that of a metaphorical resemblance but that of a metonymical third term, contiguous to both: the speaker’s desire.2828Johnson, ‘Poetry and its Double’, pp. 26–27; my emphasis.

That ‘metonymy’ of desire is not a mere rhetorical figure. Rather, the contiguity it implies is utterly concrete: it is the contiguity to the text of a long line of iterators who, one after the other, unwittingly assimilate their perspectives to that of the beholder-speaker and take it for their own. The ‘charms’ of those troubled skies, therefore, are not only predicated on a specific perspective, but in repetition that perspective is also mechanically handed on to the iterators. Apprenticed to the perspective of that ‘esprit’, they will re-enact its triangulation, and thus they, too, will learn to find the eyes of the feminized and othered ‘you’ both traitorous and sparkling. For all of the metaphoric blurriness of these lines, one thing is in laser-sharp focus: their exclusion of ‘your eyes’.

But whose eyes are ‘your’ eyes? For as long as we remain within the scenario of address that opens the poem, we will read them as the eyes of the feminized addressee — of the ‘child’ and ‘sister’. And for the apostrophized feminine other to function as a universal equivalent, the relegation of ‘her’ eyes to the exterior of that readerly enclosure obeys an iron logic. They must be seen rather than seeing: ‘brillant’ (shining, sparkling, or dazzling), closer to precious stones than to actual seeing eyes.2929This conversion of ‘her’ eyes into a shower of precious stones arguably also takes place in Poe’s ‘Domain of Arnheim’, which has been cited as another intertextual reference for Baudelaire’s poem: The Collected Works of Edgar Allan Poe, ed. by T. O. Mabbott, 3 vols (Cambridge, MA: Belknap Press of Harvard University Press, 1969–78), iii: Tales and Sketches (1978), pp. 1266–85. I am grateful to Joshua Wilner for alerting me to this intertext. They are ‘traîtres’ (traitorous) by structural necessity, since the exclusion of their perspective is the very foundation of Baudelaire’s signifying apparatus. For the same reason, they (and how and what they might see) must remain ‘mystérieux’ (mysterious) within the perceptual bubble into which the poem is herding its domesticated readers.

However, Baudelaire’s prefiguration leads his readers to exclude much more than only the ‘lady’s’ perspective. It also excludes any actual reading that is not aligned with the perspective and desire of the (imaginary) ‘speaker’ — including the ones that Baudelaire’s iterators would otherwise discover to be their own. The indexicals in these lines are not open and reversible as they were in Millay, where we could read ‘me’ and ‘you’, speaker and addressee, text and reader, here and over there in such a way as to (have them) change places. In Baudelaire, they are possessive adjectives attached to two separate perspectives, one marked as mine, and one as yours, one both speaking and seeing, the other both seen and silent, one assimilative, the other erased. The lingering effect of the apotropaic feminization of ‘your eyes’ anchors that polarization in the gender binarism. It is thus the scenario of address figured as a ‘drama of loving relationship’ itself that keeps iterators properly aligned with the ‘speaker’. That polarization splits the crowd of iterators into two camps, dividing the ‘skies’ of reading into two, and causing the astigmatism of the doubled ‘suns’: readers end up either excluded or co-opted. If co-opted, they can no longer find themselves as readers. A troubled cloud cover has been spread over their eyes: they remain under the poem’s prefigurative spell and can only ‘see’ what the poem has them see. If excluded, they can break with that spell, but their insights remain inaccessible to all assimilated readers. This divided sky, thus, programs two entirely different subject positions into which the overwhelming majority of readers sort both themselves and each other.

Baudelaire uses the word ‘ciels’. This plural form is one of two possible ones for the word ciel (sky). While the plural cieux denotes the ‘undivided universality of the celestial sphere’ and the higher powers associated with it (‘non la pluralité, mais l’universalité indivise de la sphère céleste, ou, au figuré, la Providence, le pouvoir céleste’), the form chosen by Baudelaire denotes a sky that is ‘enclosed by a determinate horizon’, such as the skies of a specific country, but also a painted sky, or even the headboard of a bed.3030Académie Française, ‘Les cieux ou Les ciels’, in Dire, Ne pas dire (11 June 2020) <https://www.academie-francaise.fr/les-cieux-ou-les-ciels> [accessed 12 February 2022]. I would like to thank my colleague Rosemarie Scullion for pointing out this fascinating detail to me. Baudelaire’s skies are thus self-consciously representations. Realizing this should help us reclaim ‘your eyes’ — the ones with which we read the poem — as our own. Let us move on to Edna St Vincent Millay’s version of these lines.

Lines 7–12: Millay

Millay’s translation reinstated, in its first six lines, a reflexive and reversible relation between reader and text. Let us now see how she reshapes the next six lines.

