Where’s cultural adaptation
A corpus-based study on translation strategies
By Anna Mauranen (University of Tampere, Finland)
Abstract & Keywords
Il perseguimento della naturalezza nella resa è generalmente ritenuto dagli studiosi di traduttologia una norma dominante, oltre ad essere un obiettivo di primaria importanza per la didattica della traduzione. Sorprende quindi la carenza di studi su larga scala che dimostrino l’effettiva predominanza di strategie di adattamento al contesto di arrivo rispetto a strategie di trasferimento di elementi tipici del testo di partenza. Corpora comparabili contenenti testi originali e testi tradotti nella stessa lingua offrono gli strumenti necessari per perseguire uno studio di questo tipo. Il presente lavoro propone un’analisi di testi di tipo divulgativo (storia, scienze, guide al fai da te), originali e tradotti in finlandese. Tale analisi sembra suggerire che il ricorso a strategie atte ad “addomesticare” il testo non è frequente come si potrebbe pensare. Le principali differenze rispetto agli standard del contesto di arrivo appaiono collegate alla diversa frequenza d’uso di espressioni culturalmente connotate in campo pragmatico.
Although there is a general tendency in translation studies and translator education to accept target-language naturalness as a dominant norm, not much large-scale evidence has been available to show whether adaptive translation strategies are more common than those transferring source text properties into the target text. Comparable corpora of translated and untranslated texts provide a method of coming to grips with the reality of translation in this respect. This paper investigates popular non-fiction texts of translated and original Finnish, arguing that translators do not use “domesticating” translation strategies as much as might be expected. The main deviations from target-like usage appear as untypical frequencies in culture-specific pragmatics.
Keywords: corpus-based translation studies, translation strategies, comparable corpora, untypical frequencies, corpus linguistics
©inTRAlinea & Anna Mauranen (2002).
"Where’s cultural adaptation"
inTRAlinea Special Issue: CULT2K
Edited by: Silvia Bernardini & Federico Zanettin
This article can be freely reproduced under Creative Commons License.
Permanent URL: http://www.intralinea.org/specials/article/1677
The leading ideology in translation studies and translation practices has long emphasised target-language appropriateness (or “acceptability”, “naturalness”, etc.) as a major criterion of a good translation. It seems to be generally accepted that, in whatever ways we may choose to see translation, it should at least function appropriately as a text of the target culture. This orientation towards the target requires both linguistic and cultural adaptation, or a “domesticating” translation strategy in Venuti’s (1995) terms. Although large cultures may have their own interests in favouring this strategy, it is common in small cultures as well. A case in point is Finland, where the domesticating translation ideal has been resorted to as a means of protecting and developing the national language and identity against more dominant cultures from which translations originate. However, the general concern for the target text has not resulted in identical recommendations for translators and students even when very similar issues have been investigated. So for instance Karhu (1997) observed that Finnish translations from English used more postmodification in nominal phrases than originally Finnish texts. She recommended that translation students be taught to change ST postmodification more often into premodification in the interests of TL naturalness. The opposite conclusion was reached by Vehmas-Lehto (1989), who felt that the frequent premodification of nominal phrases she found in Russian-Finnish translations reduced readability and should therefore be warned against.
Obviously, disputes over translation norms cannot be settled on the basis of descriptive studies alone; the conflicting recommendations above rested on a normative approach to translation and different weighting of shared norms. Yet it is hard to see how normative disputes could be settled without descriptive accounts: if translators on average resort to postmodification more often than the TL norm, it is nonsensical to chastise them for using premodification. Neither of the above studies can go very far on the strength of their empirical support, since both were based on a relatively small amount of data. The same tends to be true of other related studies, be they concerned with the translation of cultural “realia” (e.g. Klingberg 1986, Kujamäki 1998), or linguistically untypical features of translated language (e.g. Puurtinen 1995). It is only fairly recently that corpus linguistics has enabled explorations into the linguistic side of the target-language ideal on a larger scale. This is perhaps best achieved by using comparable corpora of translations and original texts in the same language, as Gellerstam (1986, 1996) has done for Swedish. This approach was suggested as a serious alternative in translation studies a few years back by Mona Baker (e.g. 1993, 1996, 1999), followed up first by Sara Laviosa (1996).
