1Korean literature was only introduced relatively recently in France: the first translated work was published there in the late nineteenth century2, but it was only in the 1980s, and more so in the decade that followed, that books translated from Korean appeared (Choi, 2014, 68-88). Most of the works are novels and short stories, published primarily by Actes Sud, Picquier, Zulma, Imago, Decrescenzo, Philippe Rey, and very recently, Matin calme, a publishing house founded in January 2020 specializing in Korean crime novels. The rise in prominence of French translations of Korean literature benefited from the Korean government’s voluntarist politics, which implemented an aid system for translating texts into different languages3, encouraging translators to work in duos.

These translator pairs follow varied work procedures that are neither very standardized nor well studied. The objective of this paper is to nourish reflection on co-translation while highlighting the practice’s salient principles. To do so, I will examine my experience translating Pyun Hye-young’s noir novel 서쪽 숲에 갔다 (literally, Departed in the west forest), published in French under the title La Nuit du hibou (literally, The Night of the Owl) (2022), with Lee Tae-yeon. Pyun Hye-young (born in 1972) is a short-story and novel writer, and many of her works have been translated into English by different translators and with various publishers4. La Nuit du hibou is a novel with a rather complex narrative structure. It begins like a crime novel would: a lawyer, Ha-in, arrives in a village searching for his brother, a guard supposed to be stationed at the forest’s entrance. Ha-in interrogates the current guard, In-su, as well as various elderly people from the village, and one can see a dark plot begin to take shape. But at the end of its first part, the novel turns toward a structure more akin to a psychological thriller or roman noir, with a rather slow rhythm, the spotlight shining more and more on the new guard, In-su, an alcoholic. The reader is led to see through In-su’s eyes, doubting everything and no longer knowing what in the story can be attributed to the character’s madness – he suffers from hallucinations – or to the diegetic reality5. It is only gradually, and always with gray areas, that one can piece together the lawyer’s brother’s story, which involves the different characters introduced in the novel’s first part. The universe is dark, heavy. The motif of confinement is central throughout the story. The characters are confined within the village, their relationships, their madness or obsessions, or a past from which they cannot manage to free themselves; this confinement is likewise represented by the forest, a fascinating and terrifying place around which the author centers the plot, the characters falling into its trap. The work’s anchoring in nature, along with its motifs of domestic violence and alcoholism, make it original.

After presenting various general elements of co-translating Korean into French, I will analyze the negotiation process carried out with Lee Tae-yeon. As I am neither a specialist in translation nor in Korean language and culture, the analysis is predominantly based on a study conducted by Choi Mi-Kyung, an experienced translator in the co-translation field, conference interpreter, and teacher at the Ehwa University School of Translation & Interpreting in Seoul. Many key sources referenced in this paper are drawn from her thesis, entitled La Cotraduction : domaine littéraire coréen-français [Co-translation: the Korean-French literary field] (2014). The dissertation was defended at the ESIT School for Translation and Interpreting in Paris under the direction of Israël, an important figure in the Interpretive Theory of Translation. While I do not claim to specialize in this approach, I was immersed in the movement for family and professional reasons6. It is therefore likely that my proximity to the theory has influenced my way of working in co-translation, leading me, in light of my experience, to draw on some of its key notions. Given my discipline of specialty, stylistics, the examples employed pay special attention to linguistic choices and their effects.

1. Co-translating from Korean to French

Co-translation is a common practice in certain language pairs, particularly those including Korean. Usually, literary translators only translate into their native language, where their capacity of expression is optimal (which is what translation specialists and professionals call “translating into your A Language”, or “translating in A” for short). Yet there exist few people whose native language is French and whose mastery of Korean is sufficient to take on translating a work by themselves. The disproportion between editorial demand and the number of translators in A available therefore forces Korean speakers with French as a second language to undertake translations into French (“translating in B”).

This situation is an effect of dissymmetrical relations between languages and cultures on the international scale, which Calvet describes as a gravitational model (1987). It involves one central language (today, English), around which a dozen languages with medium importance (including French) gravitate, then about a hundred others with lesser importance on the international scene, then four to five thousand peripheral languages; these languages are connected by bilingualism, which manifests balances of power. As Choi puts it:

The attraction exerted by the more central languages explains the imbalance when it comes to available translators, who opted to learn a language more central than their mother tongue, and rarely a more peripheral language. In other words, the large majority of French people choose English and very rarely Korean. (2014, 20)

While Korean is only taught in a handful of French universities, in Korea, French is taught in about fifty7.

It is this situation that explains why nearly all Korean literary texts are (co)translated in B (Choi, 2014, 30); that is, by translators whose native language is not French. Generally speaking, if a translator has a level of mastery of comprehension in their B language that is equivalent to that of their A language, their level of mastery of expression will be inferior8. This is why translating in B, widely practiced in academic contexts, is discouraged (by UNESCO9, for example), criticized, and, when practiced for lack of a better option, unreliable:

[Reformulation] calls for a delicate dose of editorial elegance and information reproduction, the complete preservation of the information conveyed by the source discourse not always being compatible with the target language’s rules. This dose requires a good stylistic mastery in the target language, which one rarely manages to acquire in a second language10. This is why employers seeking high quality translations and translators setting similar norms strongly recommend that one works only toward their native language. This rule does tolerate some exceptions. The most common one results from a lack of translators with the desired language pair in the market in question. Thus, in China, Japan, and Eastern Europe, the quantity of work translated into English, but also into French and other languages, is such that Anglophone, Francophone, and other translators onsite are quickly overwhelmed, and economically speaking, the only viable solution is falling back on translators whose native language is Chinese, Japanese, Bulgarian, Czech, Russian, etc. (Gile, 2005, 123)

Given the numerous difficulties posed by translating in B, especially when dealing with literary texts, for example, where the notion of a translation’s “acceptability” does not have the same challenges as in technical texts (Choi, 2014, 43-54), one might therefore be tempted to think that this practice “is a last resort to overcome the lack of translators, or a task left to the more daring translators who might also be ignorant of its pitfalls” (Choi, 2014, 35). Working in conjunction with a reviser or co-translator whose native language is French partially enables one to overcome these difficulties while meeting editorial demand. While the involvement of a reviser might not in itself guarantee “the success of the translation operation… it does increase the probability of success when done according to good work methods” (Choi, 2014, 57). On the title pages of books translated from “rare11” languages such as Korean, one will quite often see two names listed.

