Translation: A Relational Practice

Translation is a practice in which we ceaselessly engage, but without necessarily being aware of doing so. Moving between various kinds of languages in the humanities, the sciences, and the arts, it is a crucial means of communicating among different human and nonhuman agents and of disseminating knowledge. This is a thought that came to me as Emily Apter spoke during the opening conference of the long-term project “The New Alphabet” at the Berlin Haus der Kulturen der Welt in early 2019, on untranslatability in fields ranging from philosophy to computational science. A professor of French and comparative literature at New York University whose work ranges across translation studies as well as political, psychoanalytical, and critical theory, her presentation corresponded to my work on this book – which is in itself, as many publications of this kind, a space of translation between languages and concepts. The latter is also true for the project space diffrakt, a regular collaborator of the research group “Knowledge in the Arts,” where translation in a broader sense mostly takes place in the format of public conversations. Moritz Gansen of the theory collective had invited Apter to talk about her work on The Dictionary of Untranslatables and subsequently brought me on board. What follows is an account of the event in essay form co-authored by Emily Apter. It offers perspectives on translation through the lens of philosophy, epistemology, art, and human-machine communication and aims to continue to think together about translation as a relational practice in each of these four fields. It also seeks to track the theoretical and practical impulses in these different areas that arise from a relational understanding of translation. Summaries of her remarks appear indented in the following and are based on re-assembled notes on our dialogue.1

Translation is a practice in which we ceaselessly engage, but without necessarily being aware of doing so. Moving between various kinds of languages in the humanities, the sciences, and the arts, it is a crucial means of communicating among different human and nonhuman agents and of disseminating knowledge. This is a thought that came to me as Emily Apter spoke during the opening conference of the long-term project "The New Alphabet" at the Berlin Haus der Kulturen der Welt in early 2019, on untranslatability in fields ranging from philosophy to computational science. A professor of French and comparative literature at New York University whose work ranges across translation studies as well as political, psychoanalytical, and critical theory, her presentation corresponded to my work on this book -which is in itself, as many publications of this kind, a space of translation between languages and concepts. The latter is also true for the project space diffrakt, a regular collaborator of the research group "Knowledge in the Arts," where translation in a broader sense mostly takes place in the format of public conversations. Moritz Gansen of the theory collective had invited Apter to talk about her work on The Dictionary of Untranslatables and subsequently brought me on board. What follows is an account of the event in essay form co-authored by Emily Apter. It offers perspectives on translation through the lens of philosophy, epistemology, art, and human-machine communication and aims to continue to think together about translation as a relational practice in each of these four fields. It also seeks to track the theoretical and practical impulses in these different areas that arise from a relational understanding of translation. Summaries of her remarks appear indented in the following and are based on re-assembled notes on our dialogue. 1 to translation. 3 Terms like "Dasein" or "Agence" (rendered "agency" or "instance" in English), are just two examples: of words deemed untranslatable because their idiosyncratic usage defies equivalence in another language. In the case of Heideggerian Dasein: an English workaround expression like "being there" doesn't really work. And so the German term Dasein tends to carry over in most languages. For Apter this nontranslation is not really an obstacle but rather a spur to thinking. It is where "philosophizing in languages" begins, an expression she borrows from Cassin. 4 Cassin defines "untranslatables" as "what one doesn't translate, but what one doesn't stop (not) translating: after Babel with happiness." 5 Thus, the Dictionary of Untranslatables contains a whole world of different non-, mis-and retranslations of philosophical terms drawn up by a group of philosophers with linguistic expertise. Whereas Cassin's version understands itself as a philosophical and political gesture aiming "to constitute a cartography of Europ ean philosophical differences," 6 the English edition broadens the cartographic parameters, eliminating the restriction to "European" language so as to push for a global philosophical remit. 