On similar lines, a comparable corpus of translated and original Finnish was compiled at the Savonlinna School of Translation Studies (Mauranen 1998). A couple of recent studies based on this corpus have raised the question of what happens to culture-specific language in translation. Sonja Tirkkonen-Condit (2000) suggested a potentially universal translation tendency: “untranslatable” items of the TL get underrepresented in translations. Her results on certain lexical items in the “fiction” subcorpus of the database supported the idea. My findings (Mauranen 2000) on the relative underrepresentation of discourse markers with highly TL-specific pragmatics can also be taken to support the general idea, but this was found with simultaneous overrepresentation of SL-specific use of discourse markers. My findings were made in a high-prestige genre, academic texts, but the question of whether this might be different in popular genres was left open.
Yet the question of translation in popular genres is interesting; popular texts are read much more widely than academic texts, and thereby can be expected to exert more influence on the language as a whole. Nevertheless, they have not received as much attention as more prestigious genres within translation studies, or in corpus linguistics. A more theoretical interest in the behaviour of popular texts is provided by Douglas Robinson’s suggestion (1997) that translations in a dominated culture tend to render texts originating in a dominating, or hegemonic, culture fluent and accessible when they are intended “for the masses” (as opposed to texts for the elite). The hypothesis seems, at least to me, counter-intuitive, because one would assume that if the dominating culture really is hegemonic, then its discourses are familiar to members of a dominated culture independently of their social or educational standing, but the issue is best dealt with on an empirical basis.
My purpose in this paper is to look at popular prose texts translated into Finnish and compare them to similar texts written originally in Finnish, focusing on one of their most striking differences, which appear in the domain of personal reference. The results cast doubt on Robinson’s hypothesis. They also throw light on the question of overrepresentation of SL features vis-à-vis underrepresentation of TL features. While I do not intend to give recommendations to translators or translator education, the results reveal aspects of translations which differ systematically from the target language norm. Awareness of the differences is the first step towards deciding whether to encourage or discourage current translation practices.
The data I have used consists of popular non-fiction texts, including popular history, “self-help” guidebooks, and popular science. The texts constitute part of the Corpus of Translated Finnish, compiled in my project at Savonlinna (Mauranen 1998). The whole corpus, comprising about 10 million words, includes several genres and source languages, and consists of entire books published 1995-2000. The selection criteria have been external, based chiefly on publishers” descriptions. The present data constitutes a subsection defined by genre, and as it happens, only translations from English, since no translations of popular non-fiction were published from other languages during the compilation period. The subcorpora are both about half a million words in size.
3. General comparison
I set out to compare the corpora in the simplest possible way, by comparing the original and translated corpora in terms of word frequencies. I ran a comparison with the Keykey Words in Wordsmith Tools in both directions, i.e. by taking original Finnish and translated Finnish alternately as the reference lists, and comparing these to the batch lists of each individual book file (again translations vs. originals).
The most striking outcome of the comparison was the predominance of personal forms, particularly 2nd person singular, both as pronouns and as verb forms. Tables 1 and 2 give a summary.
Table 1. Differentiation in favour of translations (top 50)
Out of the 50 best differentiating word forms in favour of original Finnish texts, none were pronouns. Seven were active verbs, all in the third person and two in the (impersonal) passive.
Table 2. Differentiation in favour of original Finnish (top 50)
In all, it was the second person singular forms that populated the top end of the translations, which altogether had ten times as many instances of the lemma SINÄ (“you” SG) as originals:
SINÄ (“you”) lemma /100,000 words
Finnish originals 60
Compared to frequencies in academic writing, it is easy to see that the lemma SINÄ is much more typical of popular than academic prose. The difference between originals and translations drops from being ten times more in popular translations to about five times more in academic translations. Yet the difference remains consistent.
