Abstract

Contemporary conceptualizing of translation tends to focus on space at the expense of other dimensions (e.g., time). Obviously, this is only part of the picture, as translating involves at least as much transforming as it does transporting, but the habit stems from looking at the word ‘translation’ through the distorted lens of Latin etymology. Distorted, because translatio did not primarily concern language matters in the Middle Ages (and even less so in Roman times, when other words were used). The etymological fallacy of reading translatio into ‘translation’ is facilitated in English by the fact that there is only one word for linguistic, mathematical, and religious forms of translatio, whereas other languages use the latter term exclusively in mathematics and/or the history of religion. English is the odd one out, in other words, but due to its current status as lingua franca in Translation Studies, this idiosyncrasy is somehow becoming the ‘natural’ default position, which betrays a profoundly disturbing monolingual bias. Comparing French, Spanish, Italian, German, and Dutch terms for ‘translation’ with English usage, this article shows the limits of such generalizations. In none of the above languages is the conflation of ‘translation’ and ‘space’ suggested as strongly as in English – some reservations can even be made with respect to German. When examined through a multilingual lens, then, the so-called ‘universal’ spatial imagery of transportation turns out to be a culture-specific catachresis, at worst a fossilized metaphor and at best a suggestive image: nothing less but nothing more.

Keywords: cognitive metaphors, translatio, etymological fallacy, catachresis, translation terminology

‘I said it in Hebrew—I said it in Dutch—
   I said it in German and Greek:
But I wholly forgot (and it vexes me much)
   That English is what you speak!’
(Lewis Carroll, The Hunting of the Snark)

In 1981, the Canadian rock band Rush released what would prove to be one of its most successful albums: Moving Pictures. The more relevant part for my present purpose is not the music but the artwork on the album’s cover.[1] Rather than featuring a film reel or anything else suggesting motion pictures or ‘movies’, it depicts movers in overalls who are carrying actual framed paintings (into the Ontario Legislative Building at Queen’s Park, Toronto, as it turns out). In addition to taking the title apart by reading the adjective ‘moving’ as a gerund, the image adds a third layer of meaning by showing spectators who are ‘moved’ to tears by the pictures being moved (one of which shows Joan of Arc at the stake). Three signifieds are thus attached to the same signifier, with meaning moving between them.

Something of the sort has been happening with the signifier ‘translation’, which is often posited as a matter of movement and motion, of transporting (texts) and transferring (meaning) from A to B, from one vessel to another. In spite of its impact as a ‘moving metaphor’ (perhaps in all the senses mobilized by the Rush album cover), this interpretation strikes me as both language-specific and etymologically reductive. Granted, the notion of ‘space’ was part and parcel of the Latin word translatio, but the latter more often than not referred to phenomena other than reformulation in a different language. Reading translatio into ‘translation’ is an etymological fallacy, therefore, which is facilitated in English by the fact that no other word is available. Most other languages either use an entirely different frame of reference (I am primarily thinking of non-European languages here, see Tymoczko 2005: 1087-88) or adopted a new term during the Renaissance, when Humanists suggested traductio.[2] The resulting semantic evolution is much less visible in English, where Medieval terminology was maintained. This “cultural exception” of sorts nonetheless risks becoming the norm because of the prominent position of English in Translation Studies (henceforth TS in order to distinguish a particular brand[3] of scholarship from translation as a more general field of enquiry), thereby adding to the confusion. Were it not for the fact that English has become the go-to language for discussing translation (in increasingly monolingual venues), this would hardly be an issue. The way things stand, however, the (over)use of spatial metaphors and (over)reliance on space as translation’s supposedly ‘deep’ dimension can make us go overboard, push us in unwanted directions, and lead us astray (all metaphors intended).

Moving beyond metaphors

The confusion between translation and transfer is a fallacy much in the way the (by now ‘old’) New Critics used to speak of an ‘intentional fallacy’ (Wimsatt and Beardsley 1946). In the present case, the fallacy is not only etymological but can also be labelled kinetic, in that it reduces translation to ‘kinesis’ (κίνησις), the ancient Greek word for movement and motion: not bodily movement, as in kinesiology, but loco-motion, movement from one space (Lat. locus) to another.

