Keywords

1 Introduction

At ICIDS conferences, the issue of translating an IDN was primarily addressed as a form of translation between media (for example “adaptation of stories as a translation between media, such as literature and film”, [1]), but rarely as translation into several languages. This paper presents an overview of the different experiences in translating the literary interactive digital narrative Déprise [2]Footnote 1. Our objective was to comprehend the various strategies that the process involved. In order to do so, we have asked the English, Italian, Spanish, and Portuguese translators of the work for feedback on the translation process: Valerie Bouchardon (Loss of Grasp, 2010), Giovanna di Rosario (Perdersi, 2011), Martha Asunción Alonso (Perderse, 2013), and Diogo Marques (Perda de controlo, 2016). Drawing upon the written exchanges with the translators, we discuss the following points: 1) modes of collaboration between author and translator (programming strategy between translator and author; the translator’s digital literacy and familiarity with digital literary works; the translator’s visibility); 2) the significance of translating specific dimensions (code migration, semiotic forms and gestures, embedded voices). Additionally, we briefly discuss some future research trajectories concerning the translation of IDNs (the role of indirect translation (Gambier, 1994), the cultural dimension of the works, translation as reinvented memory). Lastly, following the four facets of translating digital literary works as identified by María Mencía, Søren Pold, and Manuel Portela’s [3] – 1) translinguistic; 2) transcoding; 3) transmedial; 4) transcreation – we explore whether such facets are found in the translator’s experiences with Déprise.

2 A Case Study: Déprise/Loss of Grasp

2.1 General Aspects of the Digital Narrative

Déprise [2] (henceforth referred to DP) is an online literary interactive narrative. The English version of the work (Loss of Grasp) won the New Media Writing Prize in 2011. This work is part of the ELMCIP Anthology of European Electronic Literature [4] and the Electronic Literature Collection volume 4 [5]. DP consists of six scenes that portray the mindscapes of a man who gradually loses grasp of his life. In scene 1, the protagonist contemplates his life and experiences a stormy moment, a loss of grasp; in scene 2, he meets his future wife on a date; in scene 3, he reads his wife’s twofold love/breakup letter; in scene 4, he reads his son’s poignant text; in scene 5, he experiences an internal struggle which culminates in a complete loss of grasp; and in scene 6, he decides to confront the situation to take control of his life again. With each scene, the reader is challenged to decipher the relationship between the computer’s interfacial imaginaries and her/his own self. The work was originally programmed in Flash in 2010 and migrated to HTML/JavaScript in 2018. Though this paper is based on the English, Italian, Spanish, and Portuguese Flash versions of the work, it is important to note that the Arabic, Chinese, German, Hungarian, Polish and Russian versions were published in January 2020 (see Fig. 1) and were exclusively developed in JavaScript.

Fig. 1.
figure 1

Homepage of Déprise/Loss of Grasp.

Full size image

2.2 The Interview Process

The initial interviews and exchanges were conducted during the same years as the individual translations, involving Valerie Bouchardon (Loss of Grasp, 2010), Giovanna di Rosario (Perdersi, 2011), Martha Asunción Alonso (Perderse, 2013), and Diogo Marques (Perda de controlo, 2016); subsequently the translators were recontacted in 2020. All correspondence occurred through email and the questionnaires were written in French and English. The languages of communication between the author and the translators were as follows: V. Bouchardon (French), di Rosario (French/English), Alonso (French), Marques (English). Multilingualism was part of the process from the outset. For instance, Diogo Marques, responsible for the Portuguese translation, used the English version as the source text and compared certain aspects of his translation with the French original as well as the Spanish and Italian versions, but communicated with the author in English. Undoubtedly, the exchanges between the author and the translators led to an intercultural and multilingual dialogue, inviting to traversing linguistic and cultural boundaries.

3 Different Forms of Collaboration

3.1 Between Text and Code

In digital works, “code” can create a certain distance between the translator and the author. The extent of collaboration between the author and the translator impacts the author’s task of programming the “translated text”. Consider for example the importance of semiotic manipulation when recreating similar literary effects in the different versions of a work. In the case of DP, most of the translators sent the linguistic text to the author without subsequent involvement in the development of the code adaptations of their respective versions.