SINÄ (“you”) lemma /100,000 words
Finnish originals 26
The first person singular pronoun did not differentiate equally well between the popular corpora, and will not be discussed further.
4. Self-help books
One might assume that references to “you the reader” and “I the writer” would be most common in the subgenre of self-help books rather than, say, popular science, history or pet care. The self-help book is concerned with giving advice about personal problems or lifestyle. Running frequencies for the lemma SINÄ for each book individually confirmed the hunch: the top four books for SINÄ frequencies were of the self-help type in both corpora.
In English, it is fairly well-established that personal pronouns are on the whole used more in conversation than in writing (e.g. the Longman Grammar 1999: 333), and conversation is particularly characterised by exophoric reference to the interactive participants. The frequent use of “you” in popular written genres, especially in the self-help genre, could thus be attributed to an apparent intention of establishing a close and confidential relationship between the writer and reader. However, the observations about English are not directly applicable to Finnish. First, although pronoun use appears to be more frequent in conversation, the pronominal forms in speech and writing differ noticeably. The SINÄ forms are written or formal, and although they have been conventionally accepted in fictional use, they are employed less and less, as fiction increasingly tends to prefer the actual spoken forms in dialogue. The tone conveyed by the written form is therefore not so much friendly and conversational but didactic or patronising. Secondly, not only do Finnish verb forms encode person, thus making pronouns more redundant, but also the “impersonal” and passive forms of the verbs can in conversation be used for personal reference, either definite but indirect, or indefinite (Luukka 1994). There is not much research on this, and none so far that would be contrastive or based on corpus evidence, but the relevant point is that not only the grammar but the pragmatics of personal reference differ in English and Finnish.
Because this genre is a relatively recent import from the US, one would assume that it also has borrowed discourse conventions of the genre from its source culture. This would explain the relatively high frequency of SINÄ in the original Finnish self-help books compared to other Finnish texts. It can also be seen by comparing SINÄ in the self-help books: it turns out that the difference between originals and translations is smaller here than in popular non-fiction on the whole, although still very clear (about eight times more instances in the translations).
SINÄ (“you”) lemma /100,000 words
Finnish originals 173
The distribution of the different forms of the lemma showed no very striking differences, apart from the subject form (nominative case) of the pronoun, which was the second most frequent in the originals, but ranked only fifth in the translations. The SG2 pronoun does therefore not appear to be used particularly often in subject position compared to its other uses in the translations.
5. Focusing on a sample
Overall frequencies apart, can we see differences in the actual use of SINÄ in the texts? To explore this, I ran concordances on SINÄ with two samples from the data: (1) for a random sample of 100 and (2) for the most frequent individual forms in both corpora. Some of the most obvious things that differ in the samples of a hundred appear in Table 3.
Table 3. SINÄ lemma Random sample of 100 from both corpora
First, a third of the instances appear in reported speech in the originals, but very few in the translations. That is, in Finnish, SINÄ is less often the writer directly addressing the reader, but an imaginary character in an example or dramatised passage; the interaction is thus removed from the ongoing discourse. In the translations, again, the pronominal reference is almost exclusively exophoric in the reader-writer interaction. Second, the number of deontic modals was quite high in both. Considering the genre, this might be expected (after all the point is to give advice for coping with one’s life), but two interesting things emerge here: the actual expressions of obligation were different, and moreover, the negated obligation was much more common in the translations. These observations are supported by frequent collocates, and point to different discoursal orientations: on the one hand telling readers what they ought to do (Finns), and on the other emphasising what they do not have to do (Americans). An additional difference is that the Finnish originals seem to encourage the reader to do a lot of thinking, but not so the translations.
6. The most common forms
For a close-up on differences in use, the most frequent forms of the lemma provide a reasonable route. Sinun (genitive) was the commonest on both lists. Since the quantitative discrepancy between its occurrences in the corpora was considerable, I took all the 56 cases in the originals, and compared the figures in percentages to those from a random sample of 100 in the translations.
