Corps de l’article

This paper seeks to contribute to current debates on the ethics of translation through the introduction of a feminist theoretical paradigm that enables the discussion to continue outside the now widely discredited notion of equivalence as the goal of translation and fidelity as the predominant model for the ethics of translation. Adopting an ethical understanding grounded in difference rather than sameness, I draw on the work of Bracha Ettinger, a contemporary Israeli artist, psychoanalyst, and feminist theorist. Previously I explored Ettinger’s psychoanalytic concept of matrix, a feminine Symbol modeled on the maternal/late prenatal infant relation, as a means of refiguring relations between readers, authors, and literary texts (Shread, 2005). I suggested that in matrixial reading, a text fosters ethical encounters through exchanges in which difference is maintained within an intimate space. I am now interested in exploring how Ettinger’s paradigm can be productively introduced into the field of translation studies.

Ettinger’s contribution to contemporary theory is relevant to translation studies firstly because her matrixial paradigm describes creative processes of transfer and transformation that provoke a reworking of traditional notions of how translations ‘serve’ original texts. Ettinger’s model allows us to further develop translation theories attuned to an ethics of difference, as found in the work of feminist critics such as Barbara Godard, Sherry Simon, Luise von Flotow, and Susanne de Lotbinière-Harwood, and in the foreignizing techniques oriented towards a recognition of difference advocated by scholars such as Lawrence Venuti. Matrixial relations also embody a respect for the ethical constraints on translating that post-colonial critics such as Gayatri Spivak have drawn to our attention. Lastly, this ethical model responds to the challenge to rethink translation from a wider cultural studies perspective by expanding our understanding of how culture is created and shared collaboratively, as opposed to ‘heroically’ by individuals.

As ethics are implemented through action, I begin by suggesting some of the ethical issues involved in translation by considering the multiple tasks accomplished by translators. After presenting Ettinger’s theory and explaining how the process of metramorphosis—as opposed to metamorphosis and metonymy—reflects a feminist re-structuring of exchange relations and accesses an ethics of difference, I tease out the implications of drawing Ettinger’s paradigm into the emerging discipline of translation studies, in which there is an increasing appreciation of the place of ethics in translation. Developing the matrixial model, I explain how metramorphosis makes space for uncanny remainders in the text. Through the cabbalistic notion of tsimtsoum, I explain how a matrixial perspective views translation in expansive terms, rather than as loss. In these ways, I explore translation processes through a paradigm that allows for creative collaboration, demonstrating that matrixial understanding is valuable to translation studies because it offers a model founded on multiplicity, instead of yearning nostalgically for the solo accomplishment associated with the mythological “original” text.

Ethics and the Tasks of Translators

Echoing Benjamin’s foundational essay on “The Task of the Translator” (1923), translation theorists invariably feel compelled to pass through the initiation rite of offering their definition of the role and responsibilities of translators. As Edwin Gentzler comments, “in Resistance to Theory (1986), Paul de Man goes so far as to say ‘that you are nobody unless you have written about this text’” (Gentzler, 2001, p. 171). Benjamin’s work has been particularly important to theorists working to counter an Anglo-American tradition based on equivalence, fidelity, and a hierarchical relationship between an inimitable original and its inevitably deficient translations. It is striking and somewhat ironic, however, that these commentaries so often offer a singular, narrowly circumscribed definition of “the” task of the translator, mirroring a long tradition in which translation has been tied to dictionary equivalents. Adopting a larger perspective, I hope to point out how this intertextual reference connects a range of theorists across their differing definitions. By considering the implications for agency and responsibility in the various tasks for translators, I begin to map out an ethical topography that calls for Ettinger’s matrixial paradigm.

Benjamin raises the ethical stakes associated with the practice of translation in his essay “Die Aufgabe des Übersetzers” (1923), originally translated by Harry Zohn in 1968 as “The Task of the Translator” and re-translated by Steven Rendall in 1997 as “The Translator’s Task”[1]. The fortuitous title of the essay has retained its enduring fascination among theorists, as a touchstone text that combines connotations of arduous duty with an assertion of responsibility implicit in “the task.” Despite the singularity of the task in the title, in Benjamin’s essay, translators are in fact entrusted with several tasks, one of which is the renewal of the receiving language: “To set free in his own language the pure language spellbound in the foreign language, to liberate the language imprisoned in the work by rewriting it, is the translator’s task. To this end he breaks through the rotten barriers of his own language”[2] (Benjamin, 1997, trans. Rendall, p. 163). Benjamin also lays the ground for what later defines functionalist approaches, which privilege the receiving culture: “The translator’s task consists in this: to find the intention toward the language into which the work is to be translated, on the basis of which an echo of the original can be awakened in it.”[3] (Benjamin, 1997, p. 159). In offering these definitions, Benjamin is not merely describing, but is in fact re-directing translators away from a conventional practice towards a more empowering and daring undertaking, one which is more taxing, since translating is no longer seen as rote reproduction, but as a creative production that innovates language and understands meaning in the large sense of the culture borne by language.