In the course of the interview, V. Bouchardon (Loss of Grasp, 2010) expressed the following: “I preferred to work with the code file. I needed to see the coding context to better understand the interplay between the code and what appears on the screen.” This statement raises the question of the structural opacity of any computer program, the question of the mediation of computation between the code and the representation on screen. Di Rosario (Perdersi, 2011) had a similar thought when dealing with the reproduction of the rhetorical figures of the text:

My primary concern has always been to reproduce "the movements" of the original text in my translation. It's not just a linguistic challenge, of course, when I think of Déprise and all its “rhetorical figures”, I still wonder if I really managed to render them all in Italian. My aim was to preserve the “invisible” digital text, and I am not thinking, in this case, about the code but about what the code allows the writer and the reader to do.

These two examples highlight the particularity of literary IDNs in delineating the boundaries between code, author and translator; how does the “code distance” between the author and the translator affect the programming of the digital work into different versions? Following the interviews, we observed a pattern: once the linguistic content was sent to the author, programmed into the new version, and shared as a trial version between author and translator, the translators would frequently request modifications to their versions after reading and interacting with their version of the work. We noticed that particularly reading time (on the screen), interaction and manipulation challenges played an essential role in the translation test version processFootnote 2.

3.2 The Translator’s Digital Literacy and Familiarity with Digital Works

The first critical approach to the translation of DP into a different language (Italian version) was made by di Rosario and Borràs [6]. The researchers pointed out that the role of the translator in the digital era is changing “as not only does s/he need to translate words, but also images and movements, and sometimes s/he is also required to have the technical competence to do it.” By “technical competence”, we refer not only to the translator’s digital literacy (i.e. computer skills) but also to a certain level of familiarity with digital works. During the interviews, di Rosario, Alonso, and Marques answered that DP was the first digital work they had ever translatedFootnote 3; and specifically, for Alonso (Perderse, 2013), she had never heard of digital works before and therefore translating DP was a completely new experience for her.

In response to this, Marques asks: “Should digital literary works of art be translated by non-specialists of digital literature?” One way to answer Marques’ interrogation is to consider the example of Alonso (Spanish version), who is not a specialist of digital literature, but who managed to translate the work through email communication with the author. However, Alonso’s translation primarily centered on working with the linguistic content and phonetics, and not with the code file as was the case for V. Bouchardon, di Rosario, and Marques himself. Considering that Marques and di Rosario are digital literature researchers, that V. Bouchardon is an ESL teacher and Alonso is a translator, we can further ponder how the translator’s professional background influences the translation of the source digital text? Is there a specific digital literacy for the translator of digital literature?

Recording voices (scenes 1, 2 and 4) as part of the translation process is specific to DP. This involves translating the transcript, identifying a voice actor/actress, asking them to read and record the translated transcript and send it to both the translator and author. For example, while working on the Portuguese version, Marques said that “he needed to find three people to voice the text and to ensure access to recording equipment”; similarly, di Rosario noted that there is also a very basic technical problem to keep in mind: “in order to translate digital works, you need several technical tools specifically for the recording”Footnote 4​. V. Bouchardon, on the other hand, employed the same software (Audacity) for recording herself as she uses with her students, thereby circumventing the need to learn a new tool. As we can observe, the level of technical integration into the different translation methodologies fluctuates depending on the professional background of the translator and on the extent of the collaboration between author and translator. Sometimes, the translators found it necessary to acquire a new set of skills depending on the challenges they encountered; for instance, we can notice that translating and recording new audio content into other languages was underlined by Marques and di Rosario as an important step in their translation methodology. This aspect presented not only technical complexities but also posed challenges related to voice performance, as will be elaborated upon in Sect. 4.3, “sound and meaning”.

3.3 The Visible and the Invisible

In their article titled “Renderings: Translating Literary Works in the Digital Age”, Marecki and Montfort [7] discuss how the role of the translator changes when translating digital literature in comparison with traditional works:

In traditional works, the translator is often invisible, a background figure, sometimes subtly credited or even not mentioned at all. In the case of digital works, the translator becomes visible, an ambassador of the work, often explaining its mechanism and the translation process.