In reality, however, co-authorship situations take on different forms. There exist many varied ways of co-translating, depending mainly on the involved translators’ fields of expertise and the roles assigned in the translation process. One can generally distinguish two modes of interaction, according to which role the target language speaker (in the present case, French) plays: reviser of the translated text or co-translator. In the case of the reviser, the target language speaker must correct and improve the translation in its acceptable but somewhat unfinished form: if they are a speaker of the source language, they inspect the text from a perspective of comparability with the original (bilingual revision). The reviser is more competent and experienced than the translator in the target language. This configuration assumes a hierarchical relationship, sometimes signaled on the title page by a mention such as “translated by X in collaboration with X” or even “translated by X and revised by X”. In cases where the target language speaker plays the role of co-translator, they participate fully in the translation process according to conditions decided by the collaborators, which can vary. It is this second procedure, in which I do not hold the position of editor but of co-translator, that my collaborator and I follow. In later sections of this paper, I will detail our co-translation system, sufficing to say for now that, despite having quite different skill sets, our work is shared in equal parts.

Lee Tae-yeon has expertise translating from Korean to French but does not master the target language enough to work alone, while my competence area is limited to French as I do not read or speak Korean. Due to our different profiles12, we would therefore be predisposed to take on the dissymmetric roles of translator and reviser. We chose, however, to distribute our roles differently, evidenced by the fact that our names are placed side by side on the title page (“translated from Korean by Lee Tae-yeon and Pascale Roux”), alphabetically according to our last names but also in an order relative to the languages (source and target). This situation is also reflected in the fact that we share our translation aids and pay equally.

2. The co-translation process

Co-translation methods are varied, in part due to material constraints such as the geographic distance between translators. Certain pairs work verbally, like Kim Kyunghee and Maryse Bourdin, who translate from Korean, or Ryoji Nakamura and René de Ceccatty from Japanese, which, according to Choi, “probably fosters spontaneity in the re-expression phase” (2014, 84)13. Other pairs communicate in writing, like Bellemin-Noël, whose knowledge of Korean is “less than rudimentary” (2012, 244) and who presented the steps of his work process with Korean co-translator Choe Ae-young in an article published in the journal Po&sie. His methodology is overall quite like ours: he works from a first version of the text sent by Choe Ae-young (which he designates as a “draft”, 2012, 245), then carries out an initial reformulation effort, exchanging a series of questions and answers with the translator before proceeding to the rewrite, which his collaborator must approve. Finally, in the revision phase, the objective is to make sure that the “literality has been respected” (246) before the “gueuloir”, the final task of reading the text aloud to feel “the music, the rhythm of the writing” (247). Lee Tae-yeon and I follow similar steps, but they are less clearly differentiated; in many respects, they are close to the process Choi specifies in her thesis (“Le processus de traduction en B [The process of translating in B]” 2012, 127-274), in which a translator whose native language is also Korean collaborates with a co-translator who does not know the source language.

In the parts that follow, I will not evoke our personal translation philosophies14, and I will present only the part of the process I am directly familiar with and not that which falls on my collaborator. I draw largely from two versions of La Nuit du hibou (and occasionally from intermediate versions), named here in abbreviated form: the initial version my co-translator sends me (V-Zero) and the definitive version (V-Pub).

2.1 The search for consensus

First, we work by chapter or by half-chapter, meaning in groupings of about five to six pages at a time. Lee Tae-yeon sends me a first version in French (V-Zero) which is to some extent a “literal” image of the Korean text; in any case, it is close in linguistic form to Korean, with minimal effort to adapt it to French, sometimes with ad hoc explanatory comments. Describing the co-translation process of the Bible, an ancient practice, Nida and Taber evoke this initial version of the text provided by the “specialist” to the “stylist”:

[The specialist] attempts to word each sentence as neutrally and colorlessly as possible from a stylistic point of view and convey all the essential elements without ambiguity. In this method, it is imperative that the specialist not produce a version that seems complete, as this would result in an intrusion on the stylist’s freedom, the stylist being the one who should truly feel like the master of this task. (Nida and Taber, 1971, 95)

This initial version comes with a number of comprehension problems that are gradually clarified through questions I raise as comments in the document’s margins. At the end of what averages to eight or nine rounds of back-and-forth (producing 16-18 intermediate versions, which we archive), the document’s meaning is more or less stabilized, its comprehension difficulties disappear, and its stylistic challenges are in hand. Working on the text in this perfectible but overall acceptable state is essentially for the sake of improvement. The exchanges that take place in the initial phase aim mostly to clarify and to validate first proposals: I ask questions and propose alternative solutions, sometimes accompanied by a comment, and Lee Tae-yeon explains the meaning and effect of the Korean text from her perspective. She validates my proposals or, on the contrary, refuses them - when she finds them too far removed from the original, when she prefers to maintain a certain stylistic trait that my revision might have erased, or keep a semantic nuance or syntactical structure. In this phase, we seek a consensus; that is, solutions with which we are mutually satisfied and that we stand behind completely as co-authors of the translation.