This echoes a point made by Alexis Nouss in his keynote address at the 2011 meeting of France’s Société de littérature générale et comparée (SLGC). Nouss detects a tendency towards what he calls ‘cinécentrisme’, not only in the overuse of metaphors like ‘bridges’ and ‘passageways’ but also more generally, in thinking about translation

[…] dans le paysage idéologique des sociétés contemporaines privilégiant les principes de mouvement et de circulation pour décrire leur nature et leur fonctionnement, marquées par le développement exponentiel des échanges et des transports à l’échelle planétaire, l’importance des migrations et les avancées technologiques de la télécommunication. En particulier, l’ensemble des réalités économiques et culturelles que recouvre le phénomène désigné comme globalisation accorde un rôle majeur à la traduction, censée répondre avec une fluidité et une rentabilité maximales aux besoins de transfert et de communication. (Nouss 2011)

By overemphasizing space, we risk throwing out the baby with the proverbial bathwater. I see three potential pitfalls. Two are conceptual in nature, while the third is more pragmatically linked to the ways in which TS has been branding itself in the 21^st century.

Using ‘translation’ as shorthand for translatio firstly overlooks the fact that the latter does not hinge on interlinguistic encounters or exchanges. During the Middle Ages, the ‘transfer of knowledge’ (translatio studii) happily happened within the confines of Latin; only belatedly did it involve ‘translating down’ into European vernaculars (Folena 1973: 65; Stierle 1996). Translatio imperii (‘transfer of rule’ or ‘power shift’) was not intrinsically linked to languages either… The Medieval notion of translatio, then, was more akin to our idea of ‘transfer’ than to today’s notion of ‘translation’ in its most common sense.

Similarly, anthropologists can write entire books on ‘translation’ without ever touching, let alone dwelling, on what is normally thought of as such. James Clifford’s Routes: Travel and Translation in the Late Twentieth Century Routes does not deal with languages but with movement and dislocation, with ‘culture in motion […] as a series of encounters and translations’ (Clifford 1997: 11). Tellingly, while his list of so-called ‘translation terms’ includes ‘diaspora, borderland, immigration, migrancy, tourism, pilgrimage, exile’ and so on, there is no trace of ‘interpretation of verbal signs by means of some other language’ (Clifford 1997: 11).

Confusing locution with location has another shortcoming. Indeed, not every switch in language is linked to, let alone dictated by, a change of locale. In many societies, long-term language contact has bred widespread bilingualism (or more precisely: diglossia) in one and the same spot. As a result, speech communities overlap, and languages share space, both private and public. Barcelona, for example, is not neatly divided into Catalan-speaking and Castilian-speaking neigbourhoods. Even the time-worn image of Montreal as being segregated along linguistic lines projects a divisive image of cultural apartheid on a reality that was much more nuanced: St Laurent Boulevard (nicknamed ‘the Main’), which divided West from East and, presumably, English from French Montreal, never functioned as some kind of Checkpoint Charlie.

In these and many more instances, spatializing language and translation does harm as well as good. The metaphor of movement acquires all the trappings of a troublesome trope. Rather than help us think, it turns an abstract concept into an obstacle to thought. It makes us confuse the (metaphorical) map with the (empirical) field, which is always more complex.

More irksome is the suggestion that the spatial reduction is somehow inscribed in the very word ‘translation’ and therefore seems ‘natural’ or ‘logical’. Even in English, where there is but one word for the many meanings (religious, mathematical, linguistic) acquired by translatio over the centuries, this train of thought amounts to an etymological fallacy, as we will see shortly. But English turns out to be a singleton in that it stands alone in having maintained Medieval terminology to describe linguistic exchange. In other European languages, translatio was replaced during the Renaissance with traductio, a word that gradually lost its spatial aura. As a result, the conflation of ‘translation’ and ‘space’ cannot be considered universal.

Metaphors: concepts and images

For all extents and purposes, then, English terminology is the odd one out, not the rule, a fact that risks being obfuscated by the wholesale adoption of English as the lingua franca in TS (on the latter, see Susam-Sarajeva (2002: 201) and Snell-Hornby (2006: 144)). According to a growing consensus in our interdiscipline, ‘Translation is movement. Translation is motion. This association is already established in the Latin word ‘transferre’, to carry across.’ This quote was taken from Klaus Kaindl’s excellent introduction to a volume on Transfiction (Kaindl and Spitzl 2014:1) but I could have used any among myriad quotes. In ‘From Transportation to Transformation,’ Rainer Guldin also feels that ‘translation implies a movement across a space in between as well as the transportation of a specific content from one shore to another.’ Seeing translation, not merely as a convenient image but as a ‘cognitive metaphor’ (Lakoff and Johnson 1980), he nevertheless wonders about the ‘function’ thus being ‘attributed to the space in between’ and about ‘the actual significance of the movement across.’ (Guldin 2012: 41) Three years later, in his book on Translation as Metaphor, Guldin has come to see ‘spatial metaphors’ as ‘a thorny issue, from which the field of translation studies is presently trying to emancipate itself.’ (2015: 50)