Lawrence Venuti is one of the contemporary champions of Benjamin’s emphasis on the translator’s responsibility to disrupt the home language and culture through the importation of the foreign. Viewing the task of translators in this light is a direct challenge to an ethics of fidelity which tends to produce domestications of the foreign, subjugating the strange by turning it into immediately recognizable forms, evaluated in terms of “fluency.” Against this model Venuti claims that “A translation ethics, clearly, can’t be restricted to a notion of fidelity” (Venuti, 1998, p. 81). For Venuti, the task of the translator is that of contestation, a straining and often unpleasant task because it is entrusted with unsettling cultural assumptions that are deeply embedded in language. Venuti thus departs from a traditional ethics of fidelity to argue in favor of “foreignizing” techniques. Rather than seeking subserviently to be faithful to the style and intention of the original, he values the ability of a source text to influence, distort, and openly manipulate the style and language of another culture via a translation. Yet Venuti’s definition of the translator’s task is closely circumscribed by the context in which he writes: his implicit concern is to resist the particular hegemonic relationship between the United States and countries under its neo-colonial influence, although it is often misread as a blanket statement applicable to all translation situations.

Some theorists have associated the ethics of translation with specific political imperatives that require alternative strategies to the foreignizing approach advocated by Venuti. Describing the French Canadian context, Annie Brisset’s work is a case in point. Brisset argues that “The task of translation is thus to replace the language of the Other by a native language. (…) Translation becomes an act of reclaiming, of recentering of the identity, a re-territorializing operation. It does not create a new language, but it elevates a dialect to the status of a national and cultural language” (Brisset, 2001, p. 346). By defining the task of translation (rather than the translator) in concrete political terms, Brisset’s definition responds to specific asymmetrical relationships, primarily between English, metropolitan French and the vernacular form of French spoken in Quebec. These relationships are a consequence of the encounter between two or more particular languages and cultures. In moving towards greater specificity in the translator’s task, Brisset’s definition is reflective of a wider appreciation of the need for context-specific ethics, able to acknowledge particular power dynamics, agencies, and historical asymmetries, all of which are necessary for a full ethical understanding. From this perspective, the task lying before translators is to engage not with situations of equivalence but to become involved with the asymmetries of difference in what Venuti calls “an ethics of location” (Venuti, 2001, p. 341). The recognition of power differentials and their extensive effects requires different strategies, but shares a general ethical mistrust of hegemonies. The task of the translator then becomes an ethico-political intervention motivated by a desire to rectify inequalities and deconstruct abusive power systems via a flexible range of strategies suited to the particular situation.

An example of the assumptions a resistant translation might challenge is the unmarked gender of “the translator” and hence for Gayatri Spivak “the task of the feminist translator is to consider language as a clue to the workings of gendered agency” (Spivak, 1992, p. 177). With Spivak the translator’s gender is finally re-defined: he becomes she, and she brings a new relation to her others, to whom she is bound by an intimate responsibility: “The task of the translator is to facilitate this love between the original and its shadow, a love that permits fraying, holds the agency of the translator and the demands of her imagined or actual audience at bay” (Spivak, 1992, p. 178). This feminist, post-colonial approach, which posits love as a bridge across differences, is a recognition of the asymmetries upon which translation is based. Spivak emphasizes the intimacy that must exist between the translator of a primary and secondary text and views this as a response to the fundamental alterity of translation. For Spivak the translator is involved in an inherently ethical relation, since “It [translating] is a simple miming of the responsibility to the trace of the other in the self” (Spivak, 1992, p. 177). The act of translating thus becomes associated with the foreign in the self, rather than serving the re-assertions of the imperial self.

One final definition that must be mentioned in any contemporary analysis of the translator’s tasks is Jacques Derrida’s re-reading of Benjamin’s text. Derrida concentrates on the role of the translator in ensuring the survival of the original, a position which turns conventional hierarchies on their head since it makes the original dependent on its translations. Derrida explains this re-scripting of the translator’s task in Roundtable on Translation: “the task of the translator is precisely to respond to this demand for survival which is the very structure of the original text. […] Translation augments and modifies the original, which, insofar as it is living on, never ceases to be transformed and to grow. It modifies the original even as it also modifies the translating language” (Derrida, 1982, p. 122). Returning to Benjamin’s notion of the translation as ‘afterlife’[4] or ‘survival’ (Benjamin, 1997, p. 153), Derrida emphasizes the way in which a translation supports, rather than fails, an original, and like Benjamin he views this as an opportunity to enrich the receiving language. This inversion of established norms is a useful strategy for re-calling the ethics of alterity, that is an ethics that acknowledges connections rather than effacing them. In a parenthetic sentence reminiscent of Levinas’s descriptions of the ethical call of the face, Derrida emphasizes that Benjamin’s essay concerns the imperatives to which an ethically responsible agent responds: “Notice Benjamin does not say the task of translation but rather of the translator, that is, of a subject who finds him/herself immediately indebted by the existence of the original, who must submit to its law and who is duty-bound to do something for the original” (Derrida, 1982, p. 122). In Levinas’ ethical philosophy this relation of indebtedness is associated with the alterity accessed by feminine subjectivity; Ettinger’s feminist theory provides us with tools for thinking through translation and the actions of translators in new ways that are inclusive of the differences on which translation is grounded.

