Loxias | Loxias 28 Edgar Poe et la traduction | I. Poe et la traduction 

Stephen Rachman  : 

Lost in Translation: Poe, Baudelaire and “The Purloined Letter”


Perdu dans la traduction.

En revisitant les lectures de “The Purloined Letter” d'Edgar Allan Poe par Bonaparte, Lacan, Derrida, Johnson et Irwin, cet article soutient qu'une erreur mineure dans la traduction de Baudelaire peut aider à recadrer le contexte critique pour l'histoire comme un modèle de signification. Plutôt que voir la lettre comme un symbole d'une signification absente ou différée, nous partons du principe que la lettre fonctionne comme une carte marquée dans un système de signification clos ou truqué.


Lost in Translation: Poe, Baudelaire and “The Purloined Letter”.

Revisiting the readings of Edgar Allan Poe’s “The Purloined Letter” by Bonaparte, Lacan, Derrida, Johnson, and Irwin, this paper argues that a minor error in Baudelaire’s translation can help to reframe the critical context for the tale as a model of signification. Rather than seeing the letter as a symbol of an absent or deferred signifier, the paper argues that the letter function as a marked card in a closed or rigged system of signification.


mots-clés : Poe , psychanalyse, traduction

géographique : Etats-Unis

chronologique : XIXe siècle


Texte intégral

I. The Destruction of Signs

1In the middle of Barbara Johnson’s exegesis of “The Purloined Letter” debate between Jacques Derrida and Jacques Lacan, she mentions an error in Charles Baudelaire’s seminal (that is, for French readers) translation of Edgar Allan Poe’s tale1. The error occurs when the detective Dupin, looking through his green spectacles, describes the specific location of the missing letter in the apartment of Minister D______. Poe writes, “At length my eyes, in going the circuit of the room, fell upon a trumpery filigree card-rack of pasteboard, that hung dangling by a dirty blue ribbon, from a little brass knob just beneath the middle of the mantelpiece.” This last sentence Baudelaire renders “suspendu par un ruban bleu crasseux à un petit bouton de cuivre au-dessus du manteau de la cheminée2.” The error consists of the tiniest of slips: “au-dessus du manteau” means above the mantelpiece when it should read “au dessous du manteau,” meaning beneath or below the mantelpiece. In this case, above and below happen to be paronyms in French.

2Johnson’s oft-referenced piece of literary criticism is dedicated to teasing out the vagaries of difference in all its telescoping complexity, and yet when confronted with an instance of simple linguistic differentiation – a mistranslation revealing the French frame of reference into which Baudelaire had placed Poe one hundred years before Lacan and Derrida, Johnson – though a capable translator of the French herself – has little or nothing to say about the error. Rather her frame of reference consists of the Gaullic infighting attendant to the notice of this error. She focuses on Derrida calling attention to the fact that it was Marie Bonaparte in her study of Poe who had first pointed out Baudelaire’s error and Derrida’s withering suggestion that Lacan’s argument deliberately conceals (and of course simultaneously reveals) its indebtedness to Bonaparte. Despite Lacan’s disparagement of Bonaparte (indeed, Lacan evidently dismisses Bonaparte in the course of his seminar as “a griller,” a kind of academic fry-cook), Derrida suggests that Lacan’s argument has more in common with the Freudian Bonaparte than even Lacan is aware.

3Rather, Johnson is more concerned with the stridency with which Bonaparte marshals an argument about the purloined letter as a sexual symbol, in particular it being symbolic of the Queen’s “maternal penis.” Johnson notes how Bonaparte finds fault with Baudelaire,

that Baudelaire’s translation … is “completely wrong.” Bonaparte’s frame of reference – the female body – cannot tolerate this error in translation. (134)

