Charting new territories: Multidimensionality of multilingualism

A discussion of the challenges and discoveries that arise in working with multilingual patients

By Olga Pugachevsky

It always starts with a patient. Or patients. They march straight into my thoughts as soon as they enter my office, and refuse to leave ever after, even when in the physical world they are already long gone. Not only that, they tend to alter my lived world. It would certainly be easier if they rearranged my office furniture instead.

One thing that each of them transforms is my speech. I have two working languages, English and Russian, fully developed as such, honed into precise professional tools, each steeped in the wealth of corresponding social, historical, and cultural discourse. I can also recognize some French and Hebrew, and that’s about it.

Nevertheless patients of mine manage to bring into being their very own, very specific enriched version of my “languageinstrument.” It is never the same again, even if it is technically pure English, or pure Russian. More often than not, however, it is a mix that might include both, as well as words and phrases from a third one, usually the patient’s mother tongue, which as they find their way into our discourse necessarily acquire new lived meanings, while the patient and I struggle to render their old meanings intelligible to both of us in the new context. Sometimes, usually when the patient’s mother tongue is utterly dissimilar to both of mine, it is experienced directly only by the patient, and is conveyed to me either through sounds, or renditions, or both, floating then around us like a multicolored maternal shadow.

Hope’s1 first language is Kirghiz, of which I have only a vague general idea, but by a twist of fate her second language, acquired at the age of six, is Russian, and her third, learned as a grown-up, in which she is also fluent, is English. At the beginning we decide on Russian as the language of our work, and proceed in that tongue, sometimes interspersing it with English expressions for the notions that it does not provide. The solution works well enough, though I am fully aware of my enormous disadvantage—the case of the aforementioned linguistic shadow. I have no direct access to the patient’s birth language, which retains the whole load of emotional, sensory, and perceptual vicissitudes linked to corporeal experiences within the primary relationship (Amati Mehler, Argentieri, and Canestri, 1990). Not only that, but because I cannot reproduce correctly even the sounds of Kirghiz, I cannot evoke by the prosody of my speech the state of the early “sonorous bath” (Harris, 1998), when my patient was “enveloped in the sonorous wrapping of the early maternal world; returning to a time when it was mother’s tongue that bathed the child in a sea of sounds” (Perez Foster, 1996, p.108; see also Anzieu, 1983). Thus I cannot take us via a direct connecting route to the patient’s early experiences, for I don’t have the keys to the door, do not possess what sometimes amounts to a single word that effectively encapsulates a whole multifaceted lived world. I don’t even know her real name, which of course is not “Hope,” but something perhaps similar sounding in Kirghiz. Or not. I am acutely aware that all the gentle nicknames, as well as all or most of the patient’s childhood traumas are interwoven with her mother tongue, as well as some of her dreams. When the time comes I’ll ask her not to translate but to retell, to concentrate on how the words sounded and felt originally, and then try to convey not only the meaning but also the thought, the feeling, the taste that they carry, and hope that it will work. We will have the shadow presence of her early Kirghiz, also the adult Kirghiz that she speaks now, different versions for different social situations, her early Russian, then again different versions of her adult Russian, and then all the variations of her adult English, from the businesslike speak she uses at work to the gentlest private whispers.

How many tongues will we really speak? Perhaps it takes a multilingual patient to throw in sharp relief the existence of all the hidden seams in the seemingly smooth structure of linguistic flow. She speaks in tongues, but also, like all of us, she inhabits many “dialects” within each tongue. Every person, whether mono- or multilingual, inhabits private, public, professional, infantile dialects, each giving voice to a different aspect of self (Amati Mehler, 1995).