Ettinger’s Feminist Matrixial Model

Ettinger’s theoretical intervention is one of several feminist responses to Lacanian psychoanalysis. Her approach is unique and innovative however in that rather than focusing on a disruptive space, as in Julia Kristeva’s pre-Oedipal Chôra, or the doubling found in Luce Irigaray’s work, Ettinger introduces matrix as a feminine Symbolic concept that supplements, rather than replaces, Lacan’s masculine phallus. The concept of matrix represents a feminine dimension relegated to foreclosure or pathology in traditional psychoanalysis. Matrix gives the feminine an access to meaningful signification where in Lacan’s theory this is impossible: Woman, Real, and Thing cannot speak; they are gaps in the Symbolic realm. For Ettinger, signification is possible within a matrixial relationship, through exchanges that transgress the usual construction of subject boundaries.

Ettinger describes a co-constitutive subjective stratum formed in late pre-natal relationship as “subjectivity-as-encounter.” Arguing that “Several comes before the One,” (Ettinger, 1992, p. 200) she describes a structure of severality that precedes individual consciousness. This radically alters founding philosophies that are based on an a priori that consciousness is always, and originally, singular. It challenges that discrete subject of Descartes’ cogito ergo sum who represents the individualistic bias of the dominant tradition of Western philosophy, by protesting that “the moment of birth doesn’t have to present a mental barrier” (Ettinger, 1994, p. 50). In the words of art historian Griselda Pollock, Ettinger’s approach means viewing “human subjectivity not only as the effect of the castrative cut so deeply lodged in cultural theory by the works of Freud and Lacan, but as operating in shifting parallel, as encounter” (Pollock, 2004, p. 25). Envisaging relationships from a model that starts with encounters, instead of one that tries to find a way in which a meeting might be possible between subject and object, self and other, Ettinger’s approach has implications for many different fields. Given the communicative challenge of translation, her work is extremely suggestive as a means of re-conceptualizing translation since it offers an alternative to approaches that are structured by a presence/absence binary.

Pollock (Ettinger, 1999, pp. 25-26) emphasizes that a matrixial model is not essentialist and that the matrixial paradigm is distinctive because, unlike other theories, it avoids repeating the dynamics of the phallic structure: “qu’il s’agisse d’un organe femelle tel que le placenta ou d’une marque/blessure indifférenciée telle que l’ombilic, il s’agit encore d’une articulation au modèle présence/absence, coupure et perte par castration, et retour hallucinatoire du refoulé. La Matrix, par contre, concerne le processus subjectivisant de plusieurs sujets-partiels […] ici la plusieurité est originaire.”[5] By starting from a position of multiplicity, rather than trying to find our way there individually, we are engaged in very different dynamics to those that typify the Cartesian frames of Western thinking. For translation this means overcoming the hard separation between original text and translation to place them in a more reciprocal relation.

Bringing Ettinger’s model to the field of translation studies, I am struck by its relevance to the encounter that takes place in a translated text between two or more languages and cultures. An awareness of subjectivity-as-encounter, and the articulation of nuanced positions that are obscured by a phallic Symbolic structure, illuminates many aspects of the practice of translation. A matrixial spatial configuration renews our visions of exchange—or translation—processes. Ettinger’s use of the term “matrix” shifts its associations from the womb as passive receptacle to that of an active borderspace that is transformed by a co-emerging I and an unknown non-I. Ettinger points out ways in which matrixial subjectivity-as-encounter is systematically overlooked in many different fields: “the internal non-I is defined, in our culture, from biology and through immunology to psychoanalysis, as negative and threatening to the I” (Ettinger, 1994, p. 42). The same can also be said of translation, in which the demand for “fluent” translations acts as an inoculation against the supposed dangers or contamination of a foreign text, and the relationship of the source text to the translation is conceived of only in terms of degenerative linear replacements.

From an ethical point of view, Ettinger’s expanded understanding of subjectivity allows for a fresh relation to foreign others: in matrixial relations, the other risks neither assimilation by the self, nor rejection. While phallocentric relations are characterized by a tendency towards polar extremes—I move towards others to recuperate them as same in an empathetic, identificatory fusion, or I move away from others in a rejection of intolerable distinctness—matrixial patterns are less clear cut: I am beside, in proximity without knowing the other, and without needing to domesticate the other through controlling familiarity, or needing to expel into exile others who are not the same as me. This more nuanced approach to encounters with difference is the matrixial ethical praxis Ettinger elaborates, whose specific implications for translation studies this article elucidates.