The interviews reveal the translators as active and versatile figures when it comes to working on digital literature. They are able to ask questions, speak to the author about stylistic matters, develop their own methodologies, acquire new digital literacy skills (if needed), and experience the act of translation through new modes of aesthetic and literary communication (such as interacting with and manipulating their linguistic text, in contrast with print literature). There is a dialogue, an exchange of ideas and at times an involvement in the coding process. Translators are recognized (as evidenced in the work’s credits) and have a voice and role in the creative process.

Moreover, visibility and invisibility in translating digital literature can be explored in a different way if we go back to the idea of the invisible digital text existing beyond the screen, and its visible linguistic translation. As di Rosario explains when sharing her translation experiences:

In a certain sense I find that the linguistic translation is “visible”, i.e. ​one can overlook the connotation of a word or its nuances, yet easily refer back to the original text, to at least check the original word​ (the word is present and fixed on the page). In digital literature, things get more complicated, because obviously one can translate the linguistic content correctly, and at the same time, part of the meaning of the work − that is also created by other movements, by manipulations, etc. − can be forfeited.

Di Rosario emphasizes once again the importance of interaction and manipulation of the work to evaluate the aesthetic effects of the linguistic text beyond the screen. It seems that within the translation process, there is certainly a negotiation between the visible and invisible text and its aesthetic outcome. For his part, Marques [8] highlights the ideas of transparency, translucency, and intersemiotic aspects when translating digital works:

It is interesting to note that the idea of transparency and translucency in digital interfaces that is being propelled by digital technology industries, can be compared to the idea of transparency and translucency in the context of translation. Namely when it comes to translating something that stops being exclusively discursive in order to become multimodal and involve multisensory perception.

Certainly, intersemiotic translation is specific to digital literature where visual, auditory and other sensory channels play an important role in the construction of meaning of the works. Marques’ and di Rosario’s observations indicate that their concerns as translators are directly related to the competencies that they already have as digital literature readers and researchers (how do we read digital literature?).The fact that di Rosario mentions “digital rhetorical figures” shows a specific aspect of the concealed dynamics of literariness within a work of digital literature. On the other hand, Marques’ reflections on the complexity of multisensory perception (e.g. App version) bring to the forefront the intersemiotic choices (e.g. shifts of time, context, and texture of semiotic resources), Bouchardon and himself had to make during the process of collaborative translation.

4 Translating the Specific Dimensions of Digital Literature

4.1 From Flash to JavaScript

DP is one of the many digital works that have encountered the challenge of programming software obsolescence. Flash, the original software in which the French, English, Italian, Spanish, and Portuguese versions were developed, stopped running in December 2020. In a bid to adapt the work to the current technological challenges, the author migrated these four versions to JavaScript in 2018. The process of loss and gain in translation was reflected on the “screen disturbances” from one piece of software to another. Certain “iconic features” of the Flash version were lost, while other “aesthetics aspects” were improved in the JavaScript version. It is important to note that this paper does not address the “re-adaptions” of the linguistic text, such as the transition from the English Flash version to the English JavaScript version; however, we are aware that a “revised translation process” is needed for the previously published versions of the work that have been recently re-programmed due to software obsolescence (Figs. 2 and 3).

Fig. 2.
figure 2

Example of XML code for the scene 2 of Loss of Grasp (for the Flash version in 2010).

Full size image
Fig. 3.
figure 3

Example of JSON file for the scene 2 of Loss of Grasp (for the smartphone App version in 2018).