The word ‘charming’ opens the stanza, turning the nominalized charms (‘charmes’) of Baudelaire’s troubled skies into an adjective, almost a verb, a process. As my reading dawns on the poem, as my mind encounters the text’s prefigurative magic and casts its own spell in turn, an open-ended process is set in motion — a doubly reflexive give-and-take, ‘charming in the dawn’. In Millay, only one sun appears, correcting the double vision of Baudelaire’s plural ‘suns’ and divided skies. This reading is conscious of its own singularity. Indexicals, too, work very differently in Millay. Where Baudelaire’s poem polarized readings — one to be metonymically aligned and assimilated (‘my spirit’), the other to be blocked and excluded (‘your eyes’) — in Millay’s text, an indexical has appeared that establishes distance and difference:

The metaphoric sun appears ‘there’. We may read that ‘sun’ as a metaphor or reflection of our metonymic presence as readers, but we do not coincide with it: it appears there (on the readable surface), while we know ourselves to be here (before it but separate from it). That is why that ‘sun’ remains ‘half-withdrawn’. Our presence here is implied, but it cannot be mapped. Whatever fleeting reflections, similarities, analogies, and shifting appearances our reading mind may discover on the textual surface — those effects never, in Millay’s poem, coincide with this actual reader. They are — akin to reflections in a mirror — performative, fleeting, incomplete, predicated on and shifting with our every move. Any movement we perform will change how they appear to us. Self-reflexive reading founds itself on its irreducible distance from representation. If in Baudelaire, the charm of ‘your eyes’ remained mysterious because it was coded as other, in Millay that ‘sun’ remains mysterious because, like all representations, it is ‘half-withdrawn’: incomplete. Whereas Baudelaire’s poem presents us with a polarized map and used the indexical ‘my’ to pull its iterators into one perspective, Millay’s ‘there’ insists on its non-coincidence with whatever appears on the readable surface, and thus preserves our freedom to differ.

The first six lines were articulated by a dash — a line of inversion. Something similar occurs in the second half of the first stanza. Millay’s lines 7–9 assumed the perspective of a reflexive look down onto a readable surface. The next three lines rotate the perspective and bring the ‘curdled skies’ into view. Textual surface and readerly skies change places, underscoring that a prefigurative representation can turn the tables on its readers and reappear as a ‘sky’ that scripts ‘your eyes’. While the sun is singular and focused in Millay, the skies in which it appears are still plural, heterogeneous, even ‘curdled’. The inequities, heterogeneity, and strife Baudelaire and others like him have installed in the cultural and literary ‘skies’ of reading still impact an effort such as Millay’s. This is why, in her poem, the metaphoric ‘sun’ and ‘your eyes’ are both treacherous, untrustworthy: both may script or have been scripted. It is indispensable to think.

Lines 15–26: Baudelaire

With the second stanza, we enter a ‘room’ — an enclosed space, decorated with oriental splendour, and suffused with fragrances and perfumes. Baudelaire begins with a list of what ‘would’ be contained in that room: shiny (but unspecified) furnishings, flowers, aromas.

The rhetoric is impersonal and constative, so much so that the belated arrival of a verb in the conditional (‘décoreraient’ — ‘would decorate’) in the stanza’s third line can no longer quite outweigh the air of factuality established by that beginning. There is a sense of beguiling luxury, but also of an eerie absence of any subject, of focus. The objects are shiny, but what they reflect is ambient unfocused light. Due to its belated appearance, the only possessive adjective — ‘notre chambre’ (our room) — seems to attach the we to the room, rather than claiming the room as ours. Deictic, pointing gestures are, with the exception of the ‘y’ in the stanza’s tenth line, absent. Aromas mingle; reflectivity remains diffuse and blurred.3131That impersonal and constative rhetoric, together with the emphasis on aromas and fragrances, recalls Baudelaire’s famous poem ‘Correspondances’. Paul de Man’s by now classic reading emphasizes that ‘Correspondances’ portrays a space that the subject cannot ‘transport’ itself out of. ‘Correspondances’ ends on a list, while stanza 2 of ‘L’Invitation au voyage’ begins with a list. See Paul de Man, ‘Anthropomorphism and Trope in the Lyric’, in The Rhetoric of Romanticism (New York: Columbia University Press, 1984), pp. 239–62. The fragrances of even the ‘most rare flowers’ merge with the prevailing scent of amber. Amber or incense is itself a mix of different fragrances. It can absorb additional fragrances, no matter how rare, without turning into anything different. And whose native language is being spoken there (‘y’), secretly, to the soul? The French possessive adjective ‘sa’ takes its gender from the feminine noun ‘langue’. We therefore cannot tell if that language belongs to ‘everything’ (tout) or to the ‘soul’ (âme). Much like the scent of rare flowers is absorbed into the mix of amber, a soul vis-à-vis Baudelaire’s poem will be absorbed, unable to differ from the poem’s language. Baudelaire’s poetics is assimilative: it adds us to the room, rather than allowing us to make the room ours. The language it whispers to our souls ‘en secret’ sets out to become our ‘native’ one by preventing our birth as self-aware subjects.