The social and political implications of allowing matrixial strata to emerge are potentially revolutionary, not so much in the sense of the disruption advocated by Venuti or Kristeva as in the sense of furthering concrete changes in negotiating practices. I have argued that in a literary context this shift significantly alters relations between author, text, and reader, and points to the ethical role of literature; in the context of translation studies, a matrixial negotiation of difference has insights to contribute to both translation practice and theory. In the next section I explain further how a matrixial paradigm extends our understanding of the processes of translation.

Metramorphosising Translation

This paper proposes shifting from a conception of translation as metamorphosis, that is, a transformation in which a primary form is replaced by a second manifestation that obscures the original entirely, and from a conception of translation as metonymy, in which the original only ever receives partial rendition in translation, towards a view of translation as metramorphosis. I draw out the ethical implications of the metamorphic, metonymic, and metramorphic approaches to translation to show how a metramorphic approach is more suited to translators who see their task as that of intervening in asymmetrical linguistic and cultural situations, as compared to translators whose main concern is with establishing relations of correspondence across what is treated heuristically as an equal plane.

Moving away from a traditional model of translation as metaphor, Maria Tymoczko advocates viewing translation in terms of metonymic substitution. She establishes the ground for this change in perspective by describing the approach based on metaphor: “translation has been conceptualized chiefly as a metaphoric process, a process of selection and substitution in which the words of one language are selected so as to substitute for the words of another language” (Tymoczko, 1999, p. 279). Countering the trope of metaphor with that of metonymy, Tymoczko argues that all translations substitute a part for the whole, that in fact, all translations are metonymies. Although Tymoczko’s metonymic model seems close to a matrixial model in its focus on “the metonymic processes of combination, connection, and contexture” (Tymoczko, 1999, p. 284), ultimately it maintains the classic separation between original and translation, accepting the deficiency of the translation within a context of possible strategies for manipulating power relations, but nonetheless re-inscribing a culture that tends to start from discrete relations rather than connections. Even while it participates in a signifying chain, Tymoczko’s metonymic approach reinforces an ethics founded on individuality, rather than alterity, since the presence of each signifier is based on the absence of the preceding ones. In contrast, I am looking for ethical relations that recognize and foster severality in translation, where presence in one language is not premised on the absence of another.

The classical concept of metamorphosis is often evoked as an image of the translation process, representing the change of form that takes place when a text migrates from one language to another. To cite just one theorist who referred to this model, Vladimir Nabokov described his own self-translation in terms of a metamorphic process: “This re-Englishing of a Russian re-version of what had been an English re-telling of Russian memories in the first place, proved to be a diabolical task, but some consolation was given me by the thought that such multiple metamorphosis, familiar to butterflies, had not been tried by any human before” (Beaujour, 1995, p. 722). I wish to shift away from the dramatic metamorphic change that Nabokov evokes in his description of translation as total mutation from one form to another, from larvae to butterfly through a hidden moment in the chrysalis, to acknowledge the fleeting moment caught by Italian sculptor Gian Lorenzo Bernini in Apollo and Daphne, a masterpiece in white marble housed at the Villa Borghese in Rome (a photograph of the sculpture is available at www.galleriaborghese.it/borghese/en/edafne.html). In Bernini’s sculpture, Daphne is in a state of becoming, one leg the trunk of a laurel tree, her wild hair vegetal already, the tips of her fingers sprouting leaves. Faced with this uncanny transmutation, Apollo’s face is struck with horror and fascination. Daphne represents the uncanny feminine ability to make space, to withdraw and contract to allow for creation: this is what Ettinger has described as metramorphosis, and this is the model of translation this article explores.

Ettinger describes the exchange processes that occur within a matrixial paradigm with a neologism: “metramorphosis.” The term is composed of “metra” and “morpheus”: it combines a play on “meta” with an evocation of “mater,” mother or womb, and “morphe”, Greek for “form,” which is also linked to Morpheus, the Greek God of sleep and dreams. Ettinger’s project can be summarized as a theorization of how the matrix offers a locus where meaning is generated rather than foreclosed, transferred rather than buried. “Metramorphosis” refers to the transformations in meaning accessed by the layer of subjectivity-as-encounter, and in this sense it promotes a view of translation as generative, rather than as deterioration, dereliction or replacement of the original.

Where in Lacanian psychoanalysis the phallus, and the entire signifying Symbolic realm, is associated with only two unconscious processes—metaphor and metonymy—the concept of matrix invokes metramorphosis as a third unconscious process. This process is distinctive in that it does not follow the routes of masculine Oedipal castration. This means that metramorphosis is not structured by the on/off binary of presence and absence that is usually taken as the only possible logic for signifying processes. Metramorphosis refers to processes that do not involve single unities acting through the condensation of metaphor or the displacement of metonymy; instead they provoke changes that mutually alter the meaning they create, without supplanting or deferring the signifier. Rosi Huhn explains clearly how metramorphosis is distinct from metamorphosis: “In contrast to metamorphosis, each of the new forms and shapes of the metramorphosis does not send the nature of each of the preceding ones into oblivion or even eliminate it, but lets it shine through the transparency, disarranges and leads to an existence of multitude rather than unity” (Huhn, 1993, p. 8). Metramorphosis supplements metamorphosis by offering a less totalizing, or complete, form of transformation: it is a transformative process, but one which does not mask or efface its origins; rather, it expands them in a connective border-state where “source” and “target” meet creatively, recognizing a shared heritage and continuing ancestral lines. This conception opens up the space that Benjamin was looking for, and found in Rudolf Pannwitz’s conception of translations where the contact of languages produces a regenerative effect by “allowing it to be put powerfully in movement by the foreign language”[6] (Benjamin, 1997, p. 163). Metramorphosis goes beyond Pannwitz’s inversion of conventional models of fluent translation by engaging a mobile two-way process that is mutually affecting.