4Johnson implies that Bonaparte’s harshness (she is rather unforgiving of Baudelaire’s slip which could have been, after all, little more than a printer’s error) is symptomatic of her general interpretive rigidity. Taking her cue from Lacan, Johnson implies that there is something ham-fisted about this. Rather than viewing Bonaparte’s querulous posture with Baudelaire as symptomatic of her sense of disempowerment (you might even say castration) in the face of this masculinist critical mastery, Johnson seems to be saying that Bonaparte should lighten up and either get a more flexible frame of reference or a more flexible attitude towards all interpretive strategies. But one might counter this by considering that far from being intolerant, Bonaparte is adamant about the translation error precisely because, for one invested in a sexual symbology, the physical positioning of the letter does indeed matter. Bonaparte’s frame of reference euphemized by Johnson as “the female body” is more precisely the anatomy of the vulva and if the brass knob does not sit in a proper clitoral position relative to the vaginal opening of the chimney mouth, then the letter cannot serve as a surrogate phallus.

5But for Johnson, the issues surrounding Bonaparte’s interpretation are merely another pretext for the thrust-and-parry of Lacan and Derrida. “A note Lacan drops on the subject of the letter’s position,” Johnson explains, “enables Derrida to frame Lacan for neglecting to mention his references” (134-5). In her eagerness to expose Bonaparte’s critical intolerance and Derrida’s re-writing of the triangulation of “The Purloined Letter” with himself as Dupin restoring the Queen Bonaparte’s note from the Minister Lacan who has stolen it, Johnson only indirectly returns to the question of whether or not Baudelaire’s error matters. For her, it is more a question of Derrida’s “framing” of Lacan and Lacan’s phallic or phallogocentric interpretation of the letter. For Lacan, the exact physical position of the letter does not matter because he believes that its signifying power is indestructible, indivisible and unstoppable. To quote in translation the last words of the seminar: “Thus is it that what the ‘purloined letter,’ nay the ‘letter in sufferance,’ means is that a letter always arrives at its destination” (39)3. Derrida argues that the position of the letter should be crucial from the point of view of Lacan’s seminar (that is, Derrida implies, if Lacan truly understood the stakes of his own position), even though Lacan denies it. Indeed, thirty years after the publication of the “The Purveyor of Truth” [“Le facteur de la vérité”], Derrida continued to insist that Lacan’s sense of the indestructible letter, “the materiality of the signifier,” was deduced “from an indivisibility that is nowhere to be found4.” For Derrida the materiality of signifiers (and signification) is always divisible and subject to rupture, interpenetration, and differentiation; this leads Derrida to find Lacan’s psychoanalytic project to be overdetermined (overdetermination is what Derrida subtly suggests Lacan ultimately shares with Bonaparte).5 Johnson tends to side with Lacan by emphasizing Bonaparte’s overdetermined psychoanalytic interpretation (it deals with symbolic anatomy as opposed to Lacanian allegories of signification, or what Derrida would call its phallogocentrism) and tellingly points out the ways in which Derrida contradicts himself in making this argument.

6If any account of this terrain always feels like one is walking in on Sam Spade, Bridget O’Shaughnessy, and Caspar Gutman in the latter stages of negotiations about The Maltese Falcon, then that is because Poe’s tale established the modern paradigm (which, as it happens, Dashiell Hammett and John Huston followed) of the hermetically sealed fiction of cross and double-cross in which spirited antagonists pursue a prized artifact of dubious or uncertain value. That is to say, what Johnson calls difference in the act, a paradigm of signification in which uncertainty prevails to the extent that one cannot be certain of the most matter-of-fact claims, is actually a different sort of paradigm related more to game theory and to the sociology or social psychology of games. As John Irwin summarized Johnson’s take on Lacan and Derrida: “The commitment to an increasingly self-conscious analytic posture that animates this cumulative series of interpretations produces at last a kind of intellectual vertigo, a not uncharacteristic side effect of thought about thought – the rational animal turning in circles trying to catch itself by a tale it doesn’t have6.” What interests me here are the ways in which the recognition of and subsequent dispensing with a concrete error, an undeniable mark of difference, leads to a series of abstractions about analytical abstraction.

7And the question remains: is there a relevant frame of reference in which Baudelaire’s error in translation matters for contemporary criticism and/or its understanding of Poe? Of course, the error matters to Bonaparte because she evidently needed to assert her own authority over the French text and Baudelaire and his pre-Freudian point of view were clearly in her way. Also, given her genital/anatomical interpretation of the room, the position of the card rack is more than relevant; it is crucial.