Table 4. Summary of sinun in the originals (all in %) and translations (random sample of 100)
Again, the proportional difference in reported speech shows itself, and so does the one in deontic expressions. A couple of observations are worth making. One is to do with the sense where the obligation is not imposed on but rather removed from the addressee. This was a large category in the translations. The top 3-word cluster in the translations was sinun ei tarvitse (“you don’t have to/ you needn’t”). The same combination of words with some permutations recurred in the top clusters. There was an overall difference in tarvitse as well, which was twice as large as in originals.
Checking its use in personal and impersonal contexts again shows that translations observe a different norm:
Impersonal originals 54.4%
Impersonal translations 17.2%
The closest a corpus approach can get to the translation process is probably through a parallel corpus, i.e. original texts and their translations. To see what happens with (sinun ei) tarvitse I consulted the FECCS corpus, the Finnish version of the Nordic parallel corpora (see, e.g. Aijmer et al. 1996). I took into account all cases where the expression appeared in either the original or the translation. In this 2-million-word corpus the 18 cases fell easily into four types.
Apart from the cases which clearly involved the generic “you” in English and which were quite predictably translated impersonally, as in (1), most of the translations used the pronoun form.
(1)“Ei siinä tarvitse olla pankkimies”, selitin kärsivällisesti.
“You don’t have to be a banker to get into it,” I said patiently.
Lit: “there’s no need to be a banker’
Translations with the pronoun fell into two main groups. The larger was rather patronising in tone like (2).
(2) Sinun ei tarvitse olla peloissasi, tyttöseni, Farthing on valppaana.
“No need to fright yourself, my dear. Farthing’s on watch.”
Lit: “you needn’t be afraid’
The second main group involved the emphatic, typically contrastive, use of “you” (3)
(3) Ei kyllä äiti on aivan kunnossa, mutta kerron sen hänelle itse, ei sinun tarvitse - hei vain”, William sanoi päättäväisesti ja paiskasi luurin alas juuri kun Barbaran käsi oli tarttumaisillaan siihen.
No, no, she “s fine but I’ll tell her myself, no need for you to - bye, darling,” William said firmly, banging down the receiver just as Barbara’s fingers closed about it.
Lit: “you don’t have to’
In addition, four cases remained where the “you” was not generic but was nevertheless translated either with an impersonal form or with a personal form of the verb without the pronoun (4).
(4) Ei tarvitse sanoa.
You don’t have to say it.
Lit: “no need to say it’
With so few instances there is no point in speculating over quantities, but some domesticating seems to be going on (personal into impersonal forms, where an explicit personal is not motivated). The same four types appear in the comparable corpora, but there the quantities reveal that domesticating solutions remain a minority.
In itself, the expression (sinun ei tarvitse) is perfectly idiomatic Finnish, to the extent that it is hard to think of natural near-synonymous expressions. For translation practice, this implies that if the referential meaning of the text is to be rendered more or less intact, then the bias is likely to remain in the text. Although some of the expressions might be substituted for impersonal ones, many of them could not - if this content is what is to be conveyed. In other words, this is a cultural difference which pertains to the propositional content and the pragmatics of what is said, rather than the linguistic expression itself. If we wish to hold on to the maximal domestication ideal, not only the expression but the content of the original needs to be changed.