A possible reason for his malaise, I submit, is the cognitive thrust of the metaphor itself. For it is easy to get carried away (pun intended) when imagining translation in spatial terms. Take the following critique of an American translation of La Guaracha del Macho Camacho (1976), a Puerto Rican novel by Luis Rafael Sánchez. The translation was Gregory Rabassa’s, one of the most accomplished translators of Latin American literature (both in Spanish and in Portuguese) into English. In the course of extolling ‘the virtues and shortcomings’ of Rabassa’s work, Gerald Guinness starts waxing poetic:

In metaphorical terms, a translator is like a ferryman who has to hump cultural baggage across the wide straits dividing one linguistic and cultural land mass from another. But desperate these straits often are, and the traduttore-ferryman who attempts the transference often finds himself left with contraband goods on his hands, a traditore in his trade. (Guinness 1981: 109; 1993: 62)

There is little question that the (geographical, linguistic, cultural, social, political) divide between Puerto Rico and the continental United States is real, but it seems fair to say that metaphors can get the better of us.

When we let ourselves be swayed by metaphors, we lose sight of the fact that they are substitutes for the realities they refer to, not those realities themselves. We know of course that the Earth is not ‘blue like an orange’ (Eluard) and that a rose ‘by any other name would’ indeed ‘smell as sweet.’ Other images, however, provide such powerful shortcuts that they superimpose themselves on that what they are meant to refer to. Sometimes, they even short-circuit our thought channels, leaving us at a loss for non-metaphorical ways of designating things. We thus speak of the ‘arms’ of a chair, the ‘legs’ of a table, the ‘wings’ of an airplane, the ‘foot’ of a mountain, all realities that can no longer be referred to in any other way. This process, known as catachresis, makes us forget that metaphors are images rather than concepts.

Reducing ‘translation’ to movement between languages across space similarly creates a fossilized metaphor. There is no need for this to happen, however, no need to overemphasize space at the expense of other dimensions (such as time), no need to forget that translating is at least as much about transforming as it is about transporting.

Transfer and Transformation

Back in 1981, when TS was awakening to the importance of translation’s contextual embedding in what would later be termed the ‘cultural turn,’ Itamar Even-Zohar called for ‘translation theory’ to be subsumed under the heading of ‘transfer theory.’ In his view, which he reiterated in 1990, it was mainly a matter of exteriorizing the ‘implicit practice’ of discussing translation ‘in terms of transfer and vice versa’; ‘sooner or later,’ he prophesized,  ‘it w[ould] turn out to be uneconomical to deal with them separately.’ (1981: 2; 1990: 73) His was not ‘a proposal to liquidate translation studies.’ Even-Zohar instead wished to promote the study of translation as a far from ‘marginal procedure of cultural systems,’ while at the same time looking at a ‘larger context’ that would ‘help identify’ what was ‘really particular in inter-systemic transfer (translation).’ (Even-Zohar 1981: 3; 1990: 74) While he did distinguish ‘transfer procedures’ from ‘direct loans’ (Even-Zohar 1990: 82) his emphasis on the former nevertheless made short shrift of the process of transformation inherent in any type of transfer.

Between the two iterations (1981 and 1990) of Even-Zohar’s article, historians studying the interface between France and Germany came up with a notion of ‘cultural transfer’ that recognized just that. According to Michel Espagne and Michael Werner (who co-edited several volumes on the topic in the late 1980s and early 1990s), transferring is less about transporting than it is about transforming:

Transférer, ce n’est pas transporter, mais plutôt métamorphoser, et le terme ne se réduit en aucun cas à la question mal circonscrite et très banale des échanges culturels. C’est moins la circulation des biens culturels que leur réinterprétation qui est en jeu. […] Tout passage d’un objet culturel d’un contexte dans un autre a pour conséquence une transformation de son sens, une dynamique de resémantisation, qu’on ne peut pleinement reconnaître qu’en tenant compte des vecteurs historiques du passage[4]. (Espagne 2013: 1)

To speak of ‘sources’ and ‘targets’ (two of the most common spatial metaphors) is to presuppose that translation always already implies movement, since it ‘springs’ from a single source in one place and ‘hits home’ elsewhere, like an arrow. These well-worn images unequivocally situate translation in space. They privilege ideas of transfer and displacement at the expense of the process, power and agency transformation requires and involves.