8But this begs the question, besides Bonaparte’s narrowly anatomical one is there another frame of reference in which Baudelaire’s error becomes significant? In a sense, the answer lies in our understanding of how we address the question of translation as a mark of difference.

9Translation errors are not the kind of difference in which Johnson is particularly interested. Fans of The Critical Difference may recall that the book’s epigraph from Paul De Man’s Allegories of Reading discusses televisions Archie Bunker dismissing one his wife Edith’s inquiries. She wants to know whether or not she should lace his bowling shoes under or over, to which Archie replies irritably, “What’s the difference?” What Archie means, as De Man explains, is not a request to learn what the actual difference is, but “I don’t give a damn what the difference is.” De Man points out that “the grammatical pattern engenders two meanings that are mutually exclusive: the literal meaning asks for the concept (difference) whose existence is denied by the figurative meaning” (v). For Johnson, the translation issue is, like the pattern of Archie Bunker’s shoelaces, trivial in and of itself; it only grows in interest to the extent that it is a pretext for yet another proliferation of the interpretive struggle enacted by “The Purloined Letter.” If the most critical differences of all are constituted by the impossibility of knowing, as Johnson writes, “whether something constitutes a description or a disagreement, information or censure,” then Baudelaire’s error does not rise to the level of a critical difference because it is an error of fact at the level of description over which easily distinguishable forms of disagreement, information and censure can be made (Poe’s text says below, not above). Furthermore, because Johnson is committed to asserting that literary origins are infinitely referable to some prior text (“The Purloined Letter” begins with an unlocated quotation and ends with a reference to Crebillon’s Atree), Baudelaire’s error lacks interest to the extent that it assigns a definite point of origin to the differential.

10Johnson’s commitment is to a theoretical frame of reference that simultaneously traces Derrida’s poststructural attack on the structural psychoanalytic criticism of Lacan while ultimately defending Lacan as having arrived at the same position as Derrida with respect to signification. Translation issues become part of the blindness that of necessity frames her interpretative insights. For Johnson, the purloined letter, not the tale per se but the letter itself, “as a signifier is thus not a thing or the absence of a thing, not a word or the absence of a word, not an organ or the absence of an organ, but a knot in a structure where words, things, and organs can neither be definably separated nor compatibly combined. This is why the exact representational position of the letter in the Minister’s apartment both matters and does not matter” (141). For Johnson, the position of the letter only matters in its symbolic dimension so that Dupin can retrace/untie a symbolic knot. To emphasize translation above/below calls attention to the discrete, nay, simple referentiality and indexical power of signs and signifying positions that runs counter to the drift and jouissance of her argument. Because Bonaparte stridently criticized Baudelaire in the service of her literal anatomical psychoanalytic interpretation, Johnson in a sense places this concern for translation to one side as being overly literal.

11There is an irony in that the missing “o” in Baudelaire’s translation might be construed in a playful way as a literal “purloined letter,” stolen, as it were, in the act of translation, or perhaps less preciously, it is lost in translation. In terms of the explicit thematic of this symposium, it is the mark and inevitable cost incurred whenever two languages meet. The presumably inadvertent erasure of the “o” marks, in the words of Genette, “the inclusive relation which links each text to the various types of discourse which it belongs to7.” Thus, this kind of error in translation – while in and of itself a minor mistake – does indeed matter. It is, first of all, linguistic variation or, if you will, mutation of the kind that occurs whenever language travels, and so Baudelaire’s error is a sign of horizontal transmission, as Luigi Cavalli-Sforza describes the process of cultural and linguistic dispersal, at the moment of Poe’s introduction to French language and culture8. Baudelaire’s translation error is a sign of another frame of reference pertinent to Poe’s transnational identity and to the theories of signification his text has spawned in France and in their return to the United States in Johnson’s criticism. Of course, whether or not the exact position of the letter matters to any given interpretation of the story depends upon one’s frame of reference, as Johnson argues, but if we were to be plain about it, the exact location and hence an error in translation concerning that location does not matter to her because it does not matter to Lacan or his theories of signification.9