The second observation on the deontic expressions concern those imposing an obligation on the reader rather than lifting one. In this case, the originals and translations differ in their preferred verb choices. The translations used the form with the strongest possible obligation (sinun on + V; “you are to”), while the originals used milder alternatives sinun pitää and sinun täytyy (“you must”, “you should”). The one preferred in the translations was odd in that it is also a much more formal alternative, while pitää and täytyy have a relatively colloquial everyday ring. This is, then a register problem again in the translations. The connection with the emphatic obligation in the translations was so strong that it also emerged from the collocations of OLLA (“be”), where the commonest collocate for on was sinun. Instead of going through the quantitative details here, I illustrate the above with concordance examples where even a reader unfamiliar with Finnish may simply see that the collocates following sinun are different.
sinun ( + deontic) originals
uksensa ja tekonsa. Se merkitsee vain sitä, että sinun pitää hyväksyä se tosiasia, että toisen
rheellinen ja sinulla on sen ikäisiä lapsia, että sinun pitää illalla huolehtia heidän
kkia kivet. Mutta kun aloitat rakentamisen, sinun pitää ottaa vain yksi kivi ja asettaa se huolellisesti
aa se huolellisesti paikalleen. Seuraavaksi sinun pitää ottaa taas yksi kivi ja asettaa se huolellisesti
päivä jatkuvasti tekemisissä sen kanssa. Sinun pitää vain ottaa toinen asenne ja huomaat, että ystävänpäivänä. Jos haluat ihmeitä syntyvän, sinun täytyy toimia itse ja reagoida joka hetkeen.
arialainen” Jos haluat lähellesi hyvän ihmisen, sinun täytyy myös itse muuttua sellaiseksi.
, jos on kysymys vastakkaisesta sukupuolesta, sinun on hyvä jossain vaiheessa ilmaista, oletko
[b[sinun ( + deontic) translations
lanteen muuttamiseksi sinun on sanottava seis! Sinun on ymmärrettävä, ettet tarvitse enää mitään
en mukaan. Joudut siis tekemään valintoja, ja sinun on tehtävä nuo valinnat ennen paaston
. Sellainen yritys tekisi vain egon todelliseksi. Sinun on muistettava, ettei egosi ole todellinen.
jos tarkkailet mieltäsi, mitä silloin käytät? Nyt sinun on luovuttava vanhoista uskomuksistasi ja
itä varmasti tulee, pidä niitä merkkinä siitä että sinun on terästettävä keskittymistäsi. Voit
silloin mahdotonta. Jotta pääsisit ulos ansasta sinun on oivallettava, ettei maallinen
s on “ristiriitoja” - joko henkisiä tai fyysisiä - sinun on syytä tutkia, miten ja miksi sallit
oleensavetävämmän. Kun etsit sielunystävää, sinun pitää ehdottomasti käydä paikoissa,
In these examples, the main problem is clearly the unusual choice of verbs from the perspective of natural Finnish. The remedy could simply be pointing out the unusual frequency of this emphatic form - changing verbs to those found in the originals would immediately soften the commanding tone. This is an example of a corpus approach showing its strength, since individual instances cannot easily be shown to be a problem - it is the repetition that brings out both the differences and the oddity of the collocational choice.
After looking at sinun in some detail, I just very briefly take up the second most frequent differentiator, sinulle (allative case, roughly “to you” or “for you”). Among the most frequent collocates in the translations were two lexical verbs virtually nonexistent as collocates of sinulle in the originals: antaa (“give”) (2 in the originals) and opettaa (“teach”) (none in the originals).
Antaa (“give”) collocates frequently with personal pronoun indirect objects in the translations, while in the originally Finnish texts the typical complements are direct objects. So, roughly, in original Finnish, you foreground the direct object; the beneficiary (often inanimate or abstract) can then be either implied or expressed in less directly personal terms. On this score, the translations could probably again be brought closer to the target norm by simply pointing out that the making the personal object explicit here is redundant, and untypical Finnish.