Translation studies can benefit from supplementing the traditional transformatory metaphor of metamorphosis and Tymoczko’s suggestion of metonymic retellings with the concept of metramorphosis through which Ettinger refers to a feminine process of change. Metramorphosis differs from metamorphosis in that it does not involve a total, complete transformation, but rather indicates an expansion or development. In a metramorphic exchange, relations of difference are established within “severality,” rather than between discrete units. Textually, this means that a translation does not overlay or efface its source through an ideal of equivalent matches, but expands it through a visible interaction in which the source is still present within the translation. Translation thus becomes not only the negotiation of differences, but also of analogies or similarities, generational lines that traverse the previously highly fortified border between text and translation.

A matrixial conception enlarges our vision of linguistic and cultural interactions that occur in translation by offering the concept of metramorphosis as an additional theoretical tool for expanding our understanding and practice of translation. I now turn to a discussion of one particular way this model allows for the potentially destabilizing effects of translation by allowing for and acknowledging affective mechanisms at work in the translation process.

The Uncanny in Translation

After recognizing the implications of asymmetrical relations on the translation process, it is also helpful to look at how translations produce, and are produced by, affective engagement. I suggest that one of the hallmarks of an affective translation is that it makes space for the uncanny within the text. In Freudian psychoanalysis, the Unheimlich is the uncanny return of the repressed familiar or homely; in the context of translation studies, I would like to look at how remainders of the foreign within the translated text disrupt and disturb a familiar or homely reading, resulting in affective readings that challenge an unproblematized identification of the homely and the foreign, thus demonstrating their inter-dependency, rather than their distinctness.

Metramorphosis offers an alternative means of describing translation because it reflects processes that involve uncanny remains and unconscious interference, rather than a smooth and complete transition or perfectly equivalent equation that would result in the traditional ideal of a fully conscious fluent translation. If we wish to problematize notions of unfrayed originals and seamless translations, by which I mean translations and originals that live complete and separate existences, the notion of metramorphosis, as manifested in the matrixial exchanges that take place between an original and its translations, gives us ways to think through alternative relations and untidy remains within translated texts.

Just as metramorphosis supplements metamorphic translation models, Ettinger also develops the repertoire of affect identifiable in the translation process, by exploring a feminine uncanny alongside the traditional (masculine) model of the uncanny. In Ettinger’s The Matrixial Gaze she returns to Freud’s The ‘Uncanny,’ to remind us that Freud made a distinction between the castration complex and the maternal womb/intra-uterine complex. When Freud (1925, Vol. XIX, p. 244) interpreted the uncanny fear of being buried alive, he recognized the existence and effects of womb phantasies, commenting “this terrifying phantasy is only a transformation of another phantasy which had originally nothing terrifying about it at all, but was qualified by a certain lasciviousness—the phantasy, I mean, of intra-uterine existence.” On the basis of this admission of a feminine uncanny, which is usually stifled or overlooked, Ettinger (1995, p. 8) presents an interesting qualification and inversion in a common understanding of the fear associated with uncanny experiences by suggesting that “while the castration phantasy is frightening at the point of the emergence of the original experience before its repression, the (what I call) matrixial phantasy becomes frightening only when the experience is repressed.” This inversion in the mechanisms of fear in the uncanny has significant implications for ethical relations since it provides a motive for identifying and fostering matrixial encounters.

Practicing Metramorphic Translation

To illustrate the theoretical points I have been making, I shall now discuss three instances where a metramorphic approach enables an understanding of the foreignizing or uncanny processes at work within a translation. Each of these examples concerns multiple languages, where a conventional approach to translation would assume that there are never but two; all reflect the need for translation studies to access more complex models for understanding the ethical negotiations of linguistic and cultural differences in both the original and translated text, and show how a matrixial paradigm responds to this need.