Lines 15–26: Millay

Rather than with a list of objects, Millay’s version of stanza 2 begins with a grammatical subject in the first person plural: ‘We should have a room […].’ To render the conditional, she could have chosen ‘would’, but opts instead for the form that articulates the desire of this subject: ‘we should have a room’, ‘tables […] should reflect’, and ‘all should speak apart’.

Baudelaire’s furniture, polished by the ‘years’ (ans), gleams with the accumulated sheen of canonical texts burnished by the (supposedly impersonal) ‘test of time’.3232This is the type of ratifying repetition Berman envisions. In Millay, by contrast, the polishing is done by a much more corporeal ‘palm’, as the scale shrinks from ‘years’ to ‘hours’ — from the authority of the ‘ages’ to the lived moment in which the text begins to shine under the gaze of a reflexive individual reader. That those hours have ‘vanished’ further underscores that what makes this surface reflexive is the transitory but embodied presence of a human life. Where Baudelaire has unspecified ‘furnishings’ (meubles) that reflect scattered light (‘luisants’), Millay has ‘tables’ — horizontal surfaces that ‘should reflect’ not just anything, but precisely the ‘rare flowers’ of self-aware, critical reading. Baudelaire’s mirrors are ‘deep’ (‘Les miroirs profonds’), but they lack the one thing that would make them actual mirrors: a reflective surface. In Millay’s poem, the mirrors are ‘deep as thought’,3333Johnson, ‘Poetry and its Double’, p. 26, notes the absorptive power of Baudelaire’s poem as a ‘mirror’ when she writes: ‘What is proposed to the woman is a place created in her own image’, and ‘the lyrical voyage, then, is a voyage through the looking-glass, a voyage into the illusory “depths” […] of a reflection’. acknowledging that what adds ‘depth’ to readable surfaces is our readerly thinking. Lines 24–25 answer Baudelaire’s subliminal strategy of assimilation with another assertion of distance: ‘All should speak apart | To the homesick heart.’ In Millay, the speaking is ‘apart’, separate from the heart, which is why the latter is ‘homesick’: it longs to return, self-reflexively, from language and representation back into itself. Precisely by maintaining itself ‘apart’ from representation, however, an indexically conscious reading can place the resignifying sheen of reflexivity and allegory over the found text — and thus turn what speaks to that homesick heart into the latter’s ‘own dear native tongue’. There is no doubt that Millay’s retranslation is highly conscious of itself as a translation throughout. In addition to radically rewriting the poem’s poetics, therefore, with this line Millay also signs as a translator who brings the poem into her own ‘dear native tongue’: English.

Baudelaire’s stanza, then, presents itself as a lavishly decorated and endlessly suggestive ‘room’ whose absorptive power is owed to its careful avoidance of even the slightest acknowledgement of our readerly presence. In Millay’s translation, by contrast, the stanza begins with a first-person pronoun, reflections appear on surfaces, polishing occurs in experienced, human time, thought attends to the mirror of language, and the ‘homesick’ heart keeps itself ‘apart’ from language. Millay’s text not only supports but calls for (‘should’) the very thing Baudelaire is blocking: a self-reflexive actual reading that can break with the absorptive power of prefiguration by returning into — and thus giving birth to — itself. Each of these trajectories reaches its respective logical conclusion in the third and last stanza.

Lines 29–34: Baudelaire

Baudelaire’s third stanza begins by telling us to ‘See […] | These vessels sleep.’ What vessels are these? And what does it mean to ‘see’ something when we read? If we assume that the poem ‘paints a picture’ for us, we will conjure a vision of a harbour in the evening light, ‘see’ merchant ships floating on canals, carrying treasures from the end of the (colonized) world. We may even go so far as to situate that scene in a specific place: ‘Clearly, the poet is inviting his beloved to a journey with him to an idealized, almost imaginary Holland.’3434James S. Patty, ‘Light of Holland: Some Possible Sources of Baudelaire’s “L’Invitation au voyage”’, Études baudelairiennes, 3 (1973), pp. 147–57 (p. 147). Patty also gives a useful overview of various scholars’ efforts to track down both visual and literary sources that Baudelaire may have drawn on in this poem. Many would place such hyperphantasia high on their list of reading pleasures. Those with more abstract inclinations, by contrast, might give those lines an allegorical, self-referential turn. Such a reader might, for instance, read the poem itself as such a ‘ship’: setting out from nineteenth-century Paris, it has travelled great geographical and temporal distances, to return enriched by ever-new readings. Both of those modes of reading — and many others — are, of course, eminently possible. But they have one fundamental limitation: by taking their lead from what the poem appears to ‘represent’, they end up obeying what it prefigures.