7 Either way, the emphasis on lingu istic difference holds, as does an emergent theory of untranslatables based on Cassin's process-based notion of "passing from one language to another." 8 The implications of translational resistance and difference for a relational theory of translation are far from clear and hardly noncontroversial. This is what Apter alluded to when she acknowledged the pitfalls of untranslatability. For her, it is crucial to mark the practice related to untranslatability as primarily theoretical. Furthermore, it is important to differentiate untranslatability from inaccessibility: Apter I often get attacked for saying that I am an advocate of permanent barriers … but what I'm really trying to do is question the assumption of a right to have access to all languages through translation. To be aware of conditions under which translation is interdicted, of instances in which the language of the original withholds something or is incommensurate; to desire to work through relational nonrelation across languages is not to say you are against communication or against translation! It's often difficult to make these distinctions clear, which creates problems for the reception of the work I am doing. … I also use the term untranslatability for spaces of translation zones understood as warzones or dissensus and disagreement (in diplomacy). Translation carries a lot of universalist baggage about the promotion of international harmony, utopian transparency (of meaning and intention) and the promise of mutual understanding. I am saying that there is something called Unverständlichkeit [incomprehensibility], which complicates this utopian narrative and that must be analyzed in relation to the history of military encounters and competing ethnonationalisms (extended to what Cassin terms "ontological nationalism"). So, I am not a fetishist of untranslatability for its own sake. Rather I want to use different notions of untranslatability to deepen our analysis of the politics of the encounter. For me, untranslatability is many things, but more than any singular idea, it is a praxis, a way of working. 9

Philosophizing in Languages
This praxis leads to questioning -through an interrogation of what stands out as a philosophical term -what "counts" as philosophy.
Apter The Dictionary of Untranslatables shored up the sovereign authority of the untranslatable by building out its translational lemmas. That said, in broadening the field of what counts as philosophy or deserves to wear its mantle, the Dictionary democratized philosophy's political estates. Dominant languages of European thought -Greek, Latin, German, French and English -were consciously entered into dialogue with Arabic, Hebrew, Russian, Spanish, Portuguese, Italian, Rumanian, Danish, Swedish, and Dutch, with an eye to soliciting, in some future Wiki iteration, an active engagement with all languages of the world. Widening the compass of languages in which the canonical history of philosophy is written, and "making-philosophical" components of language and speech that have traditionally had no purchase on philosophical standing, Cassin and her team redressed class hierarchies.
The Dictionary of Untranslatables marked as philosophical a range of eclectic terms that normally would not make it into a standard philosophical encyclopedia, such as absurd, ça, care, kitsch, life, love, sex, and gender as well as pleasure, postupok, Sehnsucht, unconscious, and vergüenza, vorhanden, word order. The practice of introducing these words as "untranslatables" worthy of being philosophized pluralizes the canon and challenges the high concept-driven hierarchies embedded in continental and analytic philosophy alike. It is a language-oriented approach that contests the hegemony of a propositional logic, 11 preferring instead to ferret out different logics as they arise from context, philological transformation, grammatical usage, and everyday idiom. Apter For Cassin the task of "languaging" philosophy involved recasting its transhistorical abstractions as live elements of a "tongue," subject to the contingencies of situational usage, the wear and tear of social exchange, and the ironies of mistranslation. In this scheme, minor words and syntactic structures exercise their right to philosophizability. Extrapolating here, one could say that "to language" philosophy is "to justice" philosophy's conceptual predicates. Justicing, made into an active verb, carries the sense of rendering unto, or calling to account, structural inequalities inscribed in histories of thought. "To justice" philosophy in the fullest sense would entail reconfiguring its classical modes and branches, from metaphysics to ethics, aesthetics to phenomenology, canon law to logic. If there is a specific political charge, it inheres in the desire to "unexceptionalize" Western philosophy's sovereign vocabulary, thereby extending the franchise to locutions and expressions routinely excluded from philosophy's standard editions.