In 2000 my husband and I co-translated Fatima Gallaire’s play Les Co-épouses (1990) as House of Wives (trans. Shread, 2008) for production in the United States. Although she is Algerian by birth, until recently Gallaire wrote exclusively in French and lived for many years in France, where her plays have been published by established French publishers; her work is regularly performed in theaters all over France; and she has received success and widespread recognition. Despite this apparently ‘unproblematic’ French surface, we discovered that, as with many Francophone texts, a repressed alterity was sewn into the work. In the process of translating the play, we became increasingly aware that beneath the French in which it was ostensibly written, there was a stratum of Arabic, which inflected the rhythm of the discourse, contributed to a certain solemnity in conversational exchange, and modulated the French. Consequently, our translation retained some formality, repetitiousness, and even apparent redundancy, particularly in greetings, blessings, and imprecations to God, which, being integral to Arabic were carried over in Gallaire’s French, and then imported to an English which commonly no longer has any use for them. To have translated a formal ritualistic Arabic greeting, where the health and well-being of each member of an extended family is enquired into one after the other, into the casual terseness of an American ‘hi, what’s up?’ or ‘how you doin’?’, for example, would have been to betray and lose an entire set of cultural referents, a specific relationship to space, time, and social interaction. We sought instead to inflect the English of our translation, allowing the Arabic to come through the French and into the English. We also held on to the pervasive use of the conditional tense as a sign of deference and respect, so important to a system of highly stratified honor, but again superseded by the direct bluntness of modern American English. In the American campus context in which the play was produced, we occasionally took advantage of our native British English, more associated with formality, to achieve this effect.

In these various translation strategies, a metramorphic approach to the text helped attune the intuition that we were dealing with more than just a French text and helped us seek inflections in the English that came not from French, but from Arabic. The play was set in Algeria, and we sought to retain the specificity of its cultural tone. The experience of linguistic and cultural interference in translating Francophone texts, where there is a particularly complex historical and cultural relationship with the French language, is common; our experience with House of Wives showed the importance of acknowledging how these relations play out in both the translation and production process.

As the translation progressed, we became increasingly aware of both the Arabic and Berber language and culture underlying the text. In translating the play, it was therefore critical that we understood that the source text was neither “fluent” nor homogenous, that the French was strained through an Arabic lining. Where a domesticating approach might efface markers that were so powerful that even the original French was subject to their distortions, a translation open to metramorphic processes was more attuned to the instability of the ‘original’ and sought to retain that uncertainty in the English. In this sense, a metramorphic approach to translation affects not just the translated text, but also the way in which the original is read. Where the conventional ideal of fluency in translation is predicated on a prior assumption that the original itself is fluent, a matrixial approach helps us hear and respond to those instances where the source text is heterogeneous. This is a far more widespread phenomenon than traditional translation strategies allowed for—in the end, perhaps all Francophone texts display varying degrees of instability, and this in turn may lead us to question even apparently classic French texts for their own regional differences and variations. By allowing for greater complexity in the source texts, a metramorphic approach helps us achieve more nuanced translations.

While a text such as Les Co-épouses may appear to be accessible to any French speaker, further examination reveals the subversive, innovative, and highly specific character of Francophone writing. Even if Gallaire’s intent in this play is not consciously militant, a guerilla metaphor used by another Maghrebi writer, Abdelwaheb Meddeb, evokes her style: “L’écriture française nous ‘livre’ à l’autre, mais on se défendra par l’arabesque, la subversion, le dédale, le labyrinthe, le décentrage incessant de la phrase et du langage, de manière que l’autre se perde comme dans les ruelles de la casbah”[7] (cited by Mehrez, 1992, pp. 123-124). An example of this guerrilla effect is the uncanny experience we encountered in translating the names of the characters in the play. As the translation neared completion, we began to feel not a homely proximity to a text we had lived and worked so closely with, but rather an increasing distance from it, a slipping away from what we had assumed to be a privileged access via the French language. This culminated in an ironic moment when a Moroccan woman whom we had asked to act as cultural advisor for the play’s production, informed us that the names of the characters, which we had unwittingly taken as ‘foreign’ Arabic names, were not in fact that at all: they were not names but adjectives, making the characters allegorical representations, rather than individuals, and hence dramatically altering our interpretation of the play. Thus, when a North African Arabic speaker looked at the list of characters she saw the outline of a plot, where we saw merely an abstract family. This foreignizing was present in the source text itself, even before the translation process had begun, but remained as a mark in the translation where it might be understood by a speaker of Arabic as that uncanny sign of cultural repression borne by the texts. When the play was published, we decided to draw attention to this textual feature by glossing the names of the characters with the allegorical associations, for instance Nahnouha, the mother in law means ‘busy body’ in Arabic, Mimia, the second wife is ‘nice’ or ‘gentle’ in Berber, and Chems, the eldest daughter, and voice of future generations, is ‘sun’ in Arabic.

My second example is the work of novelist Dany Laferrière, a Haitian writer who spent many years in exile in Montreal and now lives in Miami. Laferrière took the insight that texts may speak several languages to the realm of paradox. I have explained that as translators, we must be aware of the potential of one language to mask another within a text. This intuition lies behind a seemingly paradoxical claim Laferrière made about one of his own texts: “Ce livre est déjà écrit en anglais, seuls les mots sont en français.”[8] While his lapidary remark neatly points to the need to expand our understanding of translation to give equal weight to both cultural and linguistic factors, in interview, Laferrière also explained that given the colonial weight of French, and the fact that in any case French was only a secondary language overlaying his mother tongue, Kreyol, “je choisis de devenir un écrivain américain écrivant directement en français”[9] (Laferrière, 2005). I shall discuss the ambivalent relationship between Haitian Kreyol and French further in my third example, but for the moment I focus on Laferrière’s juxtaposition of French and English in text and translation.

