Poetics of the tangent: Dialect, translation, and the critique of linguistic ontology

Giancarlo Tursi
February 17, 2020
Online Only
"_DSC8730bw" by subotnikone is licensed under CC BY 2.0


In his 1963 For and Against Saussure, the sociolinguist Louis-Jean Calvet advanced the idea that structural linguistics doesn’t merely describe language; it invents it. Backed by Robert Godel’s reading of the manuscript sources for Saussure’s Course in General Linguistics, Calvet goes so far as to paint the invention of language as a forgery committed by the editors, who in their desire to grant the discipline autonomy put under erasure those aspects of the Swiss linguist’s thought which attended to the social and psychological determinations of language.

Though structuralism has come and gone, it may be argued that language as a whole is predicated on a system of exclusions; it is thus that the Greek word barbaroi designated not only those who could not speak Greek, but also those who could not speak Greek well, who spoke, say, with accents or in a dialect. In what follows, I explore the ways translation may be marshalled as an alternative to the linguistic paradigm. Taking my cue from a passage in Walter Benjamin’s “The Task of the Translator,” I argue that translation may be thought of as operating a linguistic tangent, one which, taking off from the circumscription of language, extends out to meet these very exclusions.  


Our translations even the best ones proceed from a wrong premise they want to germanify the sanskrit the greek the english instead of sanskritizing hellenizing anglicizing the german. They have much more respect for the usage of their own language than they do for the spirit of the foreign work […] The fundamental error of the translator is to hold on to the contingent status of his own language instead of submitting it to the violent force of the foreign language. Especially when he is translating from a very distant language he has to find his way back to the elements of language where word image sound converge he has to widen and deepen his own language by means of the foreign one we don’t realize to what extent it is possible to what degree a language can be transformed to what extent from language to language there is hardly more difference than from dialect to dialect but this not when one takes the matter lightly but on the contrary when one takes it seriously enough. (260, italics added, lack of punctuation intended, my translation of Antoine Berman’s French translation)

The only commentary Walter Benjamin provides for the above passage, taken from Rudolf Pannwitz’s 1917 Die Krisis dei europäischen Kultur, which he deems to be “next to Goethe’s reflections in his notes to West – östlicher Divan […] the best thing published in Germany on the subject of translation theory” (260), is a metaphor which seems to speak to its intrusion in the text: “Just as a tangent touches a circle lightly and at one point only and that it is this contact and not the point which assigns it the law according to which it pursues its straight path to infinity, so a translation touches the original fugitively and only at the infinitely small point of meaning, whereupon it pursues its own course according to the laws of fidelity in the freedom of linguistic movement” (261, amended translation). Indeed, so foreign is Pannwitz’s invocation of dialect to Benjamin’s notion of the “reine Sprache” (or “pure language”), it is as though the citation itself intervenes as a sort of tangent, deviating the author’s discourse on language, contaminating it with the presence of its dialectal other.  

Though he does not say so explicitly, the translation theorist Antoine Berman is one of the only readers of Benjamin who seems to seize on the significance of this textual intrusion when, in his commentary to the essay, he writes: “Benjamin’s concept of the pure language gains nothing if we seize it only in its messianic determination, since we can do nothing with this […] But if we seize it with Benjamin himself, and, beyond him, as the dialectal essence of language, then it becomes possible for us to appropriate it […]”(L’Âge de la traduction, second italics mine 181). It is thus that he proposes we read the expression “reine Sprache” as a “correlation of the Kantian expression ‘pure Reason,’ reine Vernunft,” which, “being founded on the pure language, precedes empirical languages and makes them languages” (24, translation mine) – a move which allows him to further ‘deviate’ the course of the Benjamin’s text, suggesting as it does that languages as apparently unrelated as “French and Chinese” may be seen as dialects of each other. Or, as he writes in his discussions Joyce’s translation of two fragments of his own Finnegans Wake into a “half Dantesque, half dialectal” Italian: “For the maternal heart of the maternal langauge all languages are close and related. Working most closely to this heart, the translator (of the letter) discovers the non-philological, non-linguistic relationsip between languages (Auberge, italics mine, translation mine, 142).

Indeed, everything happens as if Berman’s commentary were aimed at prolonging the very tangent introduced into Benjamin’s text by Pannwitz’s, taking off from a scrupulous and literal reading to reach greater heights, or plumb greater depths: “My commentary, as you can see,” he writes, “is not literal [‘n’est pas ligne à ligne’], it has ‘taken off’ [‘il a décollé’]. I believe that in closing – even if in too great a haste which doesn’t excuse everything – one has to take off [‘il faut décoller]” (181, translation mine).