As an example of how one "philosophizes in languages," Apter referred to Etienne Balibar's entry (in the Dictionary of Untranslatables) of the pairing "'Agency'/'Instance:'" 12 Balibar mentions that the English translation of Jacques Lacan's famous essay "L'instance de la lettre" of 1957 was "The Agency of the Letter" in the first translation by Alan Sheridan and then Bruce Fink comes on in 2002 with "The Instance of the Letter." In German you would come up with "Instanz" and the way in which "instantiating," positioning the self in relation to objects, and "Gestell" play a role here. These huge differences in meaning prompt a reflexive exercise that gets to the heart of what agency is or does in relation to language. They prompt us to query how actions get embodied in verbs, or where the agency lies in performative speech acts. How is the agency of "I do" in the marriage vow distributed across language and speaking subject or listening subjects? I think too of Lacan in seminar 20 (Encore) where he analyzes the phrase: "Pierre bat Paul." In the French tradition, "Pierre and Paul" are paired in the grammar books, going back to the eighteenth century. The phrase "Peter hits Paul" appears often as an example of what counts as "active," in the verb, of what isolates a minimal, condensed expression of human agency. This points in turn to the fascinating issue of how will, action, activisim, and violence are languagedpsychoanalytically, philosophically, and politically. Will is revealed to be tethered to force, and to the causation of harm, or the commission of physical violence.
Sexual violence and the question of will are worked over, as I mentioned, in Lacan's Encore. "Ce n'est pas pour rien que Pierre bat Paul …" ["It's no accident that Pierre beats Paul"] writes Lacan in a section on Jakobson in which he locates causative desire in the formalism of grammar: " wherein jouissance finds its cause, its last cause, which is formal -isn't it something like grammar that commands it?" Lacan then lets loose a chain of associations connecting the violence of the sex act, the question of who wills what in the embrace of the couple, the susceptibility of French grammar's masculine gender to homosexualization, and his own infamous proposition that "there is no sexual relation." 13 Transposing, Transplanting, Transforming Working with untranslatables draws major attention to the epistemic dimensions of a thinking in between languages. A closer look at the verb "to translate" and its own translations may characterize this praxis further. Apter noted that the Greek "hermēneúein" and "metaphérein," Latin "traducere," 14 and German "übersetzen" are all "freighted with associations of transport, passage and transmission." 15 This is also what the entry "To Translate" in the Dictionary of Untranslatables suggests. Among other notions, it mentions Martin Heidegger's understanding of translation as the act of "über-setzen," which in German also means to pass over from one shore to another. 16 As in Breillat's film, this passage is not merely an act of transport, but also of change. For Heidegger this comes through in the "apparently literal, and hence faithful, translation" of central terms of Western philosophy (such as "being," "thing," or "subject") from Greek to Latin, which results in their "rootlessness." 17 Since translation "takes over the Greek words without the corresponding and equiprimordial experience of what they say," it is 13

Translation, A Relational Nonrelation What remains an open question when looking at these different views of how languages move into and over each other concerns the prime mover itself; what puts them in motion in the first place. As Apter remarked at the outset with respect to Breillat's Brêve Traversée translation -the fact that this couple communicates across languages becomes a kind of cover for what is essentially an impossible relation. In the course of our conversation something she called the "relational nonrelation" emerged as an important issue: "seen through the lens of untranslatability, translation is a relational nonrelation. If we follow Jean-Luc Nancy, a relation is a rapport, a reportability." To get a clearer picture of what this implies, it is worth taking a small detour into Nancy's reset of Lacan's infamous proposition: "il n'y a pas de rapport sexuel," "there is no sexual relationship" or "relation:" 22 To say that there is no relation is then to state what is proper to relation: in order to be, it must not be a third thing between two. Rather, it must open the between as such: it must open the between two by means of which there are two. But what is between two is not either one of the two: it is the void … 23
18

The "without-relation of relation" is "its paradoxical reality" 24 in the sense that it is not possible for a nonrelation to exist: "the negation of a relation is still a relation." 25 A relation is not substantial, but what happens in the in-
between is related to actions and movements that are inherent to "relation and rapport," which "come from verbs designating the act of carrying, transporting," 26 -similar to the verb "to translate." Nancy's reading of Lacan's sentence is more of a gloss than a full revisionist interpretation. But with every approach to "rapport" he makes the idea of relation increasingly strange. He not only deconstructs the term but also denaturalizes it, turning it into an untranslatable. This requires and inspires a prismatic reading, which Emily Apter continues with Luce Irigaray: Apter Rapport is a report, a reporting or a reportability. It means a view from over there that is not over here. Irigaray uses the term "rapport à" to parry the force of the transitive and reroute the circuitry of agency (one person acting on or directly addressing another). It is going somewhere around. It is a mode of indirection and it helps us to define the space of the nonrelation and of nonmeaning that is always there as part of an informal communication or process of relating to.