Laferrière’s striking statement was confirmed by his translator David Homel who brought the ‘English’ book into English. Nevertheless, Lee Skallerup has argued that Homel gives “entirely too much authority to an off-hand and ironic comment made by the author” (Skallerup, 2005, p. 124). While Skallerup’s point that “accepting that the novel is in fact an English novel negates other non-anglo influences and linguistic tensions Laferrière exploits in his narrative” (ibid., p. 125) is well taken for the specific textual examples she presents, for instance, the partial erasure of the reference to French feminism in the chapter title ‘On ne naît pas Nègre, on le devient’ which Homel translated as ‘You’re not born Black, you get that way’, or the omission of Emile Ollivier as a reference, I would not characterize Laferrière’s comment as off-hand. Rather, given Laferrière’s long career as a canny provocateur, I think that in suggesting that his novels are more closely aligned to American language and culture than to his Franco-Caribbean origins, Laferrière is deliberately seeking to disrupt categories of thought, played out here in a questioning of the assumption that the task of translation is to translate between different languages.

Laferrière’s position as transnational writer, whose ten volume “American autobiography” spans his account of immigrant life in Montreal in Chronique de la dérive douce (1994) and his return ‘home’ in Pays sans chapeaux (1996), moving between the multiple locations of the Haitian diaspora—Port-au-Prince, Montreal, Miami—is also indicative of the inability of a traditional nation state model of translation to represent increasingly common complex transnational realities, in which it is not a matter of either/or, but rather and/ and /and also. Cronin (2006, p. 61, citing Glick-Schiller) defines this alternative allegiance succinctly as “the fact of being attached to or experiencing two places simultaneously.” With his particular predilection for mobility, as evinced by his most recent title Je suis un écrivain japonais (2008), Laferrière is not about to be characterized definitively—as Haitian, French, or English. Again, a matrixial model is better equipped for understanding how such an apparently contradictory state of being might emerge and be expressed in a translated text.

My third example brings together the concerns of the first two instances of metramorphic translation practices in a new context. In a recent translation of Marie Vieux-Chauvet’s novel Les Rapaces (1986) from a Haitian French that is for the most part a very ‘correct’, standard French into American English, I decided to introduce a third language into the translation. Although very little Kreyol is apparent in the source text, a metramorphic translation heard it emerge on the horizon of the French. This resonance alone might not have been a justification for introducing Kreyol into the text. In fact, my reasons for doing so were directly related to the intended audience of the translation: first and second generation Haitian-Americans living in the United States who do not speak French, and who cannot therefore access their Francophone literary heritage. Within this sizable migrant population, commonly known as Haiti’s 10th Department, many Haitian-Americans speak Kreyol at home. Consequently, although Kreyol is not present in the original text, its palpable presence as the subtext or context of the original helps explain its uncanny appearance in the English translation.

This metramorphic intervention was inspired by Françoise Massardier-Kenney’s discussion of her and Claire Salardenne’s translation of Claire de Duras’ Ourika (1823), in which she explains how “by translating from French into Wolof, rather than from French into English in strategic parts of the text, the translator can momentarily “withhold translation” to make the translation apparent, to restore multilingualism” (Kaddish and Massardier-Kenney, 1994, p. 23). Although the translation of Ourika was the stimulus for my decision, the specific context of my translation introduces an interesting twist, since rather than shutting readers from ex-colonial powers out as in Ourika, my importation of Kreyol acts as an invitation into the text for Haitian-Americans exiled from their literature by language. This opening of the original French text through the translation process is a form of generative expansion, an instance of the movement of tsimtsoum discussed in the following section.

Tsimtsoum in Translation

The conception of ethics in translation association with a matrixial model may be further explained with reference to a theological concept taken from Jewish cabbalistic thought. Literary critic Christiane Blot-Labarrère defines tsimtsoum as “l’impossibilité logique de la présence de Dieu au monde. Ce qui n’implique pas son inexistence. Dieu s’est retiré en lui-même pour que, dans ce retrait – en hébreu, le Tsimtsoum –, le monde ait sa place”[10] (Blot-Labarrère, 2000, p. 292). This conception of creation as generous withdrawal is very different to the Christian approach, upon which an entire tradition of individual accomplishment that makes something from nothing—l’être du néant—is based. Translation might be characterized first and foremost as a creation that recognizes that it does not start with nothing, but with a material that invites re-forming: indeed, it is precisely because of this that a chauvinistic tradition has denigrated translation as derivative and secondary. But what if we adopt the view that creation always starts from something? What if we recall that the new never emerges from nothing?