Baudelaire’s poem from the very start engineers readerly absorption and loss of indexical consciousness. Its third stanza, however, pushes both the metapoetic and the prefigurative description of these mechanisms to dizzying new heights. It begins by explicitly encouraging the very mode of reading that will reliably collapse any readerly self-awareness: a reading that sees words as containers.

Canals (‘canaux’) are channels that contain and direct a flow. The word ‘vaisseaux’, much like its English cognate ‘vessels’, also connotes any enclosure capable of holding content — jars, ships, other containers, even blood vessels. ‘Vessels’ on ‘canals’, thus, are vessels on vessels. Baudelaire’s lines model a layered repetition of the same: a reading that sees the text as a container will itself be contained. Why? Because this assumption alone — that texts represent some pre-existing content that readers merely extract or consume — obscures the fact that texts are in fact irreducibly incomplete, and that reading is not receptive but interactive and generative. It obscures the fact that we always already make decisions about reading that way and not this. Thus, if in reading Baudelaire’s lines, we ‘see’ a bucolic harbour scene, we will be unable to realize that they also prefigure what will be the case if we follow their lead: we will all turn into sleeping vessels — it will be sleeping vessels all the way up.

How can we instead pair this text with a reading that insists on its difference from the text and thus can bring itself into play? We have seen that the first change Millay makes is to reinsert indexicals. By reflexively actualizing these, readers can establish their difference from the text and its prefigurations. So what indexicals, if any, can we find in Baudelaire’s third stanza?

In the first two lines, the same indexical occurs twice:

Which way do these pointers point? If we read them as pointing our gaze towards a representation, at ‘canals’ and ‘ships’ in an imaginary harbour, we will obey the order to ‘see ships’. If we read them reflexively, we may begin to wake up to the fact that ‘these’ words could also be pointing directly at us. But Baudelaire’s lines still confound us: even if we are trying to wake up to the indexical ‘these’, the poem still holds up the diagnosis of ‘sleeping vessels’ to us. A reader in the process of waking up to themselves is thus flatly denied acknowledgement by the poem itself. If they nevertheless manage to hang on to some nascent indexical self-awareness, a new perception of these lines emerges: they appear as a prefigurative textual model that thumbs its nose at ‘these’ readers. The poem prefigures how ‘these’ readers obey, are contained, and remain asleep, unable to fully wake up to their own presence. The poem tells its readers that they cannot escape its spell.

How do we get our reading beyond this impasse? Millay has a complex and layered answer. But before we go there, let us briefly reconsider the notion of self-referentiality in literature. It is most widely understood as the idea that texts can ‘refer to themselves’, as a writing about writing. I propose that this is insufficient in several ways. It must be supplemented, first of all, with the idea that literary texts also refer to and even prefigure reading. But even more importantly, we must realize that it is actually not the text but we ourselves as readers who have the final say, at any given moment, of what we have a text refer to. And even more importantly: readers can use texts to reflect their own embodied readerly presence. The strongest way to do so is what I have called the self-reflexive actualization of indexicals. Indexicals can serve as the fulcrum around which we can turn our perception of the text from a ‘container’ of meanings into a shiny surface activated by and responsive to our presence and agency. Only from the vantage point of such a self-reflexive reading do the prefigurative dimensions of the text emerge and become readable. Indexicals do not represent, they do not prefigure anything. But they can function as a crucial fork in the path of our reading: we can use them to unmoor our minds from the text’s prefigurative grip.

So far, Baudelaire’s poem has either studiously avoided indexicals altogether or mapped them onto the ‘drama of loving relationship’ (Grossman), aligning all mechanical iterations with one perspective (‘mon esprit’) against another (‘your eyes’). The third stanza takes the issue to a new metapoetic level by confronting the problem head-on. I propose that we refer the ‘ces vaisseaux | Dont l’humeur est vagabonde’ to the very indexicals or deictics themselves whose potential Baudelaire’s poem sets out to tame. Indexicals, also called ‘shifters’, are indeed words whose references shift and wander. Yet their vagabond inclinations are activated not by the indexicals themselves but by us as readers and iterators. Their potential to disrupt the workings of a poetic apparatus such as Baudelaire’s poem, therefore, remains under control for precisely as long as we as readers can be convinced to make a habit, a very, very firm habit, of considering texts as containers: ‘See these ships …’ For the apparatus to tame the ‘humeur vagabonde’ of indexicals, and for readers to remain moored in the harbour of the text like so many vessels, therefore, ‘these’ readers — i.e. you and I — must remain asleep.