12I argue that the translation error matters as a sign of difference and a mark of other kinds of difference but these signs do not rise to the level of a critical difference because whether or not the letter is found above or below the brass knob does not in any way change its visibility. The notion that it is visible to Dupin and withheld from the reader’s vision is the letter’s crucial condition. Baudelaire may have placed the letter in the wrong place, but above or below the mantelpiece, the letter in the card rack remains, for Dupin, in plain sight. For my reading of “The Purloined Letter,” the error in marking the position of the letter reinforces a number of ideas about signification, or more specifically, what I would call closed-system signification, conventional signification, or rigged signification. I suggest that Baudelaire’s translations raise general questions of slippage and point to a specific site of erasure in “The Purloined Letter,” namely a language game derived from card games, cheating, and detective fiction. On the contrary, the purloined letter can be located by Dupin not because it is a knot in a structure that Dupin disentangles through repetition, but because it functions like a marked card – even when the mark has been altered – and its content, though never revealed to the reader, can be read to the extent that it is linked to it markings. It may be that Baudelaire’s translation may have aided in re-directing the jouissance of Poe’s signifying strategies away from cards toward the other symbolic frames that have come to dominate the tale’s theoretical importance. What has been lost in translation then is not only the letter “o,” or a debate about the relative values of post-structural and historicizing readings of “The Purloined Letter” but a debate about when a story can be used as evidence of a fundamental pattern of signification and when it offers a different kind of theoretical opportunity.

II. Questions of Translation

13In The French Face of Edgar Poe, Patrick Quinn assessed the ways in which Baudelaire’s translations transmogrified Poe’s English. In some cases, Baudelaire’s French improved or even corrected Poe’s syntax or diction; in others he mistranslated certain expressions. Quinn determined that details “may have been overlooked or improved or weakened in translation. But there is no full-scale transmutation. Baudelaire did not melt down these stories, remove their dross, and recast them in the pure gold of his French10.” Quinn engages the common transcultural question that surrounds Baudelaire’s efforts – that he lent Poe an elegance and sophistication not present in the original, thus creating the myth of the French Poe and infusing his texts with the status and value that would create the conditions for the battle of mastery between Lacan and Derrida. But Baudelaire’s translations always pose a number of connotative problems. First there is the title. “The Purloined Letter” becomes the “La Lettre Volée.” “Volée”, which is commonly construed as stolen is certainly adequate but it lacks purloin’s connotations of pilfering or filching, a kind of theft that takes place, as the OED points out, “under circumstances which involve a breach of trust.” It also has no way of conveying its origins in an Old French word that meant to put something far off, to get rid of. How can one convey in a target language an inflection in a source language that is signified by the presence of the target language? Is there a French word for stolen that has the same connotation that purloined has for English speakers?11 Suffice to say that in the very title itself there is an inevitable stripping away of connotation in relation to the nature of the crime.

14What other linguistic effects might be effaced by Baudelaire’s translation? By way of conclusion I wish to consider in greater detail, the passage with which I began about the positioning of the card rack.


“At length my eyes, in going the circuit of the room, fell upon a trumpery filigree card-rack of pasteboard, that hung dangling by a dirty blue ribbon, from a little brass knob just beneath the middle of the mantelpiece. In this rack, which had three or four compartments, were five or six visiting cards and a solitary letter. This last was much soiled and crumpled. It was torn nearly in two, across the middle – as if a design, in the first instance, to tear it entirely up as worthless, had been altered, or stayed, in the second. It had a large black seal, bearing the D_____ cipher very conspicuously, and was addressed, in a diminutive female hand, to D_____, the minister, himself. It was thrust carelessly, and even, as it seemed, contemptuously, into one of the uppermost divisions of the rack.”