Again: not all translation implies crossing borders (or bridges) between languages as if they were each encapsulated in their own space. As already suggested at the outset of this article, there is no one-on-one correspondence between languages and locales. One might add that translating takes time and therefore takes place in time, within a temporal framework. Translations, which is more, are not timeless but age as time passes. Copies carry the burden of the past, whereas originals tend to remain unblemished, much like the picture (and copy) of Dorian Gray in Oscar Wilde’s novel, as opposed to the character himself (Le Blanc 2019). Or, to use another simile: unlike some cheeses or wines, they do not always age particularly well. Their shorter shelf-life creates a real or perceived need for updates (so-called ‘re-translations’), a need that only seems to concern and affect certain originals, and only many centuries after they were first written.

There is indeed somewhat of a trend lately to revisit national classics by translating them ‘intra-lingually,’ in a modern version of the language of their original composition. The ‘No Fear Shakespeare[5]’ series published by SparkNotes provides students who dread the Bard and find him rather more awful than awesome with side-by-side translations into modern English of his plays. Other intralingual projects attest to a larger process of ‘refraction’, of ‘rewriting’ a text for a new audience with a view to not only attract new readers but also orient the way they read (Lefevere 1982). It took Cervantes scholar and novelist Andrés Trapiello fourteen years to complete his modern Spanish version of Don Quijote (whose publication coincided with the 400^th anniversary of the original), and philologist André Lanly fifteen years to finish turning Montaigne’s Essais into modern French. Around the same time, Guy de Pernon published Montaigne in a bilingual edition, which was also the formula favoured by Marie-Madeleine Fragonard for her Rabelais translation. In Italian, the example that attracted most attention is novelist Aldo Busi’s Decameron da un italiano all’altro (he subsequently modernized Castiglione’s Cortegiano, in conjunction with Carmen Covito). Still in Italy, Dante’s entire Commedia was more controversially subjected by Luciano Corona to a ‘riscrittura interpretativa in prosa e per tutti’ (a full-blown refraction in Lefevere’s sense, in other words). In all these projects the original author’s spatial coordinates coincide with those of their modern readers. No transfer in space is required, only a transfer in time. The reception of Dante and Boccaccio, Castiglione and Montaigne, Rabelais and Shakespeare, spans centuries, which suggests that time clearly trumps space as vector of transmission.

Finally, regarding the temporal dimension of translation, it is instructive to compare words used in different languages to describe versions that stray too much from their original. ‘Imitations’ is what they were called in Dryden’s England, at a time when the French started speaking of belles infidèles, using a notoriously sexist metaphor that suggests seductive manoeuvring more than actual spatial movement. Reference to time is perhaps most explicit in German, where the nach of Nachdichtungen means ‘after’. Italian rifacimenti stress refashioning and repetition (again, ‘afterwards’ and therefore situated in time), while Sanskrit anuvad (संस्कृत अनुवाद) combines the idea of ‘ritual repetition’ with that of ‘speaking after’ (Mukherjee 1981: 80).

The cultural limits of metaphors

Given all the above, why bother ‘breaking down an open door,’ as per the French idiom, ‘enfoncer une porte ouverte’ (which sounds violent but less so than its English equivalent, ‘flog a dead horse’)? Why not leave both doors and horses alone and, for that matter, let sleeping dogs lie? The answer is simple: Translation Studies. Because of its overreliance on English, not only as a practical means of scientific communication but increasingly as well as a metalanguage for talking and thinking about translation, TS is inheriting some of the blind spots inherent in its lingua franca. In saying this I am less alluding to the (obvious) political problems involved in the conversion of any culture-specific stance into something so-called ‘universal’ and supposedly devoid of positionality, than to issues affecting the very conceptualization of ‘translation.’

Instead of viewing conceptual metaphors through the lens of cognitive universalism, they can also be linked to cultural (and therefore relative) dimensions (Bernárdez 2013). Far from being universal, the English word ‘translation’ is an ‘exception culturelle’, as pointed out before. This tends to be glossed over in a discipline increasingly sold on the idea of working (thinking?) in one language only, at the expense of other traditions.

Thirty years ago, in an enlightening article entitled that still awaits its English version, Antoine Berman (1989: 679) suggested that translation theory is more informed by culture-specific terminology than we are ready to acknowledge. This is echoed by Tymoczko’s (2007: 76) contention that ‘theories of translation would obviously be considerably different […] if the international discipline of TS had been initiated and had gained strength in a context where the dominant language was Arabic or Chinese or Igbo.’ The dialogue championed by Berman, and the more recent broadening of horizons advocated by Susam-Sarajeva (2002) or Tymoczko (2005, 2007), must go beyond becoming informed about each other’s terminology, however. A next and necessary step is becoming aware of the idiosyncrasies and quirks of the natural languages that make up our metalanguages. Only then can we avoid becoming hostages to their metaphors.