In this last stanza, therefore, Baudelaire pairs indexicals with a prefigurative/antinomic representation. Even if these readers are beginning to wake up to themselves, they will find that the poem decrees them as still sleeping. That is the poem’s last-ditch effort to bring readers back from what otherwise would be the end of its world. The verb form ‘viennent’ is ambiguous. We can read it as a simple statement in the present indicative. We can also, however, read the expression ‘qu’ils viennent’ as an exhortation and expression of desire — ‘let them come’, ‘may they come’. Please, colonized readers, do come back from the end of the world, and please remain asleep, oblivious to your own labour, your decisions, yourselves. Please keep attributing all of the readerly surplus value you produce to the poem, the author, the work! In that case, ‘viennent’ would be a subjunctive, and the veneer of factuality surrounding the return of ‘these vessels’ would fade, revealing the desire that informs Baudelaire’s entire poem: to overcome the ‘humeur vagabonde’ not just of indexicals but of your and my own thinking minds. The poem, then, both prefigures and theorizes — right under the noses of its readers — how their loss of readerly consciousness is (to be) achieved. We just need to learn to read it.

Distinctly different readings of these lines are thus possible. We can visualize these as layers, stacked one on top of the other in order of ascending self-reflexivity and awareness — and hence of increasing readerly emancipation from the authority we are used to conferring on such canonical texts. We generate this ladder of metalevels as we climb it:

1. The first layer would be a reading that is a mere inert iteration. It puts up no resistance, lets itself be scripted by the text, obediently performs a mere mechanical repetition of the same: sleeping vessels on canals. (The poem says the vessels are sleeping. Sleeping vessels it is.)
2. The next layer would be a reading self-aware enough to wake up to ‘these’ indexicals and thus to begin to perceive that prefigurative mechanism — and possibly the entrapment it constitutes for readers. (Oh, the poem points at these. Is it pointing at me? I am awake. But wait, it says sleep! Whoever believes that will not wake up …) To this realization, different affective responses are possible:
   1. Irony or mockery at seeing the credulous readerly masses fooled — unable to wake up to the fact that their predicament is explained to them right before their eyes.
   2. Clandestine triumph at discovering those strategies and the power they confer. This response can lead someone to consciously deploy and thus perpetuate these strategies of control.3535Walter Benjamin’s ‘On Some Motifs in Baudelaire’ is an example of this.
   3. Finally, this realization can also lead readers to react with growing resistance and ultimately an aversion — a turn away from absorption and into a more self-aware reading.3636See Gölz, ‘Apostrophe’s Double’.
3. Such a self-aware reader, finally, may conclude, as Millay does, that the offending poetic strategies themselves have to be rethought and then rewritten.

Lines 29–34: Millay

Millay consistently figures reading as a separate layer, interacting with but distinct from the text. This is important in principle, because only a reading that finds a way to assert its separate existence from the text can also engage the latter as an interactive matrix and apparatus, and thus make strategic decisions regarding its design. In the translation of the first half of stanza 3, these layered repetitions manifest themselves very concretely. Millay published her translation in 1936. But at some later point, right underneath her own earlier translation, she pencilled in a retranslation of the first six lines of stanza 3.3737See Joan St Clair Crane, ‘Edna St. Vincent Millay’s Afterthoughts on the Translation of Baudelaire’, Studies in Bibliography, 29 (1976), pp. 382–86. She thus created a little petri dish for us to watch her retranslate her own earlier translation. A closer look at the two different versions shows that they can be precisely situated on the ladder of metalevels I have just sketched.

Millay’s first (published) translation of those six lines reads as follows:

In Baudelaire, the vessels were merely sleeping. In Millay they are stuck — moored, tied down. Their travels are over, their voyage is history. The word ‘vagabond’ has disappeared entirely. The next three lines are tinted by irony or even sarcasm: all that journeying, all that bombastic effort to cross and overcome the ‘furious deep’ (of mirrors? of thought?) has failed to deliver the promised boatloads of exotic treasure, producing instead a mere ‘trifling thing’. Even the rhymes ‘past’/‘fast’ and ‘bring’/‘thing’ chime in and parody a repetition that fails to produce any significant difference. By latching on to the most obvious reading, those readers/ships — ‘to their moorings fast’ — have gone nowhere in a hurry.

In its earlier version, then, Millay’s translation articulates the second of the three reactions I mapped out above: the spell of absorption has already been broken, an aversion is developing, but the translation still gets stuck in mere irony or mockery. It runs into the impasse Baudelaire scripts for his readers.