To give an example, I will present a case that did not produce long or deep discussions but illustrates simply what I describe here as a process of consensus. It involves the beginning of the novel’s second part, which I include below in the initial French version and a much later one, before the final proofreading phase (a):

Sur le chemin forestier, le haut soir bleu s’asseyait en silence. La colline se teignant d’obscurité devenait doucement bleue, puis s’assombrissait peu à peu.
[On the forest path, the high blue evening sat in silence. The hill dyed with darkness became gently blue, then darkened little by little.] (V-Zero)

(a) Sur le chemin forestier, le haut soir bleu s’asseyait en silence. La colline glissait dans l’obscurité, bleuissait lentement, s’assombrissait.
[On the forest path, the high blue evening sat in silence. The hill slid into darkness, slowly becoming blue, darkening.]

The second sentence had undergone few modifications after V-Zero, and the first one, none, staying literal. The wording “the high blue evening sat in silence” seemed acceptable to me in French, though its figurative weight might be rather heavy. Still, we decided to revise it as the adjective “high” bothered one of our reviewers.

What is at play here, from my perspective, is the tension between the strangeness of the translated text, when it maintains a certain degree of literality, and the disturbance this strangeness can provoke when reading. It involves a delicate balance: the signals of the target text’s strangeness can, upon reminding the reader of the text’s translational nature, be an obstacle to immersion in the diegesis. This has particularly strong implications in a noir novel, and even more so here, as the sentence is the opening of the book’s second part and a turning point in the story. Eco, following Humboldt, differentiates between “strangeness” and “the strange” to describe the effects the translated text can have on reading:

Humboldt (1816) proposed a distinction between Fremdheit (which one could translate as “strangeness”) and Das Fremde (translated as “the strange”). Perhaps he did not choose his terms very well, but his idea seems clear: the reader senses strangeness when the translator’s choice seems incomprehensible, as if it was an error; on the other hand, the reader senses the strange when faced with an unfamiliar way of presenting something he recognizes but has the impression of seeing for the first time. (Eco, 2003, 204)

This is a fine distinction and one that is difficult to objectify as the effect produced by the text is different depending on the reader, but it does indicate what an unusual linguistic form can generate in the target text. It had seemed to me that the effect of the quoted sentence was “foreign” and that any “strangeness” was tenuous, perhaps merely a poetic turn of phrase, but the reader’s remark about “high” made me doubt, especially since, as Eco notes, there is always the risk of the translator’s choice being interpreted as an “error”. According to Choi, it is often “delicate to find the right balance between respecting the original language’s form, which contributes to the making of meaning, and the target language’s constraints, particularly when the two languages are completely different” (2014, 104). Another translator might have chosen not to reword the sentence, but, due to the text’s genre and the sequence’s position within the narrative, we opted to revise it.

This choice was also motivated by the fact that one can assume that in reality, the disturbance expressed by this reader was not only caused by the epithet “high15” but also by a series of phenomena. The image is very saturated: there is metonymy (the word “evening” qualified by “high” and “blue” actually refers to the sky while also specifying temporality) as well as personification (the agent-subject of the verb “sit” is inanimate). The two prepositional phrases also pose problems in interpretation. The first, “on the forest path,” is likely to be interpreted not as a detached adjunct, but a locative complement of “sit”, even if one can doubt that such an interpretation would come spontaneously to the reader’s mind; the second “in silence,” placed just after the verb, has a pronounced effect as it is interpreted as an adjunct of manner that describes the verb, which is not relevant here.

Therefore, we negotiated to find a solution for this series of issues in order to reduce the semantic and figurative conflicts that prevent readers from immersing themselves in the text. The four successive proposals are in bold, followed by Lee Tae-yeon’s (LTY) and my (PR) comments:

(a) Sur le chemin forestier, le haut soir bleu s’asseyait en silence. La colline glissait dans l’obscurité […]
[On the forest path, the high blue evening sat in silence. The hill slid into darkness […]]

- PR: Can you explain what the meaning of “high” could be, or suggest some synonyms? We could write “the tall blue sky”, with the verb; that might give the impression that the evening is described as a grand character, like a king. Or we could just remove “high”.

- LTY: When it is light out, the sky is high and blue. When the sun sets, the high blue evening sits = described as “getting dark.”

(b) Sur le chemin forestier, le grand ciel bleu prenait progressivement des teintes nocturnes. La colline glissait en silence dans l’obscurité […]
[On the forest path, the tall blue sky gradually took on night hues. The hill slid into silence in the dark […]]

- LTY: I prefer the expression “sits”. Here’s what I suggest: [see c. below]. Even without “evening”, one will know from the next passage that night is gradually falling.

(c) Sur la route forestière, le grand ciel bleu s’asseyait en silence. La colline glissait dans l’obscurité […]
[On the forest road, the tall blue sky sat in silence. The hill slid into darkness […]]

- PR: I propose a different image [see d. below]. It’s the blue sky (the day) that submits to the night, as one would bow to a king. What do you think?

(d) Sur la route forestière, le grand ciel bleu s’inclinait peu à peu devant la nuit. La colline glissait en silence dans l’obscurité […]. (V-Pub : 109)
[On the forest road, the tall blue sky bowed lower and lower to the night. The hill slid silently into darkness […]]

- LTY: Oh, good idea!