Though it helped turning ‘translation’ into a master metaphor (even prompting wishful thinking about an imminent ‘translational turn’ in the humanities), the emphasis on spatial coordinates – that is, translation being a matter of someone ‘carrying’ something ‘across’ – has obscured other important aspects (think of ‘time’ or ‘transformation’). Moreover, by positing that the very word ‘translation’ is ‘already a spatial metaphor’ because it ‘impl[ies] transportation across a middle space’ (Guldin 2015: 49), scholars working in English – whether they be native speakers or not – make ‘pre-theoretical assumptions’ (Tymoczko 2005: 1094) that do not necessarily ‘carry over’ into other languages.

Rather than merely making such a claim without backing it up, I would like to, in the second part of this article, investigate to what extent the spatial sense of translation applies (or does not apply) to a few other languages that are within my reach – inviting scholars with different backgrounds to do the same for their respective languages[6].

The Romance languages: French, Italian, Spanish

As far as I can tell, English is alone among European languages in not possessing two different terms for translation-traductio (interpretatio in Classical Latin) and translation-translatio. Cognate languages like German (or Dutch) clearly distinguish interlinguistic Übersetzung (D. vertaling[7]) from mathematical Translation (D. verschuiving), i.e. the geometrical movement of points over the same distance and in the same direction. It certainly does not apply to the Romance languages: in French, Italian and Spanish, the latter is respectively called translation, traslazione and traslación. And in none of the five languages just mentioned is there any confusion possible between these terms, which describe spatial movement, and turning an utterance from one language into another.

Even more telling of this divorce is the fact that the verbs used to describe the latter process (i.e., traduire, tradurre and traducir) are not spontaneously felt to be linked to space by native speakers of French, Italian[8] or Spanish. More often than not, students have to be reminded of their Latin origin (trā(ns)dūcĕre, ‘to lead across’) in order to notice there once was a spatial connection. True, traces thereof survive in a few expressions, in particular in the specialized jargon of the legal profession, well-known for its archaic penchant in matters of vocabulary. Thus, in French, one can be ‘traduit en justice’ (‘taken to court’) but that phrase is opaque to all but the most educated speakers (or those with a legal background). Similarly, the authoritative online dictionary of the Italian language, Treccani[9], dedicates the very last lines of the relevant entry to a specifically bureaucratic use of ‘traduzione’ as referring to the transfer of inmates (whether to prison or from one prison to another). Another remainder (and reminder) of antiquated usage is ‘la tradotta’, the transport of military convoys by train (this detail alone says enough about how dated both word and reality are by now). As with the French example above, it is fair to say that these are not widespread but technical uses, pertaining to what Mario Wandruszka (1971; 1973: 49-50) called ‘technolects’. Passively recognized by few and actively used by fewer still, they do not, in other words, invalidate the claim that the spatial dimension has all but vanished from current usage.

As it happens, ‘tradurre’ forms part of a larger series of mots savants coined in the Middle Ages which have largely lost their spatial anchorage, have become unmoored from dūcĕre (‘to lead’). In Table 1, only the verb to the far right still actively refers to space. But even so, Fr. conduite, Sp. conducta and It. condotta can simply mean ‘behaviour,’ without reference to anything being actually ‘conducted’:

Table 1

No obvious semantic link subsists either between traducteur and conducteur (‘driver’), traductor and conductor, traduttore and conduttore (who is actually a TV host, the Italian word for ‘driver’ being autista). The very prefix tra- stopped being transparent in the Romance languages with the exception of Italian (where tra means ‘between’). This does not, however, as one might think, infuse traduttore with a particular spatial dimension: as it happens, Italian favours another form of paronomasia, associating tradurre with tradire (‘to betray’), as in the all too well-known formula, traduttore traditore.

Latin (Saint Jerome)

Yet the word traduttore was derived from Latin only in the Renaissance. True, traductor is attested in Classical Latin, but it was far from common and, more importantly for the present discussion, had nothing to do with languages. Most dictionaries quote the same passage from Cicero’s correspondence (Att., II, 9) where he calls Pompey a traductor ad plebem, which has been variously translated as a ‘conveyor’ or ‘plebeian-maker[10],’ i.e. somebody who conveys not meaning but people, promoting or demoting them from one station in life to another. Cicero was not a big fan of social mobility. This also comes across in his famous treatise on The best kind of orators (De optimo genere oratorum), where he chastised translators not only for their lack of eloquence but also for their lack of training, compared to a skilled and patrician rhetorician like himself (see McElduff 2009, Folena 1973: 61-63; Copeland 1991: 16-30). Quite logically, given what we just saw about the meaning of the word traductor, these (in his eyes) less-qualified hacks were never called traductores by Cicero. He called them instead interpretes, more often than not indiserti, ‘ineloquent’ (De finibus, III, 15).