Only in the retranslation, inserted by hand below the earlier printed translation, does Millay succeed in reshaping this most highly reflexive moment in the poem in a way that goes beyond diagnosing the problem. Only the retranslation reclaims its vagabond inclinations — and invites us to do the same (Figure 1):3838A small but significant formal feature registers the addition of another layer in this retranslation: only in this retranslation does the syllable count break out of the strict pattern of Baudelaire’s stanzas (four pentasyllabic rhyming couplets, each followed by a seven-syllable line) by adding a syllable to all lines except for the first. The fact that reading and translation adds itself to the original is thus registered by the added syllables. Millay discusses the syllable count in her preface to Baudelaire, Flowers of Evil, trans. by Dillon and Millay, pp. v–xxxiv. Figures 1 and 2 are from the copy of the book in the University of Virginia Libraries, pp. 77 and 74 respectively.

The first thing Millay’s line ‘See there on these streams’ does is reorient the indexicals to register a reflexive turn. ‘See there’ distances the text from our here. With that distancing, this reading unmoors itself from the representations offered in the text. It becomes fluid and uncontainable, as ‘these streams’ of readerly awareness reflexively wash over the text, reopening it to change. Millay’s first line no longer prefigures channels and vessels. Instead, the indexicals register the turn into reflexivity (from ‘there’ to ‘these’) and reopening of flow — ‘these streams’. The word ‘fleet’ connotes an accumulation of many vessels rather than individual ‘big ships’. It thus underscores the fact that assimilated readings tend to cluster — they can be herded into predictable places — while also hinting at their mercantile and even military functions. But the word ‘fleet’ also recalls the fleeting nature of what such readerly accumulations actually are made of. This mighty ‘fleet’ actually has nothing solid about it. It might dissipate at any moment. The adjective ‘errant’ similarly invokes the wandering that Baudelaire’s verses aimed to neutralize, while also hinting that the readings that have ‘come to berth’ and given up their mobility may have made a mistake, an error. The fact that the errant fleet is ‘laden down with dreams’ emphasizes that assimilated readers are, ultimately, trapped by their own ‘dreams’ — their obedient visualizations reinforce a (lack of) imagination and self-awareness. If we read the poem aloud, however, we can hear in the word ‘berth’ the homophonous possibility of a new ‘birth’ — the moment when the homesick heart returns into itself and succeeds in converting the found language into its ‘own dear native tongue’.

The two versions, Millay’s printed translation and her handwritten retranslation, then, climb the ladder of increasing readerly reflexivity. Her earlier printed translation still remains, however resistantly, caught in the diagnosis of entrapment. Only her handwritten retranslation takes us to the next level and recovers its vagabond humour.

Lines 35–40: Baudelaire

Baudelaire begins the final section of the third stanza with a dash. According to French punctuation rules, this may signal that there is a new speaker.3939I would like to acknowledge Marie Culpepper, who made this important point in my seminar on Baudelaire in the spring of 2020. The description that follows is completely factual and constative, as if from no perspective in particular. The setting suns ‘re-dress’ (revêtent) fields, canals, and town — they place a layer of ‘hyacinth and gold’ over the landscape. And the whole world goes to sleep:

With those lines, Baudelaire’s poem has arrived at its telos: the whole world’s going to sleep (‘s’endort’) is presented as a ‘réalité accomplie’.4040Brugmans, ‘L’Invitation au voyage’, p. 6. ‘Couchants’ is an adjective formed of the verb coucher. When it modifies the word soleil (sun), it means to ‘set’ — ‘the setting suns’. But there is also a transitive version of the verb coucher that means ‘to put to bed’. Coucher les enfants, for example, means ‘to bring the children to bed’.4141Couchant, finally, can also mean ‘crouching’ — as a dog or lion might. The reflexive version se coucher means ‘to go to sleep’. We can thus also read Baudelaire’s ‘soleils couchants’ as ‘suns’ that put those canals, fields, town, and finally the entire world to sleep. But what are those suns? I would suggest that they are the various elements — both metaphoric (prefigurations) and metonymic (indexicals) — that serve to prefigure, align, and orient the perspectives of readers. A new layer blankets or ‘re-dresses’ the canals, fields, and city. That layer is ‘d’hyacinthe et d’or’. Of course, we can just read this as describing the colours of the evening light. But Hyacinth, in Greek mythology, is Apollo’s lover and thus a figure associated with male–male desire. ‘Gold’ can be read as a gesture towards the function of the othered feminine as object of exchange and universal equivalent. ‘Hyacinth and gold’, then, neatly sums up the two-pronged and gendered interpellative strategy of Baudelaire’s text, which instals both assimilative and apotropaic traffic signs in the poetic ‘fields’, thus promoting the absorption of reading by polarizing the readership: the ‘sister’ is offered as a universal equivalent (gold) that facilitates the unselfconscious alignment with male homosocial desire.4242See Eve Kosofsky Sedgwick, Between Men: English Literature and Male Homosocial Desire (New York: Columbia University Press, 1985). With the last six lines, the poem declares its trajectory completed, its goal achieved: the whole world, everyone (tout le monde also means ‘all people’, ‘everyone’) has been put to sleep.