There is not much to say about the other frequent translational collocate opettaa sinulle (“teach you”), since there were no examples of it in the originals, but it seems to continue the strain of didacticism noted earlier. Out of curiosity I wanted to see how the converse item OPPIA “learn” compared to “teach”. It was equally frequent in both corpora (52/100,000), and it occurred with a modal auxiliary in both, mostly VOIDA (“can”). Those were the similarities. As to the differences, the verb did not take the 2nd person singular form in the originals at all, and the reader as the intended learning subject was otherwise made explicit in only 3% of the instances. In contrast, nearly 70% (68.4%) of the translations made the 2nd person learner-reader explicit. The originals also usually had some impersonal epistemic modal preceding the infinitive oppia, (i.e. an expression of the type “it’s possible to learn”) but in the translations much less so, apart from a few instances of voit (“you can”), again a 2nd person singular form.
opitaan hoitamaan ja niistä saatetaan joskus oppia hyvinkin paljon. Huipputason urheilujoukkueen
ta voisi vapautua? Voisiko pelkojensa kanssa oppia tulemaan toimeen? On pelkoja joista voi
ullista on, että opitusta tavasta on mahdollista oppia pois. Pakoreaktiosta ti pääseminen edellyttää
ä jo osaa. Epäonnistuminen on mahdollisuus oppia uutta! Epäonnistumisesta voidaan puhua
sia psyykkisiä taitoja. Kilpailtaessa järkevästi voi oppia vertailemaan ja näkemään omassa
tulevaisuutta ja elämää varten. Ennen kaikkea voi oppia voittamaan itseään, joka on taidoista
ristöään. Toisaalta aidosti ja ehdoitta itseään voi oppia rakastamaan vain suhteessa toisiin ihmisiin.
mmärretään, että erehtyminen on inhimillistä, voidaan oppia rakentavampia tapoja selviytyä
ja vanhojen tapojesi vapauttamista. Toivon, että opit tekemään äänelläsi kolme aivan uutta asiaa. 1.
ään muuta saisikaan tästä ohjelmasta, toivon että opit tämän taidon. Istu kokeeksi merenrannalla ja
aat, että pystyt olemaan rauhallinen, mutta lisäksi opit arvostamaan elämää entistä enemmän.
äset menestyksen makuun ja harjoittelet sitkeästi, opit pian elämään myrskyn silmässä. Suhtaudu
tuomita itseäsi ihmissuhteittesi laadun takia. Voit oppia niistä kaikista. Tarkkailijan asemassa tiedät,
etella tätä asiaa. Jos puoliso on jättänyt sinut, voit oppia itsenäisyyttä. Tällainen tilanne auttaa sinua
oskaan ennen, mikä sinulle sopii ja mikä ei. Voit oppia itsestäsi muutamien paastonjälkeisten
What does this imply for translation? If the ST writer chooses to address the reader explicitly, to what extent can the translator manipulate the text to make it fit the ideal of maximally domesticated text, or maximally natural TT? In many of the cases with OPPIA here the “you” appears to be used in a fairly generic manner, and the Finnish impersonal form would then be a natural substitute. In others, however, it seems that the text would have to be changed radically to exclude the recurrent direct addressing of the reader. In brief, the issue is one of discourse pragmatics, with some of the changes required for faithful domestication being more socio-cultural than merely matters of linguistic idiom.
7. Target language specifics
As a final example, I want to take up one item which differentiates in favour of original Finnish, not translations, to follow up on the idea introduced at the beginning that TL-specific features may get underrepresented in translations. Among the top differentiators favouring originals were two impersonally used lexical verbs, kannattaa and pyritään. I focus on kannattaa (” is worth it” “is a good idea”), since it is the more frequent and relates to the didactic tone already discussed. There were more than twice as many cases in Finnish original self-help books as in the translations.
kannattaa in self-help books (n/100,000)
Kannattaa is a mild way of suggesting something, and it is typically used impersonally, since it takes no grammatical subject or personal forms (even though there are other means than the traditional subject position in attaching person to the verb). In the original data, the personal expressions associated with kannattaa (as beneficiaries) were all category terms, like “a sportsman”, “elderly women”, “those worried about food”, or “everyone”. In the translations, however, sinun (personal “you”) was again a common beneficiary. The other distinguishing feature in the translations was collocation with expressions denoting “remembering” or “keeping in mind”, which did not appear even once in the originals. The concordances below illustrate the different combinatory tendencies in originals and translations.