pede exchange, but the fact of position: such that you cannot know what is on that other side of that language. This dialogue also illustrates how noncommunication is exacerbated by false belief. Within the bounds of intersubjective relation, you can only believe you know, you never actually know.
For me, Chomsky's "from where I am sitting" physically embodies the epistemic limit and the spatial articulation of relational voids in subject relations. There is no translational transindividual! There is the absolute singularity of "where I am sitting," which is intractable, untranslatable. Chomsky basically says to Marr that sheer power difference prevails over identity.
Apter has taken the notion of untranslatable difference to heart in her conception of comparative literature. It animates her critique of world literature, 28 which in its more recent institutional guises valorizes market-friendly translation and -ready comparison, ignoring the power dynamics of dominant languages and culture industries. A number of questions were raised during our conversation at diffrakt at this point: Which languages and works of literature are considered worth translating? What is considered a "minor" and a "major" language? How does language politics play when the EU claims that "it speaks your language," 29 when it in fact authorizes speaking only in nationally recognized tongues? Why does the United States bill itself as a monolingual country when it is anything but? How do we decolonize English (or Globish)?
Rediscovering Babble through Machines The question of untranslatability was then extended to digital and machine languages that have become an integral part of political and social structures, hailed not only as facilitators of communication among humans but also between humans and machines. The relations between natural language and machine languages (AI, algorithms, the techniques of deep learning) has, in Emily Apter's view, blurred "the distinctions between natural language and code." 30 Apter Substituting an algorithmic baseline for a philological one, machine translation has consequences for how we work in the comparative humanities, how we think about plurilingualism or define language in the philosophy of language: N. Katherine Hayles speaks of our relation to a "cognitive nonconscious" (another term for the "intelligence" of machines). She is referring to problem-solving that is "unthought" in the human sense. We might say that one way to construe the untranslatable is, similarly as a gathering term for machinic translation that is unthought. Machine translation also draws attention to the increasing complexity of defining what a unit of translatability is. When digital typeface softens the distinction between alphabetic letter and stroke, pixel, and point, when alphanumeric characters introduce computational units into translational alphabets, how does this transform the field of translation studies?
On the phonetic level, corresponding effects can be observed in the automated transcript of our conversation at diffrakt, which was prepared by software based on what is known as natural language processing. When Emily Apter speaks, for example, about Balibar's take on "agency," proper names and non-English titles are interpreted by approximating sounds, turning "Lacan's" or "L'instance de la lettre" into the new word "Flocons" or the phrase "a sale of land stones," while her speech gets reorganized by punctuation that only loosely follows its rhythm: Flocons The transcript was generated on the Auphonic platform using the Google Cloud Speech API. Since this transcript turned out to be experimental, another transcription was done by an unknown human worker from a transcription service. In the end, most parts had to be retranscribed because of many unclear details and a lack of meaningful punctuation, perhaps also due to the quality of the recording.

She makes this last observation apropos of artist Nina Katchadourian's Talking Popcorn (2001/2008), a sound sculpture that decoded the pops of a popcorn machine into Morse code signals and generated speech from them.