This feminine perspective is the subject of Nancy Huston’s Professeurs de désespoir in which she argues that “Les philosophes européens érigent en modèle de l’être humain un homme solitaire, rationnel et autosuffisant”[11] (Huston, 2004, p. 25) despite the indisputable fact that “Ils ne surgissent pas de nulle-part”[12] (ibid., p. 77). Huston argues that “tous ces systèmes, sans le savoir et donc forcément sans le dire aussi, sont fondés sur l’oubli de l’enfance, l’oubli de l’enfant, de l’enfantement, l’ignorance de ce que tous les autres, obscurément, savaient sans le dire”[13] (ibid., pp. 36-37). Taking Huston’s point a step further, it is clearly a systematic oversight of the patriarchal perspective that encounters with unknown others are not considered a part of the development of subjectivity. Construed here paradigmatically in terms of the encounter between the mother and post-mature infant in late pregnancy, this severance of an interconnected stratum of subjectivity becomes culturally endemic. As a result, the opportunity for an alternative grounding for inter-subjective relations is lost.

The denial of inter-dependency is particularly acute and poignant in the field of translation, as Michael Cronin points out in Translation and Globalization (2003). Countering a prevailing cultural perception that consistently views dependency in negative terms, Cronin recognizes the mauvaise foi of philosophical models that will not recognize “the connection between human affliction and our dependence on others,” particularly given the “moral agents who are themselves presented as though they were continuously rational, healthy and untroubled” (Cronin, 2003, p. 39). Making the same point in a positive light, Cronin suggests that translation “shows the ethical value of dependency” and that it “reveals our multiple dependencies and the connectedness underlying the consoling fictions of absolute autonomy. It may be the sum of our debts that constitutes our true wealth as peoples” (ibid., p. 40). The notion of indebtedness as wealth counters the hegemonic free-market economy which seeks profit over disinterested gift and derides exchanges which manifest a more generous principle of care. Translation is potentially an ideal field in which to draw attention to this wealth, and yet, up until now, this potential has rarely been explored.

Ettinger’s psychoanalytic theory gives us the tools to amend this predominant philosophical bias. In a discussion about tsimtsoum with Levinas, Ettinger explains her artistic process, in terms that are analogous to the practice of translating: “Écrire comme suivre un centre qui fuit continuellement, peindre comme s’étirer avec, se rétracter, en se dérobant devant la connaissance”[14] (Ettinger and Levinas, 1997, p. 18). Ettinger’s description of her painting process, premised on her withdrawal, is parallel to the process whereby an original text withdraws to open up a space for its translations. It is in this sense that Derrida’s notion of the survival of the original is so intimately linked to its translation—a translation that expands an original from within, rather than detracting from its outside. This conceptualization of the space of translation is difficult from a phallic perspective that interprets the space generated by translation as either a no-man’s land or a threatening void. Vladimir Ivir’s discussion typifies this common response to “the existence of ‘lacunes’” (Vinay and Darbelnet, 1958), ‘blank spaces’ (Rabin, 1958), ‘gaps’ (Ivir, 1973, 1977), ‘voids’ (Dagut, 1978)” (Ivir, 1987, p. 36). Ivir’s anxiety over “filling the gap,” which generates a collection of compensatory strategies, posits the entire project of translation in terms of deficiencies, rather than a regenerative process.

One of the challenges of keeping the textual border a threshold, rather than a frontier, is that to do so is to insist on an intimate area of ignorance, the foreign that is within; this ignorance signals the original text as the ethical space of subjectivity-as-encounter. In a manner similar to Eve Kosofsky Sedgwick’s closet (Sedgwick, 1990), the matrixial model conceives of this space as a substance rather than a lack. In contrast to Lacanian phallic associations of the lack with the Real, Woman, and the Object, from a matrixial perspective this lacking space is signifying. Where the gap troubles phallic structures, producing an impulse to fill in the space of the foreign within the text, a matrixial model allows for such uncanny events. Furthermore, the encounter in the transitional borderspace of the translated text within the original offers the grounding for an ethical opportunity.

I have suggested that conventional models of translation do not adequately account for the further possibilities a matrixial model identifies. From a matrixial perspective, the creative space of translation is not a lack, an absence of the original, but rather a space created within the original by tsimtsoum, by the matrixial withdrawing and metramorphic processes that involve other presences in a joint space of severality. One of the tasks of translators is thus to enable an ethical space of creative encounter through the withdrawing that brings this severality into relation and allows for the emergence of subjectivity-as-encounter.

Conclusion

In recent years there has been increasing interest and recognition of the heterogeneity of the translated text, a heterogeneity that derives from linguistic and cultural effects and that reflects both the historical and creative complexity of translated texts. By endorsing the creative interplay that shows through a translated text, in the place of a new, smooth surface, I have suggested that Ettinger’s matrixial model enables us to understand more fully and engage more subtly in translating processes. I hope that my suggestion of metramorphosis will help expand beyond the notions of metonymic part replacement and metamorphic processes where one form is replaced by another, by drawing attention to instances where the substitution process is neither seamless nor complete, but where the accumulated layers of interaction come together in a translation that embodies an ethical space of encounter.