« A la longue, mes yeux, en faisant le tour de la chambre, tombèrent sur un misérable porte-cartes, orné de clinquant, et suspendu par un ruban bleu crasseux à un petit bouton de cuivre au-dessus du manteau de la cheminée. Ce porte-cartes, qui avait trois ou quatre compartiments, contenait cinq ou six cartes de visite et une lettre unique. Cette dernière était fortement salie et chiffonnée. Elle était presque déchirée en deux par le milieu, comme si on avait eu d'abord l'intention de la déchirer entièrement, ainsi qu'on fait d'un objet sans valeur; mais on avait vraisemblablement changé d'idée.  Elle portait un large sceau noir avec le chiffre de D... très en évidence, et était adressée au ministre lui-même. La suscription était d'une écriture de femme très-fine. on l'avait jetée négligemment, et même, à ce qu'il semblait, assez dédaigneusement dans l'un des compartiments supérieurs du porte-cartes.”

15What Baudelaire has partially effaced from Poe’s text is not so much verbal clumsiness but a level precision and a host of paranomasiac effects that point directly to card play as a motif. “Trumpery,” “card rack,” and “visiting cards” all point to the letter finding its home in a symbolic order but not in the phallogocentric anatomical symbolic space below the mantelpiece rather in the logic of card games or rather the logic of cheating at cards. In a story in which a minister or jack assists a queen in finessing another jack in order to trump the king, Poe has positioned the much abused letter in a rack that displays itself like a hand. The minister reads the mark on the letter. The Queen knows that the card is marked but does not want to give up the game. Poe presents us with the moral equivalent of a hustle, Dupin is offered a cold deck and palms the ace of spades. Like modern poker players who are loath to yield any visual cues, he dons his famous green spectacles. Baudelaire’s French mutes this, especially in its emphasis of the shabbiness over the fakeness and, I suppose, there is no way to convey the full connotative force of “trumpery-fillagree.”

16In a sense, the common critical frames of reference pertinent to “The Purloined Letter” have been skewed toward increasingly sophisticated acts of reading by literary critics, psychoanalysts, and analysts of analysis. Lacan concerns himself with reading the sign of the letter. Derrida, taking his cue from Poe’s tales, begins his commentary by observing that everything begins in a library, implying that literary perusal is the framing concern of all the Dupin stories and thus “The Purloined Letter.” Johnson and others (myself included) have observed the ways in which Poe frames his text with other texts, deflected acts of reading which point to a further range of texts so that reading appears, however far we may pursue it, like an ever retreating event horizon. This last framing fosters a sense of the act of reading as central to the concerns of the story (not peripheral) and a mode of analysis that creates the conditions of telescoping, intertextual referentiality.

17Within all of these frames, the exact position of the letter in the room is overshadowed by the condition of its partial destruction and that the tale makes inaccessible its specific verbatim contents. Dupin does not locate the letter by reading it (in the sense of deciphering its contents), rather he recognizes it by its markings – in particular the seal of Minister D_____ and that of the Queen (her handwriting). As John Irwin has shown, Lacan and Derrida play a game of evens and odds over the structure of “The Purloined Letter” (Lacan contends that the structure is triangular and Derrida that it is quadrangular while Johnson refuses to take a numerical position.) In this battle of one-upmanship, Irwin demonstrates that Poe’s text reveals the limits of critical argumentation. As Irwin summarizes the cultural work of the debates surrounding “The Purloined Letter,” “in its translation from fiction to criticism, the project of analyzing the act of analysis becomes in effect the program of being infinitely self-conscious about self-consciousness” (11). I want to add to this summation the suggestion that analytical self-consciousness is not the only arena of gamesmanship and that, in reworking Poe’s essential materials, Lacan, Derrida, and Johnson are playing a game of evens and odds over what might be termed a marked deck of cards. If the argument between Lacan and Derrida over the nature of signification engendered by Poe’s “The Purloined Letter,” turns on a question of whether or not letters always reach their destinations wherein Lacan insists that they do and Derrida that they do not, then the lesson of Poe’s tale and Baudelaire’s translation suggests several frames of reference in which they are both right, but not because the tale reflects the problems of signification they have in mind.