My first intervention consists of simply proposing to replace “high”, evoking the possibility of deleting it (in order to measure its importance), and exploring effects of personification (the adjective could be used not literally but as a synonym of “majestic”). My comment does not come from stylistic or technical linguistic metadiscourse but from attempts to clarify through simple rephrasing. In my early experiences as a translator, my commentary was longer and more technical, but I quickly noticed that they put me in a position of authority that hampered dialog and the search for consensus. In her response, Lee Tae-yeon makes the meaning behind the image clear, but she keeps the adjective “high”, thereby showing me the importance placed on the seme of height and the literal sense of the word. My second intervention is a proposal that, while keeping the semantic content from Lee Tae-yeon’s rewording (the phrase means that the sky is darkening), normalizes the image to the extreme through a more standardized formula (“take on night hues”) that is indicated as such in my comment (“this is not original in French”), the hypothesis being that it might be a common expression in Korean. Lee Tae-yeon refuses this suggestion, keeping the verb “sit”: we see here that she resists the proposed normalization and makes her desire to keep the original image known. Not managing to find a satisfactory formulation with this verb or a comparable image, I then searched for a solution using personification of the sky, with the image of the “high” sky descending towards the night, suggesting a different metaphor with the verb “bow”, likely to be interpreted in the literal sense (the movement) or figuratively (the gesture of respect). This image is explained in my comment so that my interlocutor can clearly perceive the sought out effect in French. Lee Tae-yeon accepts this suggestion, a consensus being reached: we are both satisfied with the solution found.

This dialogue, rendered here in order, did not take place all at once. The margin comments it consists of are found in five successive versions of the document, where we exchanged in parallel our thoughts on a whole series of other difficulties, the fragmentation of these exchanges each time requiring us to retake mental hold of the problem and its challenges. However, this also allowed us time to reflect, test solutions, and research alternate phrasings. An oral, face-to-face dialogue would probably allow us to work more quickly and spontaneously, but writing buys time for maturation, especially since, except for the intensive final proofreading phases, the time difference helps maximize our time. I generally work in the evenings: first, quickly reading over Lee Tae-yeon’s responses in the morning, letting my mind work passively on some problems throughout the day, then going back to work in the evening.

From this relatively simple example, one can imagine to what extent it is necessary to be patient, placing oneself on an extended timeline, but also how gratifying the work is once a mutually satisfying solution is found. For this reason, my co-translating experience is far from the loss which many translators experience. Rather, mine is an experience of gain: going from a text in an unsatisfactory, unacceptable, or incomprehensible state (V-Zero) to, at the dialogue’s end, a successful text in a satisfactory state. Regardless of the quality of the final product, the process itself is extremely gratifying.

This process involves a relationship of explicit trust: Lee Tae-yeon is the only one who can guarantee the rapport with the source text, which I am not capable of accessing. On my end, I make sure that my suggestions are apparent in the text, and I leave comments as soon as I think an explanation is necessary so that my collaborator can make an informed decision. We venture on the fact that this particular form of collaboration is productive and creative due to its unique features, as the next parts of this paper endeavor to show.

2.2 Questioning, comprehending

In the search for consensus, my primary difficulty is related to attitude and interaction. It consists of putting myself in a position of dialogue, not interventionist authority, as well as dosing out my interventions and commentary on the text. This need is connected to my role as co-translator, not reviser, and interaction with my collaborator. On the one hand, it is a matter of ensuring I understand before intervening in the text as well as not acting as proofreader but as co-author of the translation, and on the other hand, asking the right questions to encourage my collaborator to reinspect the original Korean, if necessary. In the early versions of our work, I ask a lot of questions to understand the processes’ meaning and organization, clarify anaphora, and pinpoint stylistic challenges. These questions must be clearly formulated and simple to limit the time it takes to read them, and it is pointless to ask every question that comes to mind during the first reading, as the answers depend on each other, and one should avoid the document becoming unreadable and the dialogue unmanageable due to too many questions that could have a negative effect on my collaborator16.

Furthermore, it is tempting to intervene rapidly on the text, rewording it to increase its readability and facilitate the task of revision. I have learned from experience that one must not introduce any modifications too quickly, but first take care to verify comprehension of the first version, the strangeness of which, as I mentioned, should be attributed to the nature of this V-Zero, which is unfinished and serves as the basis for our negotiations. Indeed, it is much less costly to leave a comment bubble explaining a point over several back-and-forths of the text than to have to go back over a hastily made modification: in the latter case, one finds themselves needing to dig through previous versions and sometimes find one’s way back to the path that led from one of the text’s states to another, which is time-consuming and psychologically grueling, as one then has to “unravel” what has been done. Facing a passage that poses comprehension problems, it is preferable to advance gradually through the different versions exchanged in order to grasp the meaning, rather than trying to remove all the difficulties straightaway, “straightening out” the language and “cleaning” the text, so to speak, deleting the comments. This is, of course, a posture that is easier said than done for a teacher who, by professional instinct, is used to quickly “correcting” texts perceived as linguistically defective.17

Among the passages that required a particularly dense exchange of questions and answers are the argumentative sequences, mostly because of Korean syntax, which “favors the liaison of phrases and simple sentences using “and” (Choi, 2012, 82) and which forces us to reconstitute logical links starting from V-Zero. Bellemin-Noël, co-translator from Korean, also presents this syntactical and rhetorical difference between Korean and French and the problems it poses to the translator:

An economical, concise language that demands much more intuition than our so-called “analytical,” more talkative European languages, where everything is spelled out, even that which often seems to serve no purpose. Korean is a language, for example, where brevity is the norm, while for us it is a stylistic effect; where the three lines that correspond to my average sentence in French are seen as an affectation borrowed from Proust. (Bellemin-Noël, 2012, 244)

The frequency of simple juxtaposed clauses or those coordinated by “and”, which do not stand out in Korean, leads one to reproduce connectors in the translation, especially temporal or logical ones, which requires a significant effort of comprehension.