Full size image

With this in mind, it can be argued that DP undertook a true process of transcoding (moving from Flash to JavaScript) and transmediation (transitioning from PC to an App for smartphones and tablets). Rethinking the scenes with a tactile dimension (given the absence of mouse pointer on smartphones) implied a challenging transcreation task for both the author and the translators. The translation between machine -readable codes has had an effect on the visual – and audio – glitch effect apparent in the JavaScript versions. For example, scene 1 (the protagonist’s stormy moment, complete loss of grasp) is rendered differently with JavaScript in both the web-based and the App version causing the speed of semiotic displacement to decrease (Fig. 4). Similar examples include the interaction with the love/breakup letter written by the protagonist’s wife in scene 3 wherein the act of reading the text aligns more cohesively with upward and downward motions (JavaScript) compared to the right-to-left approach (Flash). Additionally, the appearance of the individual phrases in scene 6 is perceived as slower and less graspable in the JavaScript rendition than in Flash. It seems that the “compositional principle” [9] of the work could not be expressed in the same way with the shift from Flash to JavaScript.

Fig. 4.
figure 4

a) Excerpt from scene1 (Flash version); b) Excerpt from scene1 (JavaScript web-based version).

Full size image

4.2 Semiotic Forms and Gestures

The interplay of semiotic resources (words, images, audio, gestures, etc.) plays an important role in digital literature translation practices. Can we truly achieve intersemiotic cohesion in the translation of certain works of digital literature? For example, when recreating similar literary effects (i.e. figures of speech and rhetorical figures) in the different versions of the piece. In DP such literary and rhetorical effects are possible thanks to “figures of manipulation” (meaning gestural manipulation). These figures rely on a gap between the reader's expectations while manipulating the text and the result on the screen [10]. Considering this, we argue that the translatability of the literary effect of the figures of manipulation calls for a careful intersemiotic translation in all the current target languages (English, Italian, Spanish, Portuguese).

For example, in scene 2 (the rendez-vous when the protagonist meets his wife), a question mark appears on the screen, waiting for the reader to explore its functionality. A few seconds later, the reader understands that s/he needs to click on the question mark for a series of questions to randomly appear on the screen: “Who are you?”, “Do you like…”, “What do you think about…”, “Where are you from?”, “Where are you going?”, “Do you think…” (Fig. 5a). Following Mencía, Pold, and Portela (2018), this is a key moment for the translation of the piece at different levels: 1) translinguistic: insofar as the translators had to translate the different versions of the linguistic text from the source language to the different target languages (Fig. 5b, 5c, 5d); 2) transcoding: insofar as the author had to reproduce this effect on the five new JavaScript versions as well as on the App (migrating platforms); 3) transmediation: insofar as the translators and the author had to integrate certain semiotic modalities into the App version (e.g. multilingual visual narration of the woman); 4) transcreation: insofar as the author and translators had to reproduce the literariness and translatability of the “figures of manipulation” (interaction and manipulation) that unveil the image of the woman’s character. The translation process aimed to create a harmony between interaction, manipulation, and the linguistic texts in different languages that construct the same image. The aim is not only to grasp the meaning of the aesthetic elements of expression but also to search for literary patterns.

Fig. 5.
figure 5

Déprise, Flash version, scene 2 target languages: a) English, b) Italian, c) Spanish, d) Portuguese.

Full size image

4.3 Sound and Meaning

DP is a polyphonic creation. Translating the different voices has played an important role in the translation process. The voices of the narrator, a telephone operator and an adolescent appear in different scenes. These voices act as unidirectional conversations at different rhythms that create an atmosphere of digital heterophony.

In scene 2 (the rendez-vous when the protagonist meets his wife), the author plays with what he calls “alterations” or misunderstandings between the voice being heard and what actually appears on screen. Interestingly, in the original French version, this effect was created accidentally. That is, the original idea was to use a speech recognition software programme to generate these alterations or misunderstandings.

  1. a.

    Vous habitez la région depuis longtemps ? (Vous évitez la légion depuis longtemps

  2. b.

    Et vous travaillez dans quoi ? (Et vous travaillez l’envoi ?)

  3. c.

    J’ai l’impression qu’on a beaucoup de point communs (J'ai la pression et la pinte en commun)

  4. d.

    Je vous trouve vraiment très jolie ! (Chevaux, brousse, bêlement… prés jolis)

  5. e.

    J’aime votre façon de sourire (Gêne, votre face a des soupirs)

  6. f.