18Rather, the detective tale as Poe has conceived of it is a fixed form (a mystery constructed expressly, as he says, for the purposes of being unraveled) and a contrivance (more air of method than method), and the mode of signification it presents is closed rather than open, rigged rather than random. The letter at the center of tale, of which so much has been made, is a letter in name only; it functions more like a playing card doctored for purposes of cheating. In rigged systems such as a marked deck we need only see the back of the card to understand its value and so the letter however torn, may be indivisible from its markings as Lacan suggests. If however, an o is lost in translation, changing the position of the letter in space, then the letter’s necessary pre-condition for its identification – its location – is called into question, as Derrida suggests. The letter’s whereabouts become ambiguous in much the same way that its markings (in the original story) were altered or partially damaged, as if someone had made up their mind to destroy the card and then, mid-tear, thought better of it. As if an apparently meaningless frame of reference traveling at the speed of thought, suddenly became relevant, as Johnson suggests. In a sense all of these forms of signification are present in “The Purloined Letter,” one in the frame of the tale itself and another in its translational frame, both lost and found in translation.

Notes de bas de page numériques

1  Barbara Johnson, The Critical Difference: Essays in the Contemporary Rhetoric of Reading (Baltimore: Johns Hopkins University Press, 1981): 134. All subsequent references to this work will appear parenthetically in the text.

2  Edgar Allan Poe, “The Purloined Letter” The Collected Works of Edgar Allan Poe, vol. 3: Tales & Sketches II, ed. T. O. Mabbott (Cambridge: Belknap Press of Harvard University Press, 1978) 3:975. All subsequent references to this work will appear parenthetically in the text. Charles Baudelaire, “La lettre volée”, Histoires extraordinaires, Paris: Michel Lévy frères, 1856.

3  Jacques Lacan in The Purloined Poe: Lacan, Derrida And Psychoanalytic Reading Eds. John P. Muller and William J. Richardson (Baltimore: Johns Hopkins University Press, 1981): 39.

4  Jacques Derrida, Resistances of Psychoanalysis (Palo Alto: Stanford University Press, 1998): 60.

5  Derrida writes that this “materiality of the signifier” “always seemed and still seems to me to correspond to an “idealization” of the letter, to an ideal identity of the letter, which was a problem I had been working on elsewhere along other lines for quite some time” (60). Here Derrida attempts to debunk the Lacanian position suggesting that he and Lacan are working on the same problems, and that Lacan’s work is everywhere marked with traces of Derrida’s grammatological investigations.

6  John T. Irwin, The Mystery to a Solution: Poe, Borges, and the Analytic Detective Story (Baltimore: Johns Hopkins University Press, 1994): 11.

7  Gérard Genette, Introduction à l'architexte, Seuil, Poétique, 1979, p. 88.

8  Luigi Luca Cavalli-Sforza and Francesco Cavalli-Sforza, The Great Human Diasporas: The History of Diversity and Evolution (New York: Addison-Wesley, 1995): 212-14.

9  For Johnson, Poe’s mode of solution is not a specific piece of information yielded by the obtaining of a singular piece of data or a clue, but achieved through the reproduction of a sequence of entanglement, a mirroring rather than an undoing. Given that her metaphor is spatial – a knot – knowing where something is located might be relevant, even on a non-literal level.

10  Patrick F. Quinn, The French Face of Edgar Poe (Carbondale, IL: Southern Illinois University Press, 1957): 134.

11  An 1857 translation of the story by William L. Hughes offered “La Lettre dérobée,” suggesting “concealed” or “backdoor.”

Pour citer cet article

Stephen Rachman, « Lost in Translation: Poe, Baudelaire and “The Purloined Letter” », paru dans Loxias, Loxias 28, mis en ligne le 15 March 2010, URL : http://revel.unice.fr/loxias/index.html?id=6017.


Stephen Rachman

Stephen Rachman enseigne dans le Département anglais à Michigan State University. Il est le coéditeur de The American Face of Edgar Allan Poe et de l'auteur de beaucoup d'essais sur Poe. Il est actuellement le président de Poe Studies Association. Stephen Rachman teaches in the English Department at Michigan State University. He is the co-editor of The American Face of Edgar Allan Poe and the author of many essays on Poe. He is currently the president of the Poe Studies Association.