This terminology is echoed four centuries later in the title of Jerome’s letter to Senator Pammachus, De optimo genere interpretandi, wherein the Patron Saint of (Bible) translators quotes and misreads Cicero. Alongside the Classical verb, interpretari, Jerome often resorts to transferre to get this point across. Medieval terminology would coagulate around the latter, starting from the supine translatum and leading to translatus, translatio and a new verb, translatare, which starts competing with the more common transferre in the 7th century (Buridant 1983: 96). Translatare would become a key word in the Middle Ages, before being itself displaced in the early Renaissance by traducere in most European languages, with the singular (both single and curious) exception of English.

The ecclesiastical Latin used by the Church of Rome was replete with loans and loan translations from Greek. In Jerome’s context, then, translatio was the equivalent of metaphora. It was used for the transferral of a dead person’s body (or a Saint’s relics) for appropriate burial. Many languages still use the term for that specific purpose[11], albeit with the usual phonetic or morphological inflexions: Italian traslazione, Spanish traslación or Portuguese tra(n)slação, Norwegian translasjon or Dutch translatie. Jerome uses it in this particular sense at least once in his letter to Pammachus. He quotes the following passage from the Acts of the Apostles (7: 16), ‘translati sunt in Sychem et positi sunt in sepulcro quod emit Abraham,’ which reads in the King James Version: ‘And were carried over into Sychem, and laid in the sepulchre that Abraham bought[12].’ Modern translations hesitate between ‘carried back,’ ‘taken back’ or ‘brought back’ but none use ‘translated’. A simple and logical explanation would be that in the minds of English readers, the term ‘translation’ presumably conjures up a change in or substitution of language(s) much more readily than a change in setting.

English (Hemingway)

Even in English, then, it seems that the association between translation and movement could be far from ‘natural.’ Instead of taking my word for it, let us look at Ernest Hemingway, renowned for his keen ear for spoken language and natural speech. In his first novel, The Sun Also Rises (originally published in Britain as Fiesta), he has his characters travel to Spain in order to attend Pamplona’s famous bull-races, the San Fermines (in which Hemingway actually participated). San Fermin, the narrator informs us, ‘is also a religious festival.’ Part of the ceremony involves actual translatio: ‘That afternoon was the big religious procession. San Fermin was translated from one church to another. In the procession were all the dignitaries, civil and religious.’ (Hemingway 1926: 156 and 158-9) Hemingway uses the verb in its obsolete sense of transferring religious relics. He does so in order to create a stylistic effect a few pages (and many drinks) later, when he has one imbibed character say about another: ‘Come on, […] Let’s translate Brett to the hotel.’ (163) For this sentence to work, and for the effect of estrangement (ostranenie) to obtain, two conditions need to be met. First, the original spatial meaning of the verb, ‘to translate,’ must have become obsolete or at least obscure, overshadowed by its modern meaning. This, in turn, allows Hemingway to resurrect the old meaning in his description of the religious procession, thereby re-injecting etymology into current usage. Were it not for this two-step semantic operation, the sentence about Brett being ‘translated to the hotel’ would sound more than a little awkward (even in a novel that indulges in back-and-forth bantering between explicit English and suggested Spanish). The drunken debauchery of the novel’s tourist characters, furthermore, stands in stark contrast with the solemnity of the Spanish procession, thus turning the entire scene into a carnival of Bakhtinian proportions.

Predicated as it is on the polysemy of the Latin translatio, which only English seems to have inherited, Hemingway’s wit is bound to be lost in translation. In a recent Spanish version, M. Solà (whom I have not been able to further identify) keeps the religious vocabulary, a logical choice in that it maintains the curious distance between the characters’ antics and the festival’s ceremonies: ‘Por la tarde tuvo lugar la gran procesión. Se trasladaba a San Fermín de una iglesia a otra.’ (Hemingway 1979/1985: 81); ‘Vamos – dijo Bill – traslademos a Brett al hotel’ (84, ‘let’s transfer’). By contrast, the novel’s French translator, Maurice-Edgar Coindreau, did away with Church terminology altogether and opted for the much more general ‘transporter’: ‘La grande procession religieuse eut lieu dans l’après-midi. San Fermin fut transporté d’une église à l’autre.’ (Hemingway 1933: 175); ‘Allons, dit Bill, transportons Brett à l’hôtel.’ (180). In this version, Bill’s comment still echoes the narrator’s description but there no longer is anything strange or funny about it, as Coindreau either fails to or decides not to pick up the pun about translatio.