One is reminded of Gilles Deleuze and Félix Guattari’s notion of the “line of the flight,” itself a figure of acceleration and also aimed at the ‘minorization’ or ‘deterittorialization’ of language. Of course, the sense of haste Berman refers to is quite real, this being the last lecture of the seminar series; it is precisely for this reason, however, that we would do well to dwell with this figure, giving it the time he was unable to owing to his early death at age forty-nine.


Drawing on Klaus Wagenbach’s study of the way the Prague writer’s German was influenced by Czech and Yiddish –, Deleuze and Guattari refer to Kafka’s writing in A Thousand Plateaux (1980) as “designat[ing] the line of flight or deterritorialization that carries away all of the assemblages” of the standard language (88-89). Their somewhat mathematical treatment of his work – they refer to it as a function – the “K. function” – which “differentiates” the respective “horizontal” and “vertical” axes of “submission” and “deterritorialization” of his milieu (88) – reveals an important homology between the line of flight and the tangent, that of speed or vectorization, which, as they write of the Canadian piano player Glenn Gould, “transforms musical points into musical lines” (4). Translated into linguistic terms, such a transformation supplants notions of discreteness with ones of continuity deducible from a “rhizomic,” dialectal situation, where, unlike “a structure, a tree, [or] a root,” “there are no points or positions” (4):  

There is no language in itself, nor is there a universality of language, only a competition of dialects, of patois, of argot, of specialized languages […] It evolves by subterranean stems and flows, along fluvial valleys, or train tracks; it spreads like an oil stain. (4, translation mine)

Where Deleuze and Guattari deploy images of bleed and spill, Pannwitz’s text mimics the idea of dialectal continuity to through its disregard for punctuation, pointing to the crisis in linguistic categorization brought about precisely by the realization that there is “hardly more difference” between a language and another than there is between a dialect and another (260). And yet, as Berman’s translation suggests, just as the lack of punctuation in Pannwitz’s text may be said to teem with ghostlier punctuations which the reader must interpolate if they are to make sense of the text, so it is not so much a matter of there being “no more difference” (“plus de différence”) between individual languages, but of there being no more difference between languages than between dialects (“plus de différence que”) – a difference (or “différance”) which only yields ‘more’ (“plus”) difference.

Of course, Deleuze and Guattari also seize on the importance of this dynamic between continuity and discontinuity when they refer to the “relative” (9) functions of point and line in the rhizome, writing that “every rhizome contains lines of segmentarity according to which it is stratified, territorialized, organized, signified, attributed, etc., as well as lines of deterritorialization down which it constantly flees” (9, Massumi’s translations). And it is this ambiguity, this indeterminacy, this relativity of point and line which characterizes the moment when a tangent crosses the curve of a circle, leading Descartes in the second book of his Geometry to refer to the tangent as “not only the most useful and most general problem in geometry that I know, but even that I have ever desired to know” (508, cited in Langer). For if, as a kind of “ghost of lines,” to use phenomenologist Tim Ingold’s words, the “infinitely thin,” “abstract, rational, conceptual” (47) line of Euclidean geometry could circumvent the issue by conceiving – as Apollonius did in his Conics – of the tangent as a line “such that no other straight [one] could fall between it and the curve” (22), differential calculus, from Leibnitz on, would problematize this solution by insisting on the possibility of infinitesimal subdivision, forever deferring the moment of actual contact. One is reminded of the Heisenberg uncertainty principle, which states that just as it is impossible to measure the position and velocity of an object at the same time, so it is impossible to distinguish a particle from a wave; similarly, one cannot say with certainty whether what happens at the moment when a tangent crosses a curve is a transformation of points into lines – as Deleuze and Guattari would have it – or a transformation of a line into an infinite set of points – as Pannwitz’s spectral poetics suggest.

Deleuze himself homes in on the figure of the tangent in The Fold: Leibnitz and the Baroque. Taking off from Paul Klee’s studies of the figure as it crosses, not a circle, but “an active, spontaneous line” (14), Deleuze refers to the inflection point, or the “point-fold,” as “the pure Event of the line or of the point, the Virtual, ideality par excellence […] the World itself, or rather its beginning, as Klee used to say […] ‘a site of cosmogenesis’, ‘a nondimensional point’ ‘between dimensions’” (15). “Weightless,” being “neither high nor low, neither right nor left, regression nor progression” (15), suspended between convexity and concavity, inside and outside, ownness and otherness, the “intrinsic singularity” (15) of the point of inflection might, in other words, be said to index a state of ontological non-belonging, one whose implications for translation studies we would do well to explore.


In his seminal 1953 Languages in Contact, Uriel Weinreich wrote of the “structural no man’s land between two phonetics systems” (14, in Martinet; italics mine) produced by the “actual sounds” of the bilingual – a formulation which might be brought into alignment both with Derrida’s claim in Monolingualism of Other (1996) that “a language doesn’t belong” (“une langue, ça n’appartient pas”) (18) and Emily Apter’s appraisal of the stakes of Barbara Cassin’s Dictionary of Untranslatables (2014), as she refers to the “glimpses of languages in paradoxically shared zones of non-national belonging” which a study of the way Untranslatables translate affords us (xv, italics mine).