The exchanges also frequently focus on point of view, on the scale of a chapter or a whole sequence, or even a sentence or clause, as is the case for the part in bold below (from an intermediary version):

La vérité n’avait jamais attiré Ha-in, et voilà qu’il s’y heurtait à cause de la disparition de son frère. Cette idée le plongea dans la mélancolie. Et ce n’était pas l’ivresse, qui aurait pu le rendre émotif, ni son échec à retrouver son frère, qui aurait pu rendre son humeur changeante.
[The truth had never attracted Ha-in, and there he was facing it because of his brother’s disappearance. This thought drowned him in melancholy. And it was not the drunkenness, which could have made him emotional, nor his failure to find his brother, which could have made his mood swing.]

- PR: Do you think this sentence should be understood as denial, that is: he is saying it is not his drunkenness nor his failure, but we should understand that it actually is these two things governing his thinking?
If your answer to this question is no: who is making this judgment, the character or the narrator?

- LTY: If we include this paragraph in the previous one, would it be more understandable? I think he feels frustrated or melancholic because of the idea that he must face the “truth” because of his brother, since for him, as a lawyer, what mattered to him above all was “fact”.

- PR: But does the sentence really mean that he is drunk but believes that alcohol is not at the root of his feelings (the narrator is the one telling us that alcohol is the cause) / he is drunk but alcohol is not at the root of his feelings (that is what he is saying, and the narrator, too)?

- LTY [underlines the first of the two possibilities above, adding:] I see that he is drunk all the same.

- PR: Okay, he is drunk, but we have 3 solutions regarding the narrator’s position: 1) The narrator thinks drunkenness is the cause of his feelings (HI, on the contrary, does not); 2) The narrator (like HI) thinks that drunkenness is not the cause of his feelings; 3) We do not know the narrator’s position, we only have HI’s thoughts.

- LTY [underlines the third proposal]

One observes that the dialogue about point of view is difficult and goes through a number of rewordings and hypotheses that aim to clarify both the statement’s content and the identity of the speaker correlated to it; it is challenging for me to clearly articulate the fact that my question is centered on point of view and not on the character’s feelings, and Lee Tae-yeon’s responses are evidence of this. The exchange led to a radical reformulation, reorganizing the information in the paragraph, which, in the published version, became:

Ha-in sentit qu’il s’enfonçait dans une grande mélancolie. Quand on boit, on peut devenir émotif. Et quand on reconnaît avoir échoué, on peut basculer subitement dans la tristesse. Mais s’il était d’humeur sombre, pensa-t-il, ce n’était ni à cause de l’alcool, ni de son échec à retrouver son frère. C’était parce qu’il constatait que la vérité, pour laquelle il n’avait jamais éprouvé la moindre attirance, se mettait aujourd’hui en travers de son chemin.
[Ha-in felt himself sink into a deep melancholy. When one drinks, one can become emotional. And when one recognizes having failed, one can suddenly fall into sadness. But if he was in a somber mood, he thought, this was not because of alcohol, nor his failure to find his brother. It was because he saw that the truth, for which he never felt the slightest attraction, was now standing in his way.] (V-Pub., 102)

The final version reorganizes the statements and makes the point of view clear, thanks to the parenthesis “he thought” and the presentative, the latter of which allows one to attribute the statements’ logical structure (a negated cause followed by an asserted cause) to the character (Rabatel, 2000). Not managing to decide if the link between “drink” and “become emotional”, and the one between “recognize having failed” and “suddenly fall into sadness”, should be attributed to the character or the narrator, I took advantage of the malleability of “on” in French (“one”) and the gnomic present, a proposal which Lee Tae-yeon approved. Similar situations arose often, as unraveling each sentence is complex, especially when multiple statements can be traced back to either the character(s) or the narrator.

In this process, we are each the other’s reviser, so to speak. Lee Tae-yeon revises my suggestions based on her relationship with the Korean original (bilingual revision), while I revise hers using criteria specific to the French language (monolingual revision). This process, however, is not self-evident given that I cannot access the Korean text – nothing guarantees that the version sent to me, the one I work from, is not lacking or flawed; furthermore, it is sometimes difficult for me to ensure that my proposed reformulations do not distort the text. As Choi emphasizes, this type of co-translation is “a task done in tandem from the translation phase until the final revision” and, for the collaboration to succeed, the French co-translator must “make the effort to apprehend the text’s content, even its slightest nuances” (2014, 63).

At one point, it so happened that this comprehension effort, and the dialogue it was focused on, left us at a standstill; that is, in her response, Lee Tae-yeon was unable to provide me with the elements that would allow me to grasp the meaning and thus propose a translation. This was a rather particular case as it dealt with the topography of In-su’s lodge near the forest; more specifically, the location of its different yards, the front gate, the grounds, the entry to the basement, all described in Chapter 10 and later evoked in rich detail in Chapters 25 and 26. Certain elements in our translation of Chapter 10 were not consistent in relation to the rest of the story. To no avail, we arrived at a point where we mediated through drawing (a map of the premises), finally making the decision to contact the author directly18 and submit our map of the lodge, which seemed plausible despite not exactly corresponding to what Lee Tae-yeon could glean from the text. Pyun responded that she no longer remembered how she had imagined the place, as considerable time had passed since writing the novel (it was published in Korea in 2012), so she let us decide, particularly concerning the placement of the basement’s entry and the positioning of the front gate, according to what we deemed most effective for the story. We therefore agreed on the premises’ configuration, giving ourselves some leeway, that we will examine by comparing this paragraph from V-Zero with its published version:

Derrière la maison, à côté de la double porte donnant sur la cuisine, il y avait un petit grenier. Juste à son côté, un grand portail de fer était installé entre son mur et la cour. De l’autre côté de l’accès de la cave, les arbres avaient poussé de façon clairsemée et ils se prolongeaient vers la forêt dense. À cet égard, le portail ne séparait pas la résidence et la cave, mais la résidence et la forêt, faisant office de clôture.
[Behind the house, next to the double doors opening to the kitchen, there was a small granary. Right next to it, a large iron entry gate was installed between its wall and the yard. On the other side leading to the cellar, the trees had grown sparsely and extended towards the dense forest, serving as a fence.] (V-Zero)

À l’arrière de la maison, devant la double porte de la cuisine, il y avait un petit local. Un haut portail de fer joignait l’angle de celui-ci et le mur du pavillon, auquel il était perpendiculaire. La cour devant la cuisine se trouvait ainsi en partie isolée de l’extérieur. Sur la droite, au-delà du portail, se trouvait une autre cour, avec du gravier, devant la porte d’accès au sous-sol du pavillon, puis des arbres, qui avaient poussé de façon clairsemée et rejoignaient la masse dense que formait la végétation forestière.
[At the rear of the house, in front of the kitchen’s double doors, there was a little building. A high, iron entry gate joined it and the house’s wall perpendicular to it. To the right, beyond the gate, was another yard, with gravel, in front of the door accessing the lodge’s basement, then trees that had grown sparsely, rejoining the dense mass making up the forest vegetation.] (V-Pub., 115)

The changes introduced are numerous; certain elements borrowed from later chapters (for example, the fact that one of the yards is graveled) aim to clarify the place’s topography. Of course, one could reproach us for having specified this, having made the place’s description more realistic, perhaps adding coherence where it was lesser in the source text. Our objective was in no way to rectify anything we might have identified as a flaw in the text, but in this case, it was impossible for me to find any satisfactory or even acceptable formulations without making the effort to clarify. Perhaps this is an effect linked to the co-translation process as we practice it: to be able to restate the text in French according to Lee Tae-yeon’s version, I feel the need to understand everything, or at least know that, in such a part of the text, I must make room for ambiguity between two possible interpretations. Consequently, I probably have a very low threshold for tolerating vagueness. One can also hypothesize that if I could access the original text, I would be able to identify the zones of uncertainty or wavering in the text and endeavor to render them in French.

2.3 Deverbalize to reverbalize

The second difficulty in the search for a translational consensus is cognitive, in the psychological sense of the term; that is, it concerns the means and mechanisms of knowledge acquisition, namely that conveyed not by the original text, but the image of it relayed by my collaborator in V-Zero. This linguistic image of the Korean text in French is something that, at least at first, I try not to consider as a text to improve, but a linguistic object that I must above all receive as it is: an image of a pre-text from which we are able to create the translated version. When I started working with Lee Tae-yeon, language issues attracted my attention during my reading of V-Zero, preventing me from forming an affective or psychological bond with the story. Since then, I have developed a bit of a particular reading capacity that is rather difficult to describe. I now manage to enter a state of reception where I do not pay attention to the language’s materiality, I detach myself from it (my brain does not identify syntax, lexical, or style problems as pertinent), and by letting the text produce an effect on me, which I would describe as being made of both mental images and emotions. This effect remains anchored to and associated with the ensuing text, from which I work to pose questions and propose reformulations. If I had to describe concisely what happens to me, I would say that, although the wording is a bit paradoxical, starting from a linguistic form, I endeavor to access what is expressed from somewhere beyond language.

My hypothesis is that this state is somehow related to deverbalization, which adherents to the Interpretive Theory of Translation describe as a phase of the translation process: “Deverbalization is the phase known in the translation process between understanding a text and re-expressing it in another language. It involves an emancipation from the linguistic symbols that are concomitant with the grasp of a cognitive and affective meaning.” (Lederer, 1994, 213) The three phases of the translation process are represented below in triangle form, following Seleskovitch’s proposal19:

The first phase of the translation process is that of comprehension, during which the translator seeks to extract the meaning from the text, knowing that this meaning “is not limited to the establishment of a relationship of designation between the linguistic chain and a referent” (Laplace, 1998, 98). For me, this operation, in which the interpretive dimension is constituent, takes place over the course of the questions and answers exchanged with my collaborator. The “meaning”, in the sense these theorists give to the word, can then be deverbalized - it becomes a “cognitive memory” (Hurtado Albir, 2005 [1990], 175). The notion of deverbalization draws from cognitive science, among others:

Specialists in cognitive science clearly delve deeper than linguists in exploring the conceptual, speaking of neuronal traces, mental images, mental objects that physically exist in the brain, even a system of meaning, going as far as confirming the existence of “brain activation maps” and a “geography of comprehension” [Changeux and Ricœur, 1998, 123]. For them, firmly anchored in the materialist movement, “the mental image should not be understood in an evanescent or immaterial sense, but on the contrary, as a well-defined, cerebral activity” [Changeux and Ricœur, 1998, 111]: the representations of language take shape in the cortex and are therefore purely the product of cerebral activity. (Tatilon, 2007, 168)

Starting from this cognitive memory, this mental image, it is possible to move to the final phase, that of reverbalization in another language. I have neither the skills nor the experience to take a stance on this description of the translation process, but the round trips I take in my mind between the verbal and non-verbal, working from a French version of the text toward a satisfactory re-expression, seem similar to the deverbalization process. This is Choi’s position, in any case, for whom the co-translation tasks “are not shared, but divided into two” between collaborators: the one she calls “reviser” “passes through all the levels the translator has already traversed to arrive at the reformulation, except that the reviser, does this work in a single language, the source language” (2012, 64). For me, the key is accessing what is expressed in the V-Zero text while keeping distance from its linguistic form, clarifying its gray areas, and finally materializing it anew in the language. Reverbalization is sometimes very close in form to V-Zero. with only minor changes, and other times more different. The deverbalization-reverbalization process generally arises in my work starting with V-Zero and the above-mentioned grasping of what it expresses, but it can be necessary to deploy it at any stage of the endeavor, including in the last proofreading and revision phases, or to tackle a difficulty.