But why does Baudelaire mark off these last six lines as spoken by a new speaker? Who is that speaker? The poem’s concluding lines are devoid of any indexicals, personal pronouns, or possessive adjectives. With the exception of the initial dash itself, all other traces of this new speaker are thoroughly effaced. The last six lines of the stanza are wrapped in an aperspectival, flat factuality. Who is it that could confirm that Baudelaire’s strategy worked, that tout le monde has indeed lost consciousness, that everyone has in fact gone to sleep? Clearly, this confirmation can come neither from the poem itself nor from its implied ‘speaker’. Rather, the confirmation must come from a reader or other iterator. But surely, in order to confirm that the entire world has gone to sleep, that person would have to be awake? As a matter of fact, they must not. What they have to do, rather, is accept the script the poem hands them, and repeat it somnambulantly, mechanically, unselfconsciously.

Lines 35–40: Millay

Millay, too, opens this last section of stanza three with a dash. In English, this horizontal mark does not signify a new speaker. Instead, the dash acquires significance because, in Millay’s poem, it is a repetition. We recall that Millay added a dash in stanza 1 (at the end of line 3). This second dash now, together with the earlier one, creates an interior frame within the poem. The first three and last six lines are outside of that inner frame, as transitional zones surrounding the interior framed by the two dashes. It is in these transitional zones that personal pronouns appear especially prominently (‘me’ and ‘you’ at the beginning, ‘we’ at the end). If the first dash, marking the entry into the inner frame, was followed by the word ‘sleep’, the second dash (in line 35) is immediately followed by the word ‘now’, which punctures Baudelaire’s indexical-free, sleep-inducing sunset scenario with a wake-up call that brings us back into the present moment:

The descending ‘sun’ tints a ‘dyke’ — a barrier fortifying the canals — as well as town, field, and canal. The word also adds the feminine counterpart to the male homosocial implications of the word ‘hyacinthe’, restoring a gender symmetry that was missing in Baudelaire. But the final critical moment does not arrive until the last two lines of the poem, in which a first-person, actualizable, and ungendered ‘we’ concludes: ‘All that we behold | Slumbers in its ruddy light.’ Here, someone is indeed awake. Millay’s last six lines present the landscape not (as in Baudelaire) from a quasi-objective non-perspective, but rather as seen by a ‘we’ that ‘behold[s]’ the text and perceives that slumber. This ‘we’ includes, in every iteration, this reader, the one who is reading or citing the poem now, and who hopefully has agreed to join Millay’s lead and my own, as we have been thinking along with and re-minded her text. What sleeps are ‘all things in sight’ and ‘all that we behold’ — not we ourselves. We, rather, are now departing from the slumbering world of the poem, leaving it behind for other readers to re-mind it and wake up to themselves.

Barbara Johnson also comments on that dash in Baudelaire. She does not invoke its function in French punctuation — that of introducing a new speaker — but interprets it as a mark of an impersonal movement: Jacques Derrida’s notion of ‘spacing’. Writing, as continued différance, indeed cannot arrive in any specific present:

In spite of the demonstratives (‘ces canaux’, ‘ces vaisseaux’) and the present tenses (‘ils viennent’, ‘le monde s’endort’), the trip’s end-point seems curiously missing. More curiously still, this eclipse of the end-point is inscribed as such in the text by the use of a dash […]. If the poem’s language is thus organized around its disappearance, that disappearance turns out to be not an asymptotic limit external to the text — its end or origin — but its own necessary and inherent discontinuity, the very principle of its spacing, its articulation, and its rhythm.4343Johnson, ‘Poetry and its Double’, pp. 39–40; emphasis in original.

Johnson’s reading of the dash as ‘this eclipse of the end-point’ precisely articulates the effect of Baudelaire’s poetics: readers repeat mechanically, deferring to the text, and eclipsing awareness of their own presence and actions. To any such self-effacing reading, ‘spacing’ will continue to appear as the impersonal ongoing diffèrance of writing in general. But in fact, it is Johnson’s reading itself that enacts this ‘spacing’ by displacing and eliding another possible reading, one that could have acknowledged this act of repetition in its irreducible singularity and actuality. Johnson’s text articulates this, too, with amazing precision: ‘The truly unreachable utopian place, the place which is par excellence unknowable, is not some faraway mysterious land, but the very place where one is.’4444Ibid., p. 41; emphasis in original. By stopping just short of using an indexical such as here or I to mark this place, Johnson obeys Baudelaire’s prefiguration, which systematically withholds open indexicals that would invite readers to come to — to wake up to themselves.4545For a related earlier argument, see Sabine I. Gölz, ‘One Must Go Quickly from One Light into Another: Between Ingeborg Bachmann and Jacques Derrida’, in Borderwork: Feminist Engagements with Comparative Literature, ed. by Margaret Higonnet (Ithaca, NY: Cornell University Press, 1994), pp. 207–23.