    Vous voulez marcher un peu ? (Nouveau-nés barges et il pleut)

Except for V. Bouchardon, the translators were not aware of the use of speech recognition software programme to generate these alterations, and therefore tried to reproduce this effect in the English, Italian, Spanish and Portuguese versions. The stylistic challenge was experienced differently by each of them, di Rosario (Perdersi, 2011) explains:

Another complicated part was what Bouchardon called the “alteration”: in fact, in one scene there are some sentences that will be changed. I left some of the meanings, especially in the main sentences, but I played with the sound of the words in the “alteration”, like Bouchardon did in the French version so the meaning of some sentences is totally different.

For her part, Alonso (Perderse, 2013) notes:

When I had to translate Déprise, I worked on the alterations in order to obtain the same effects of surprise as in French. For that, I played quite freely on phonetics. I consider this aspect (the recreation of rhythms and sounds) to be the main challenge of any translation. This seems to me to be of major importance in the case of a digital piece, directly appealing to all the senses of the readerFootnote 5.

In his experience, Marques (Perda de controlo, 2016) comments:

I found particularly challenging to ‘‘transcode’’ into Portuguese that specific French humour present in all of Serge's interactive fictions. For instance, in the case of cultural transfers, such as thorny idiomatic expressions and the added difficulty in translating homophonic words.

For V. Bouchardon, the experience was a little different. The translation of this part was the result of an ongoing discussion with her husband (the author). Moreover, as an ESL teacher, V. Bouchardon is used to her students confusing sounds and words in English. She relied on her teaching experience to render the “alterations”.

Translation examples scene 2

French (original):

  • “J’aime votre façon de sourire”

  • (Gêne, votre face a des soupirs);

English:

  • “I like the way you smile”.

  • (I light the west aisle);

Italian:

  • “Mi piace il tuo modo di sorridere”

  • (So il tuo ruolo nell’uccidere);

Spanish:

  • “Me encanta tu sonrisa”

  • (Andar por la cornisa);

Portuguese:

  • “Gosto da maneira como sorris”

  • (Gosto da bandeira como só ris)

Translating the dichotomy of meaning and sound is one of the greatest challenges in the translation of scene 2. The translators’ imagination and stylistic strategies proved a true process of transcreation, considering that the original version was made with a speech recognition software programme. Certainly, the linguistic and phonetic exercise triggered the following questions: how to translate cultural humour in a work of digital literature? How do different cultures interpret DP? The fact that DP continues to be translated into different languages gives it a “prismatic translation” effect [11]. On the one hand, the voices on scene 2 belong to different cultural contexts (France, Spain, Italy, Portugal) and individuals; and on the other hand, random cultural elements were used in the translations to fit each individual context. Moreover, these translations show the inextricable connection between multilingualism and multiculturalism as a product and as a challenge of translation.

5 Future Trajectories

5.1 The Role of Indirect Translation

The French (original) and English versions of DP were released simultaneously in 2010. In the interviews, the translators expressed that for some of them the source language alternated between the French and the English version. Therefore, there are two things to consider: 1) the translator’s knowledge of other languages, 2) the existing translations at the date of a new version. We know for a fact that the source text used by V. Bouchardon was the French version (2010). In the case of di Rosario (2011) and Alonso (2013), they also worked with the French version (2010) but had the possibility to make aesthetic and linguistic comparisons with the English version (2010); whereas, Marques’ (2016) source text was the English version, but he had the linguistic competencies to read the original French version, and make comparisons with the Spanish and Italian versions. The interconnections among the source and target languages open the possibility to speak of a process of indirect translation in digital literature, namely a translation of a translation [12].

As we can observe, V. Bouchardon’s English version (Loss of Grasp, 2010) stands as a referent considering that English acts as a language of reference and encounter between translators (lingua franca). With this in mind, we can ask, up to what extent has V. Bouchardon’s English translation influenced other versions of DP? Which would be considered as the source text: Déprise (2010) or Loss of Grasp (2010), or maybe both? What is the visibility and literary impact of indirect translations in the future versions of DP? In a future critical approach, we will certainly investigate the methodological possibilities of indirect translation in literary IDNs (borrowings, comparisons, dissimilarities), taking as a case study the six new versions of DP (Arabic, Chinese, German, Hungarian, Polish and Russian).