Here, one would need to adduce still broader philosophical critiques of ontology, as Apter herself does in her Against World Literature: On the Politics of Untranslatability, as she evokes Graham Hartman’s notion of a “vicarious causation” (190). According to this model, “forms do not touch one another directly, but somehow melt, fuse, and decompress in a shared common space from which all are partly absent,” meeting only “on the inside of a third,” a space which it might be helpful to think of in terms of translation’s own mediating thirdness. (190). One would be tempted to enlist Henri Wissman’s Thinking Between Languages, if it weren’t for the fact that it considers that “middle language” as “still a language” (“encore une langue”) (20) – a notion which leads him to exclude Germany’s “dialectal diversity” from his Humboldtian conception of a “nation as an individual” (74) and a “linguistic individuality” (75, translations mine).


More useful, again, is Berman’s notion of a “third language,” which, though most fully developed in his reading of the role German plays in Pierre Klossowsky’s French translation of the Aeneid, actually consists in placing a dialectal interval between the source and target languages of a text, as he observes apropos of Joyce’s translation of two fragments of his own Finnegans Wake into a “half dialectal, half Dantesque” Italian:

Moving in its non-normed space, the translator works, as we have said, on what is most maternal about his language […] For the maternal heart of the maternal language all languages are close and related […] This, another great 20th century could teach us exemplarily; that which Joyce did into a half dialectal, half Dantesque Italian of a fragment of his Finnegans Wake. Passionately, and almost maniacally, searching out the maternal heart of language, Joyce knew – instinctively – that the only way of prolonging (accomplishing) the polyphony of koine of his great work was to convert it into a dialectal polyphony, to fold the concert of intertwined koinai within the maternal space of dialects. (142, translation mine).

Benjamin’s (or was it Pannwitz’s ?) figure of the tangent may be said to find its ultimate prolongation here, in this middle, “non-normed space” between languages which, upon final analysis, doesn’t even belong to language.


Apollonius Of Perga: Treatise On Conic Sections With Introductions Including An Essay On Earlier History On The Subject, Ed. Heath, Thomas, New York, Barnes And Nobles, 1961. 

Apter, Emily, “Preface,” Dictionary Of Untranslatables: A Philosophical Lexicon, Ed.  Cassin, Barbara, Apter, Emily, Lezra, Jacques, Wood, Michael, Princeton, Princeton University Press, 2014. 

Benjamin, Walter, “Task Of The Translator,” Trans. Harcourt Brace Jovanovich, In Selected Writings, Ed. Bullock, Marcus, Jennings, Michael, Cambridge, Harvard University Press, 1996. 

Benjamin, Walter, “La Tâche Du Traducteur,” In Oeuvres, I, Trans. Gandillac, Maurice De, Paris, Gallimard, 2000. 

Berman, Antoine, La Traduction Et La Lettre Ou L’auberge Du Lointain, Paris, Seuil, 1999.

Berman, Antoine, L’épreuve De L’étranger: Culture Et Traduction Dans L’allemagne Romantique, Paris, Gallimard, 1984. 

Berman, Antoine, L’âge De La Traduction: ‘la Tâche Du Traductuer’ De Walter Benjamin, Paris, Vincennes, 2014. 

Berman, Antoine, L’âge De La Traduction: ‘la Tâche Du Traductuer’ De Walter Benjamin, Trans. Wright, Chantal, New York, Routledge, 2018.

Calvet, Louis-jean, Pour Et Contre Saussure: Vers Une Linguistique Sociale, Paris, Payot, 1975. 

Deleuze, Gilles, Guattari, Félix, Guattari, A Thousand Plateaus, Trans. Massumi, Brian, Minneapolis, University Of Minnesota Press, 1987. 

Deleuze, Gilles, Kafka: Pour Une Litterature Mineure, Paris, Minuit, 1975.

Deleuze, Gilles, The Fold: Liebnitz And The Baroque, London, Athlone Press, 1993. 

Derrida, Jacques, Le Monolinguisme De L’autre, Paris, Éditions Galilée, 1996.

Ingold, Tim, Lines: A Brief History, London, Routeledge, 2007. 

Langer, R.E., “René Descartes,” In The American Mathematical Monthly, Vol. 44, No. 8, Oct. 1937, Pp. 495-512. 

Wissman, Heinz, Penser Entre Les Langues, Michel, Paris, 2012. 

Weinreich, Uriel, Martinet, André, Languages In Contact. Findings And Problems, Berlin, De Gruyter Mouton, 1979.

All by
Giancarlo Tursi