Baudelaire’s poem works to create aura, that is, the appearance of distance, by convincing its readers, no matter how close they may be, to send their imaginations ‘là-bas’, to some faraway place. We are to forget that we are here, repeating, translating, and, if need be, rewriting the poem right before us. The dash is thus read in two incompatible ways: either it marks the ‘curious’ and unfathomable ‘eclipse’ of the very reading which itself enacts this ‘spacing’, over and over, by virtue of remaining a mystery to itself; or, alternatively, it can alert us to the appearance of a new speaker — ourselves. The second reading is the conditio sine qua non for the development of a theoretical framework that could begin to describe the functioning of this metonymic shadow docket of mechanical ratifiers. For us to reclaim our own agency and to bring about change, therefore, we need to break with a central taboo of post-structuralist literary theory: we must learn to think the effects and potentialities of our own unrepresentable readerly presence. We are unrepeatable, but we do the repeating. The moored ‘fleet’ of normalized readers huddles in the centre of the normal distribution. But each one of us, again and again, can be the singular one who makes a difference.

Millay’s Retranslation of the Refrain

The refrain is the only other passage in this poem for which Millay added a pencilled retranslation. The published version rendered the refrain like this:

To my perception, these two lines feel strained and conflicted. ‘Restraint’ and ‘order’ are in tension with the claim of ‘luxury and voluptuousness’, which suggest excess and abandon. Even stranger, then, that the former should ‘bless’ the latter. That translation seems to testify above all to Millay’s still ongoing struggle with Baudelaire’s poem.

In the retranslation, the traces of tension have disappeared. If anything, its affirmation that all is in order ‘there’ is less qualified and more definitive than in Baudelaire. Where he has a privative ‘nothing but order’ (‘Là, tout n’est qu’ordre et beauté’), Millay’s retranslation affirms straightforwardly that ‘all is order’. And where Baudelaire had a more ambiguous ‘calme’ (which, given the troubled skies in stanza 1, may well be followed by a storm), Millay’s final version achieves ‘peace’ (Figure 2):

Millay jotted down the new version of the refrain next to the French original in the bilingual edition. We can speculate that the word ‘there’ in the second version of the refrain confirms the final success of her own retranslation and the ‘peace’ that her repair of the poem grants us.

Literature and Life Transform Each Other

Henri Meschonnic is a rare translation theorist who points out that how we read and translate is critically informed by our theories of language. They make us perceive texts in certain ways — and lead us to neglect other possible perceptions. Meschonnic goes so far as to argue that in translation (but also, we may add, in reading) the theory of language that guides our actions is ‘the major and even only problem’: ‘Without knowing it, when we think we are translating a text it is our own representation of language that we are showing, and that interposes itself between the text to be translated and the intention of the translator.’5050Henri Meschonnic, Ethics and Politics of Translating, trans. by Pier-Pascale Boulanger (Amsterdam: Benjamins, 2011), p. 57. The challenge of reading translations, therefore, ‘consists of recognizing which representation of language is at work’. Meschonnic specifically critiques the limiting effect of a theory of language that is centred on the sign: ‘if we think language is in keeping with the sign’, then ‘our whole attention [is] turned to meaning, since the sign knows nothing else’.5151Ibid., p. 58. What we overlook as a result, he argues, are what he calls ‘poetics’, ‘rhythm’, ‘voice’ — and, most importantly, the ways in which language and life transform each other.

The poem, I repeat again and again, because it must be said again, I define this as the transformation of a form of language by a form of life and the transformation of a form of life by a form of language. Hence a poem is the maximum relationship between language and life. But a human life.5252Ibid., p. 137.

Meschonnic is helpful because he is a rare theorist who draws attention to these processes of mutual transformation, and thus offers a starting point for a theoretical enquiry into how exactly this mutual transformation of language and life can be understood. To such an enquiry I hope to have contributed with this essay. As I have tried to show, Baudelaire’s poem is an apparatus that works towards having texts script life. Millay’s (re)translation revises his poetics to reopen the inverse possibility — for our life to transform language. If the importance of her achievement — and many comparable ones by other (predominantly, though not exclusively) women writers — seems to have gone unnoticed, this is precisely because the hegemony of that poetics of control extends to our theories of language and literature as well. For most existing critical approaches, her alternative poetics remains systematically unreadable. Our critical apparatus, too, needs to be repaired.