German (Heidegger)

German comes closest to English in highlighting the spatial dimension of ‘transfer’ in the translation process. Übersetzen (‘to ferry across’), less often übertragen (‘to carry across’) and their corresponding nouns, Übersetzung and Übertragung[13], were introduced as literal translations of Renaissance traductio. The older Latin word, meanwhile, became more specialized and was kept for either religious Reliquientranslation or, more commonly, mathematical Translation.

Yet even in German, there is more than meets the eye. When Martin Luther defended his version of the Septuagint in 1530, his letter on the matter was entitled ‘Sendbrief vom Dolmetschen’ – a word whose meaning, like Latin interpretari, has narrowed considerably with the 20^th-century advent of conference interpreting. As far as I can tell, Luther’s letter does not contain a single occurrence of übersetzen. Perhaps the word had yet to be coined? We know from the etymological dictionary prepared by Wolfgang Pfeifer (1993) that dolmetschen remained current for quite some time afterwards and was only permanently displaced by übersetzen at the outset of the 18^th century.

Both übersetzen and übertragen are linked to space in much more obvious ways than any of their Romance equivalents. This spatial dimension was teased out by a philosopher who was fond of wordplay and loved pushing the acoustic envelope of words. In his 1955-56 lectures on ‘The Principle of Reason’ (Der Satz vom Grund), Martin Heidegger claimed that great translations ‘transpose (übertragen) a work of thinking or poetry […]. In such cases, translation is not only interpretation (Auslegung) but also tradition (Überlieferung). As tradition, it belongs to the innermost movement of history.’ (tr. by Heyvaert in Berman 1992: 177) Earlier on, in his 1943 lectures on Heraclitus, Heidegger had separated (severed) the prefix über (‘across’) from setzen (‘put’) to stress his point:

Hier wird das Übersetzen zu einem Über setzen an das andere Ufer, das kaum bekannt ist und jenseits eines breiten Stromes liegt. Da gibt es leicht eine Irrfahrt und zumeist endet sie mit einem Schiffbruch. (Heidegger 1979/1994: 45)

Here translating (Übersetzen) becomes a transporting (Über-setzen) to another shore, which is barely known and lies on the other side of a broad river. This easily leads to roaming about, and most of the time ends in shipwreck. (tr. Groth 2017: 137)

The German language is like a building of LEGO bricks: rather than inflecting word-endings, it strings morphemes together, a process known in linguistics as ‘agglutination,’ which leaves the individual units visible and facilitates the deduction of meaning. Heidegger makes clever use of this morphological feature – which exists in many other languages: German’s close cousin Dutch, for instance, but also completely unrelated languages like Turkish or Basque. Unfortunately, it allows him to suggest a few misguided things.

As the quotes above suggest, Heidegger saw formal similarity as a sign of semantic similarity. Similarity does not, however, constitute identity. In fact, the ‘wordplay of übersetzen in German does not […] come about through a homophonic relation […] but through homography’ (Dizdar 2009: 92). While the two infinitives may look the same in German, they surely do not sound the same. The verb which literally means ‘leading across’ is stressed on the first syllable (übersetzen), whereas the accent is on the second syllable in the verb with the metaphorical meaning (übersétzen). Actually, even in terms of writing, there is no complete overlap. In German, phrasal verbs with a concrete meaning can be separated, whereas verbs used in the figurative sense cannot: compare Der Fährmann setzt uns über (‘the ferryman transports us’) with Ich übersetze den Text (‘I translated the text’). This also explains why the two verbs have different past participles (übergesetzt vs. übersetzt, the difference residing in the presence vs. absence of a grammatical morpheme). The German equivalent of the sentence, ‘the ferryman ferried us to the other shore’, thus becomes ‘der Fährmann hat uns auf das andere Ufer übergesetzt,’ whereas ‘he translated that sentence word for word’ would be ‘Er hat diesen Satzt wortwórtlich übersetzt’ (examples provided by the Swiss-based Catalan scholar Germà Colón [2001: 168]).