From the point of view of the normal multiplicity of self, denoting the existence of individual self-states and their respective systems of lived relational meaning in constant kaleidoscopic flux, “each self-state a piece of a functional whole, informed by a process of internal negotiation with the realities, values, affects, and perspectives of the others” (Bromberg, 1996; Davies, 1996), with only a healthy, acquired, developmentally adaptive illusion of a cohesive “me,” an experiential state, it makes perfect sense that each self-state is encoded in its own “dialect,” the carrier that intertwines with the carried. The dialect comes into being as a necessary component of each relatively unlinked self-state, as it originates based on a particular internalized self-other dyad and reflecting the level of cognitive organization and developmental sophistication at the time, as well as self and object representations, experience of somatic body-self, and predominant affective tone. So it can be as sophisticated as a nuanced philosophical discourse or as simple as baby talk.

In that sense the analytic endeavor can be conceived as a continuous effort of translation, undertaken by both the patient and the analyst as they are attempting to translate the patient’s composite, disjointed, dynamic narrative into the common “language of analyses,” a permanent engagement in “Nachträglichkeit”—rearrangement, reordering, retroactive resignification (Canestri, in Amati Mehler et al., 1985, p.108). Whatever they can’t translate remains repressed, dissociated, lived out somewhere else. “A failure of translation— this is what is known clinically as repression” (Freud, 1896, p. 235).

The task is akin to translating oral poetry, where it is not the plain meaning of the text that is the primary objective but all the wealth of imbedded evocations, associated imagery, whether encoded in the word meanings themselves, or the rhythms, the rhymes, the sounds of the verse.

It is essentially an eternal search for corresponding signifiers, which in the end might result in approximation only, however close.

Here again a multilingual patient presents an extreme example of unavoidable “losses in translation” that apply to monolingual patients as well, but perhaps, being much more subtle in the latter cases, tend to be overlooked.

No translation is perfect, but there are many obstacles to even a “good enough” rendition of the original narrative. To begin with language in itself is not as much a system of neutral signs as a living substance, composed of endless sets and subsets reflecting and at the same time defined by the multiplicity of historic and social factors, constantly recording in itself the histories of societies as they evolve. Any text or personal discourse is inherently polyphonic, inhabited by multiple voices, styles, and enunciations, coexisting side by side without necessarily integrating, but often contradicting each other (Bachtin, 1981).

Every language has its own specific and peculiar structure, expressed at the phonematical, grammatical, and lexical levels. Hagege (in Amati Mehler, Argentieri, and Canestri, 1993) gives a good example regarding the sound structure of language.

He states that it is impossible to evoke by the same means the melancholy of the nasal vowel sounds when translating Verlaine into Japanese, because Japanese has no nasal vowel sounds. Melancholy has to be conveyed in other ways, which by definition will not evoke the same shades of feeling. One can add that the very pronunciation of different sounds would constitute quite another physical experience for the speaker, because it would involve the use of alternative muscle groups, and would thus activate diverse associative pathways.

As for grammar, one can think of languages where inanimate objects and animals are assigned a gender, which also vary from tongue to tongue. For example “death” is feminine in Russian but masculine in German, “vagina” is feminine in Russian but masculine in Hebrew and French. Sometimes words just do not exist: there is no proper word for “privacy” in Russian, for example.

Once a patient of mine bilingual in Spanish and English indignantly noted that in the first sentence of the original Spanish version of One Hundred Years of Solitude by Gabriel García Márquez, the character remembers how his father took him not to “discover” ice, as it is in the English translation, but to do something that is best but not precisely conveyed by the verb “to know,” with the implications of a deep, intimate, perhaps life-changing experience. She felt that the verb “discover” denoted the moment of coming upon something new, the moment of “uncovering,” rather than an encounter that leaves a permanent trace: the father taking his son “to know” ice.

Perhaps the situation is best expressed by Hagege through Jacobson’s formula: “Languages do not differ according to what they can or cannot express, but according to what they compel or do not compel one to say.”

At this point I am approaching again what I described earlier, but now locating myself at the theoretical rather than experiential coil of the same inquiry spiral. So, the next question goes, in what language are we, the patient and I, trying to translate the patient’s poetic narrative inscribed in a particular dialect, or rather, in what dialect of what tongue?