5.2 The Cultural Dimension

DP contains passages strictly related to Western culture. For instance, in scene 3, the reading from left to right of the love/break up letter in comparison to reading from right to left in the Arabic version; also in scene 3, the inclusion of Georges Bizet’s Carmen as background music; in scene 4, the concept of “hero” expressed by the protagonist’s son; also in scene 4 the mention of the word Zoïle (a Greek critic); to name but a few examples. The adaptation to the different versions of such aesthetic and poetic effects raise the following questions: how do we substitute or compensate for such cultural aspects in the target culture? How do we deal with cultural transferences when translating works of digital literature? In the example of the love/breakup letter in the Arabic version, we find that the double meaning and visual effect of the linguistic text is relevant to the meaning of the work, but not culturally consistent with the target language. A similar thing happens with Bizet’s Carmen where the Opera plays an important role in the construction of the literary atmosphere of the scene − as the protagonist reads the letter, but it might not be suitable for the different contexts of the translations; or the translation of the word (persona) “Zoïle” into the different languages.

In her article, “Digital cultures: A view from French studies and literature” [13], Erika Fülöp expresses a desire to “counter the stereotype of a homogenous global culture in the Digital Age”, insisting on the fact that digital literature preserves the traces of pre-digital cultures. Fülöp’s reasoning is closely related to the translation of literary digital creations today. Should the role of the translator (into English, for example) be to dilute cultural references so that an English-speaking audience may identify with the work, or on the contrary to reproduce any traces of cultural specificity, thus emphasizing the cultural diversity of the productions of digital literature more than their international dimension [14]? If so, is cultural diversity expressed uniquely through the linguistic dimension of digital literature? What about the relation and equilibrium between digital aesthetics and cultural transferences?

5.3 The Digital: A Reinvented Memory

If we consider translations of literary IDNs as archives of cultural and technological elements: computer software that stores ways of expression of a specific digital software period (Flash obsolescence and perhaps in the near future, JavaScript), cultural translation of idiomatic expressions, cultural adaptations of the fine arts; a link could be made between translation and preservation. The added value of digital technology is not where one expects. The digital medium is not a natural preservation medium but on the contrary it is hell for preservation. But digital technology makes us enter another universe which is a universe of reinvented and not stored memory [15].

From an anthropological point of view, this model of memory is more valuable and more authentic than the model of printed media which is a memory of storage (the book that one stores on a bookshelf just like the memory that one would store in a case of one’s brain). Indeed, cognitive sciences teach us that memory does not function on the model of storage and conservation. Preserving is thus permanently reinventing the content − just like translating. The issue is to have an accurate and faithful invention, a reconstruction in which the changes are explicit and commented upon. On a similar level, translation highlights the digital age as a move from a model of stored memory to a model of continuous reinvented memory. Thus, from this point of view, considering the interplay of intersemiotic forms in literary IDNs (words, images, audio, gestures) and keeping in mind Reynolds’ proposition of “prismatic translation” [11], literary IDNs can be regarded as a good laboratory to experiment with translation in the digital age.

6 Conclusion

The insights drawn from the interviews underscore that translation processes generate a creative tension that interweaves media, semiotic forms, programmed writing, aesthetic experience, and cultural aspects. These translation processes – spanning the translinguistic, transcoding, transmedial and transcreative dimensions (Mencía, Pold and Portela, 2018) – prove to be ways of traversing boundaries and barriers.

Culture is present within literary IDNs in a plethora of forms; therefore, translation can act as a vehicle for the dissemination of cultural works, reaching diverse audiences. Interviewing translators yields a comprehensive perspective of the different processes and methodologies used when translating a literary IDN, particularly when the goal is to cover more than one target language. The growing number of literary IDNs, characterized by a diversity of genres and technological apparatus, will not only shape the evolution of translation methodologies, but will also help to redesign research trajectories within the field itself.