To sum things up: space is not irrevocably inscribed in the German term for translation either, even though it arguably comes a lot closer than the Romance languages. Heidegger’s hyphenation, then, is less a way of actualizing the word’s potential (to put it in Aristotelian terms) than simply manipulating it. In the eyes of Turkish-German translation scholar, Dilek Dizdar (2009: 92), Heidegger’s über-setzen (with a hyphen) becomes itself sort of a ‘translation of übersetzen’, albeit one that ‘no longer contains the wordplay characteristic of this word’ and ends up being ‘counterproductive for the conception of the theory and practice of translation’. Since it is unconceivable that Heidegger would not have been aware of the aforementioned grammatical features of his native language, he more likely chose to ignore them as they did not serve his purpose.

Ultimately, however, wordplay of this sort amounts to little more than ersatz thinking. Again, do not take my word for it. The diagnosis is that of writer and translator Georges-Arthur Goldschmidt, who fled Nazi Germany as a Jewish child and became French. In his essay, À l’insu de Babel, Goldschmidt highlights the morphological flexibility of the German language, where many verbs and nouns can be combined with at least twenty prefixes or suffixes, resulting in what he sees as a simulacrum of thought:

Rien n’est plus aisé à fabriquer qu’un pastiche du langage philosophique [allemand]. Il suffit de combiner les éléments entre eux pour paraître obtenir de la pensée. […] Ce qui semble ainsi pensée et ‘profond’ n’est rien d’autre que du hasard combinatoire. (Goldschmidt 2017: 122 and 126)

In addition to using and abusing paronomasia, Heidegger was one in a long line of theologians and philosophers who believe that etymology casts such a long shadow over words that their meaning is enshrined once and forever. Nothing could be further from reality. Etymology does not turn words into relics, does not stabilize meaning. Words evolve semantically as well as phonetically. They can even end up meaning the opposite of what they originally meant, hence the very real possibility of etymology becoming a fallacy. Take the word ‘school,’ σχολή (scholē) in classical Greek, where it stood for ‘leisure’ (as in ‘intermission of work’) before undergoing a 180^o turn. First extending its meaning to ‘that in which leisure is employed’, it became shorthand for ‘leisure for learning; learned conversation, debate; lecture; meeting place for teachers and students,’ and eventually to our current ‘place of instruction[14].’ Today, the idea of ‘leisure’ has all but vanished (ask any school-aged…) Etymology, then, is neither essential nor eternal but rather a contingent source of meaning.

In short

Our detour via Heidegger nevertheless proves useful in showing the importance of etymological reductionism in what I have labelled the kinetic fallacy underlying much current thinking about translation as transportation. For centuries, from late Antiquity through the Middle Ages, translatio meant transfer in space: of bodies and empires (or their remains) first and foremost, later of texts. The Renaissance would transform this terminology by adding traductio to the mix. Though ostensibly related to translatio on the level of morphology (trans-ferre and trans-ducere), the more modern term signalled a shift on the semantic level in that it became increasingly unanchored, disconnected from the initially implied spatial dimension.

This shift was recorded in most European languages, as we have seen. Except in English. Not only was the old word translatio (entered into English via Middle French translater) maintained, it also kept intact its three semantic fields (religion, language, mathematics), thereby encouraging an overlap between entirely different realities, some of which are inherently spatial while others are not always and not necessarily. As a result, complex polysemy risks being (and has in fact often been) reduced to simple synonymy: translation is translatio, linguistic exchange is movement. In no other language that I am aware of is this the case. Not even German, where the two very close variants of the verb übersetzen are not quite identical but are almost ‘false friends’ (des faux amis). Nor do they extend their meaning to mathematical applications or religious translatio. What may seem like a logical step to a mind trained chiefly or exclusively in English, therefore, turns out to be a step too far. Taking a step back, it is plain to see (especially when looking through a multilingual lens) that the transportation metaphor is mostly a suggestive image: nothing less, but nothing more.

About the author(s)

Rainier Grutman is a Professor of French and Translation Studies at the University of Ottawa (Canada). Initially trained in Romance Philology (Namur, Leuven, Complutense Madrid), he received his Ph.D. from the Université de Montréal. In addition to studying bilingual writers and heterolingual writing (Des langues qui résonnent, Montreal 1997, 2nd ed. Paris, 2019), he has focussed on literary translation and, in particular, self-translation. He is co-editor with Alessandra Ferraro of L’Autotraduction littéraire: perspectives théoriques (Paris, 2016), he has contributed articles on the topic in many journals as well as in works of reference such as the Routledge Encyclopedia of Translation Studies (1998, 2009, 2019), the IATIS-Yearbook (2013), and the Wiley-Blackwell Companion to Translation Studies (2014).

Email: [please login or register to view author's email address]

•  Previous article
• Next article