It has been noted that “as patient and analyst struggle to understand each other, to work through defenses, and to define and articulate the patient’s particular internal state of the moment they are in fact reenacting the intersubjective operations of the child and important other who long ago negotiated a consensual word meaning for and understanding of that very self-state” (Foster, 1992, p. 72–73). When analysis is conducted in one language, that language becomes a frame of reference within which the dialect of analysis becomes located. The dialect itself comes into existence because of the very nature of the process, in the repetition of the caregiver-child matrix. The very same course applies to a multilingual patient, except that here our cocreated common dialect is cobbled from bits and pieces of two or more framing languages, all of them present to various degrees at the same time, none of them ever silent, though some spoken and some not.

An access to the patient’s early lived experience can be gained through evocation of the palpable visceral feeling of the early speech, where every word within the interaction of the early self-other dyad was not just said, but produced physically as well as emotionally, where sound, tone, melody, volume mattered no less, but probably even more than the plain meaning, and where every utterance was not just heard, but received. That exchange needs to be revived within the unique dialect of the analytic couple, being described and rendered rather translated. Hope is trilingual, but even if we had one language between us, the process would be essentially the same ,though with less pronounced shifts, more subtle, elusive, less identifiable as such.

“What were the words that your grandma used to say to you as she was brushing your hair in the morning? Please try to remember what they sounded like in Kirghiz, imagine them.” We are speaking Russian. Hope closes her eyes, thinks for a minute or two, then, with her eyes still closed, she smiles. Then she tries to explain to me in Russian and English. Not to translate because it is not quite possible but to describe and explain. I try to understand the nuances as well as I can. Then she says firmly: “Yes, I totally forgot about that, now I’ll think of it every morning.” We had been looking for that little bit of nurturing that made it possible for her to survive all the rest that happened during the dark miserable years of her childhood for a quite a while, and now here it was. In Kirghiz, Russian, and English.

We both know that it is the beginning of a long road, for we have yet to access most of Hope’s earliest memories, coded initially in Kirghiz of course, and to retell them in our common tongue, doing our best to render the lived meaning, creating the new signifiers as we go, aware of the necessary revisions, and the fate of all that will remain lost in our translation.

  1. The patient’s name as well as any identifying details have been changed to preserve confidentiality.


Amati Mehler, J., Argentieri, S., & Canestri, J. (1990). The Babel of the unconscious. International Journal of Psychoanalysis, 71, 569–583.

Amati Mehler, J., Argentieri, S., & Canestri, J. (1993.) The Babel of the Unconscious. Madison, CT: International Universities Press.

Amati Mehler, J. (1995). The exiled language. CanadianJournal of Psychoanalysis, 3,87 –104.

Bakhtin, M. M. (1981). The epic and the novel: Towards a methodology for the study in the novel. In M. Holquist (Ed.), The dialogic imagination: Four Essays. Austin: University of Texas Press.

Bromberg, P. M. (1996). Standing in spaces: The multiplicity of self and the psychoanalytic relationship. Contemporary Psychoanalysis, 32, 509–535.

Davies, J. M. (1996). Linking the “pre-analytic” and the postclassical: Integration, dissociation, and the multiplicity of unconscious process. Contemporary Psychoanalysis, 32, 553–576.

Foster, R. P. (1992). Psychoanalyses and the bilingual patient: Some observations on the influence of language choice on the transference. Psychoanalytic Psychology, 9, 61–76. Foster, R. P. (1996). The bilingual self: Duet in two voices. Psychoanalytic Dialogues, 6,99–121.

Freud, S. (1896). Letter 52 from Extracts from the Fliess Papers. In J. Strachey, A. Freud, A. Strachey, & Alan Tyson (Trans.), The standard edition of the complete psychological works of Sigmund Freud, volume I (1886–1899): Pre-psychoanalytic publications and unpublished drafts, 233–239. London: Hogarth Press and Institute of Psycho-Analysis.

Harris, A. (1998). Psychic envelopes and sonorous baths: Siting the body in relational theory and clinical practice. In L. Aron & F. S. Anderson (Eds.), Relational perspectives on the body, 39–64. Hillsdale, NJ: The Analytic Press.