Deverbalization is particularly productive when it comes to detaching oneself from an image or turn of phrase. For example, it is this method that allowed me to move from the metaphor with the verb “sit” in the first sentence of the text’s second part, to the one with the verb “bow”. I “contemplated”, so to speak, the mental image of the described landscape by making an effort to forget its relation to its linguistic formulation and then, once the image was well-formed, I tried to put it back into words. Distancing oneself from syntax also sometimes requires forgetting linguistic form using deverbalization. Take, for example, three versions of the novel’s first sentence:

Dans la forêt, la plupart des choses, si elles n’ont pas de taille assez grande, n’attirent guère le regard.
[In the forest, most things, if their size is not so big, barely attract a glance.] (V-zéro)

Dans la forêt, les choses qui ne sont pas assez grandes n’attirent en général pas le regard.
[In the forest, things that are not so big in general do not attract a glance.]

Dans la forêt, le regard n’est généralement attiré que par ce qui est grand.
[In the forest, one’s gaze is generally only attracted by something big.] (V-Pub., 9)

The intermediary version was left unchanged for a long time. During the final proofreading, it bothered me because of its two successive negations as well as the modal qualifier “pas assez grandes (not so big)”, implying the presence of a subjective source holding up a measuring stick to evaluate size. Having identified these problems, and knowing that the meaning was clear to me, I tried several alternative phrasings in vain (in particular, removing a negation by using the expression “les choses trop petites (things that are too small)”. It was the deverbalization process that allowed me to reverbalize this, finding the solution that was ultimately kept in the final version. At the lexical level, the changes are minor; the main transformations are in category and syntax: a prepositional phrase (“en général (in general)”) becomes an adverb (“généralement (generally)”); a complete negation (“ne…pas (not)”) becomes a restriction (“ne…que (only)”); a noun (“les choses (things)”) becomes a pronoun (“ce qui (something)”); the sentence goes from active to passive voice, requiring a rearrangement of the agents around the verb. This example, especially the rearranging of the verb’s agents, illustrates a phenomenon I have often observed anecdotally: managing to extract oneself from syntactic form is often difficult, so deverbalization is a precious resource to help the brain “undo” itself from the syntactic relationships that can prevent it from rewording.

2.4 On the edge of my seat

The point I would like to end on is related to the fact that I do not have access to the original text: unlike most translators who first read the whole source text, often several times, and have a solid knowledge of it before starting their translation, I discover it bit by bit as I work. This peculiar situation is only partly offset by my collaborator: at the beginning, she gives me general indications on stylistics that allow me to determine a general register, tone, and way of writing, and we regularly discuss this point. However, we decided by mutual agreement that she not compensate for my ignorance of the story: when we work, I am in the position of a reader who does not know what will happen next in the novel. I commit to not knowing the book’s main points, particularly the denouement, and my collaborator always respected this wish. Of course, this poses an additional difficulty, but I trust Lee Tae-yeon, who knows the book well and who always makes sure the text and narrative are consistent with what follows in the novel. My ignorance, however, sometimes requires the reworking of certain passages once an element is unveiled later in the story, as well as a considerable amount of time for the final revision.

This progressive discovery of the plot has a very strong psychological effect on my relationship with the translation task. It generates a powerful desire to work; that is, to advance in the text, and the pleasure I feel when receiving a new chapter largely makes up for the difficulty of reading V-Zero. If I want so strongly to understand the text sent to me, it is not simply because I want to produce the best possible translation, but because I want to understand what is being recounted, what happens between characters in the story. It is this desire that allows me to accept that this work takes up so much of my leisure time, with psychological motivations similar to those mobilized by seriality. This is how I endeavor to maintain the narrative tension as long as possible, for the sake of the plot, in accordance with the description proposed by Baroni. I cite it at length as he explains very precisely how I can feel, not only as reader of what Lee Tae-yeon sends me, but also as co-translator rebuilding the plot with her, anticipating “the virtualities of a world in movement”:

The plot, when we consider it not as a static configuration, but more like a form in movement, like a transformation whose fundamental nature is to introduce and potentially resolve a tension, bears more resemblance to a labyrinth than to some beautiful classic architecture, symmetrical and organized.
The plot makes room for surprises when it strays from the expected path.
It induces suspense when it recounts important events and those whose development remains partially undetermined.
It provokes curiosity when these events become difficult to interpret, when they are presented in an incomplete or mysterious way.
Thus, the plot develops when we hesitate at the crossing of paths, when the storyline suddenly becomes unpredictable or mysterious in nature, and this tension invites us to anticipate the virtualities of a world in movement. If one considers the plot while it is being carried out, the denouement seems from that moment on like a secondary function, like one of multiple possible futures for the narration, and not its basis. To understand the plot’s dynamic, it is essential to consider at the same time the text’s effective end and its possible endings, the textual structures inscribed in the story but also an act of reading that articulates the actualized structures with the structures that are actualizable, the stories that have a virtual, potential, or alternative mode of existence. (Baroni, 2013, 12-13)