When the machine accidentally self-immolated during an exhibition in 2008, the built-in computer preserved the machine's "last words." These were written in alphabetic text on a plinth on which the burned out machine would then be displayed. Katchadourian asked different scholars and writers, among them Emily Apter, to interpret the words. The first phrase on the pedestal reads: "QOCRETETI NEIIHF HEMTLEERA CE SA CFII FAUSE." Apter asks: "FAUSE, shouldn't that be a word?" 33 For her, the artwork is "recovering the space of babble, i.e., language in its pre-edible form, and it allows us to listen to it." Talking Popcorn's "last words," moreover, raise further questions regarding the relation between humans and machines, and the place from which we look at them: Apter What gender is Talking Popcorn? Why did it break down? Did it recover after going into rehab because it was built again, what is its relationship with the artist? Are they co-creating something? What kinds of relationality are being staged in this piece? And then there is the larger question about the relationship of the human subject to machine talk, and by extension, to nonsense, gibberish, and gobbledygook.
It is perhaps the gobbledygook that we ought to listen to in order to avoid common pitfalls inherent in the language we use when speaking about computers, entangling vocabulary from very diverse fields, as Apter highlighted during her talk at the HKW: Machine learning, artificial intelligence, smart technologies, deep neural networks. All these metaphors from Turing to Kurzweill to Dreyfus indicate an indifference to the pathetic fallacy that arises with the assignment of cognitive function to machine processing. … Machinic intelligence, I'm suggesting, is a translation into the language of human understanding of programming automatized algorithmic work, clustering, outputting, and pattern recognition. The point here is that AI has no other mode of representing how it thinks. 34 At the same time, new kinds of machinic expression are co-facilitated by artists working with them. This is also true for the musician composer Tomomi Adachi, who teaches an AI named tomomibot to mimic his style of improvisation in order to become his all-too-human and all-too-machinic co-performer. 35 Apter I think what both Katchadourian's and Adachi's work show really well is that language is not something purely abstract or metaphorical. It is material, right? It is raw material, built up through a web of contingencies, labor conditions, modes of relationality, and nonrelation. There is an adage that language is simply a dialect surrounded by an army, defined by the various power structures that organize phonetic values and visual graphemes into something called a language that becomes marked by an ethnos or a sovereign nation. But AI and machine translation denaturalize the social construction of language, putting us back in touch with prelanguage, and making us aware of the distributions of its units according to given power relations.
With Apter, I would like to consider languages co-created by artists and machines as a potential source of untranslatables that confronts us in a new way with the radicality of untranslatability. Different from untranslatables in philosophy, the tomomibot and Talking Popcorn do not expose us to that which does not translate in discourse, but to the experience of that exposition: the moment of not understanding, alienation, stuttering, radical uncertainty, and doubt, and this, in turn, is core to doing philosophy. One crucial aspect of that experience is the confrontation with machines speaking in tongues, using vocabulary that we cannot look up in any dictionary. As a new kind of glossolalia, machinic speech "preserves only the envelope of semantic intention." 36 Another excellent speaker of such prelanguage is nimiia cétïi, a neuralnetwork-aided co-creation of the artist Jenna Sutela and the bacterium bacillus subtilis. Nimiia cétïi can be perceived by humans as a typeface that is based on the raw movements of the bacteria and as a voice speaking in tongues of Martian language. The voice is produced by a neural network trained on Sutela's voice that matches what it sees in the bacteria's movements with its mimicking of Martian language. 37 In this way, Sutela tries to connect with the "non-human condition of machines" (wich she considers " aliens of our creation") as well as with other-than-human forms of life and their intelligences that already live with(in) us, e.g., in the case of microbes that are part of the gut-brain-connection. 38 And she shows how the co-created languages of machines and nonhumans contribute to new ways of reflecting on how we relate to language tout court. What all three of these artistic works have in common is that they acknowledge the creative potential of alienating humans from "natural" language. In this case, machines are not merely being used as "perfect others" or to serve human needs of pragmatic translation. 39 They do not just confront us with Unverständlichkeit -incomprehensibility -but also reveal that languages are subject to culture, to cultures conditioned by humans and nonhumans. In this respect these works hint at the limits of human interpretation and direct