This mode of reception is naturally the one adopted by anyone reading “literally”, a fortiori for a noir novel, but in the case of the co-translation procedure, it is intensified due to temporality, which is much slower than that of normal reading and closer to that of critical reading. It is also augmented by the fact that the blurring produced by the narration itself is combined with that of V-Zero comprehension problems, which are resolved only gradually. Thus, for me, Baroni’s words take on a meaning relative to my experience co-translating as, just after the passage cited, he affirms: “the plot presupposes an intriguing narrator that addresses an intrigued narratee. In other terms, the tension that builds the plot must be considered in the framework of an interaction, and it is first a temporary disorder in the transmission of a message” (13). I identify with this “intrigued narratee”, desirous of navigating the labyrinth that is the text, of a message doubly blurred by the narration and the fact that I have yet to lay eyes on part of the text. And I fully play what Baroni calls “the plot’s game”, which involves the reader refraining “from reading the end of the novel in advance, to at least in part respect its linearity, so as to not spoil the tension” (13).

The second psychological effect of my ignorance of what happens next in the story is that of emotional adherence20. This induces a playful complicity with my collaborator as I often express my reactions or share my conjectures or questions, and she humorously keeps me in the dark. In addition, and this is even more decisive in terms of writing, I endeavor to take advantage of my adherence to the story and its narration in the (re)creative process: I use the emotions stirred up by my gradual discoveries as a spring to propel my writing/rewriting. The emotions initially felt – the fear provoked by a scene, the poetic emotion from the description of a landscape, the compassion for a character, surprise, laughter – later stay associated with the text as we rework it. I particularly remember several moments while working on some frightening chapters when I felt a powerful fear, to the point of having to stop myself from working21. Thus, my position relative to the text is not comparable to that of the original author who, even if she did not plan the story’s entire plot in advance, at least has expertise in it, nor to that of the Korean reader, who advances more quickly in the story and whose attention to detail is inevitably less strong, nor to that of my co-translator who, at least in the work phase, knows the plot, and whose attention is brought into action by issues related to the change in linguistic system.

It is evident that my initial perception of the version sent to me by my co-translator, V-Zero, includes a not insignificant amount of subjectivity, at work in all dimensions of interpretation but particularly manifest on the emotional level. Every translator – and co-translator – is above all a “reader subject”, according to the definition given by Rannou in the field of literature didactics:

Associating the term “reader” to that of “subject” underlines the interpretive responsibility engaged by the act of reading, to be sure, but also all the complexity of the literary experience, including affect, transfer, reformulation, appreciation, pleasure, and rejection, as much as appropriations and singular adventures. (Rannou, 2013, 6)

From this point of view, the co-translation process as I have described it is contradictory to critical reading, at least that which is “ordinarily practiced and valorized in the institution of education” and requires “coming out of immersion, going beyond the mimetic illusion, and losing interest in the plot, the characters’ psychology, or pathos in general.” (Baroni, 2013, 11). On the contrary, without trying to keep them at a distance, I rely on the emotions provoked by the text, in which variables completely foreign to it can be at play, such as the material or psychological context in which I work. In my opinion, passages like the violent scenes in the novel conserve the stylistic marks (parataxis and noun phrases, for example) of the emotions initially felt while reading (surprise and fear), which remained engraved on my memory and were therefore stronger the more I discovered the story. These traces of my emotional reception of the passage, which were kept in the text under my collaborator’s control and with her agreement, can partially be interpreted at least as a mark of translational restatement by a person who is not only co-translator but also reader and subject of emotions.

The work, of course, is not limited to interacting with my collaborator: after the comprehensive revision that aims to reconfirm harmony with the original, resolve any lingering problems, improve style, and ensure consistency throughout, we solicited proofreaders and went back to the text following their remarks and before discussions with the editor who, in turn, proposed revisions. I will not evoke this part of the work here, which was little more than a process of consensus with plenty of compromises and arguments. We had to give up a certain number of stylistic traits that were important to us, especially concerning discourse representation, but the work resulting from modification requests undeniably allowed us to improve the text in numerous places.

Not all translation experiences are alike – the procedure and its genesis are quite particular in the case of a co-translation, especially when one of the translators is not a speaker of the text’s original language. But this experience manifests two phenomena that are likely at work in any translation. First, the determining characteristic of the translation’s material conditions that partially influence the resulting work: here, for example, the method of working on the files, archiving them, discussing and tracking modifications through comments, managing schedules with the time difference as well as the distinct demands for a professional translator and a teacher-researcher whose work takes place after hours. Next, this experience highlights some creative principles at the heart of all translations, the processes of negotiation, particularly visible here thanks to the interactions between the various parties involved, namely Lee Tae-yeon and I, but also the editor. Translation is often described as a negotiation process, between two languages, two systems, two cultures, two aesthetics, two texts – this is certainly Eco’s position in Dire presque la même chose (2007). In the context of translating La Nuit du hibou, I experienced two rather distinct modalities of negotiation: first, consensus, mostly in interactions with my co-translator, which consisted of finding a solution that satisfied both parties involved, namely a linguistic proposal with which both identified and of which both were ready to assume authorship; second, compromise, especially with the editor, which consisted of finding an intermediary solution acceptable to both parties but not necessarily satisfactory for each. Even if one has a natural inclination to assign a more positive value to the first method than to the second, and the psychological experience of consensus is more gratifying than compromise, both result in improving the text as they push one to continuously reinterrogate the relationship to the original as well as the linguistic solutions found, of which there are always countless possibilities.