ämässä ihan tarpeeksi joka tapauksessa, joten kannattaa miettiä hyvin tarkoin, mitä vaikeuksia
an, alamme tiedostamattamme toteuttaa sitä. Kannattaa miettiä myös miten paljon aikaa tarvitaan
kuitenkin käyntiin. Ongelmia ratkaistaessa kannattaa antaa arvoa pienellekin kehitys- tai
oittaa myötätuntoa lapsen pahalle mielelle. Kannattaa antaa hänen surra surunsa ja lohduttaa
llisesti erilaiseksi. Kadonnutta avainnippua kannattaa ensin etsiä järkevästi, mutta jos se ei löydy,
a, että mieluummin kuin syiden pohdintaa, kannattaisi etsiä ratkaisuja tilanteisiin. “Kuinka tähän
n varalle tai useamman päivän turnauksiin kannattaa harkita mielekästä tekemistä ja ajankulua
an tulevaisuuteen. Rentoutua voit juuri nyt. Kannattaa muistaa, että rentoutunut ihminen voi olla
i ette odota niistä vakituista ja jakamatonta. Kannattaa muistaa, että vasta parisuhteensa päättäneet
ä antaa anteeksi, mutta yllä mainittu painotus kannattaa pitää mielessä, sillä siitä on paljon apua.
euraava A Course in Miracles -teoksen kohta kannattaa pitää mielessä: “Tämä on mieletön
ostasi paastoamalla. Kaikkein viimeiseksi sinun siis kannattaa nyt panna sinne lisää haitallisia
ä (joiden päässä on kumi). Tämä harjoitus sinun kannattaisi tehdä istuen oman pyhän paikkasi
e on realistinen elämänasenne. Sen sijaan sinun kannattaa epäillä omaa arvostelukykyäsi. Muistuta
The negated form (ei kannata) follows exactly the same lines: twice as many instances in the originals, none involving “you”, while 40% of the cases in the translations use “you” (roughly indicating “it’s not a good idea for you to do X”). Once more, sinun is linguistically redundant in the case of personal beneficiaries, and could perhaps be weeded out of the text with a suitable linguistic cleansing operation. However, where it collocates with “reminding” verbs, it is a matter of socio-cultural discourse pragmatics, i.e. you do not say such things in Finnish, not very often at least. It is not customary in Finnish writing, even self-help literature, to keep telling readers to remember something. At the same time, it does seem quite acceptable to tell readers to think - we may only guess at the cultural undercurrents reflected here.
This last example also illustrated how overrepresentation of TL features occurs simultaneously with underrepresentation of SL phenomena. The item kannattaa itself is a TL-specific expression, belonging to a Finnish verb category which is limited the 3rd person singular form, and which (in most currently prevalent analyses) does not take a subject. That it occurred more than twice as often in the originals provides support to the hypothesis that translations tend to have proportionally fewer cases of features specific to the target language.
This paper has explored comparable corpora of popular Finnish texts, self-help books in particular, observing consistent differences between originals and translations. The relative excess of personal pronouns in translations looks like a straightforward breach of a TL linguistic norm, which translators and translation students could perhaps easily be taught to avoid. But the greater number of 1st and 2nd person forms of verbs is a different matter - it conforms to local linguistic norms of the TL, but still deviates clearly from TL usage. Together these two phenomena contribute to foregrounding the writer-reader relationship, which amounts to culturally untypical target language pragmatics. Reader-writer relations are not of course missing from Finnish text, but they tend to remain implicit and be backgrounded most of the time.
The dissimilarities reflect preferences in linguistic usage which would be difficult to detect without the help of corpus methods, since it is their recurrence patterns that reveal the differences, not so much individual expressions in themselves. This combination seems to support a view of translations as “hybrid texts” or “third code” (e.g. Frawley 1984)They also reflect separate cultural practices, as translations appear to retain source cultural pragmatic features distinguishable from those of the target culture, while at the same time exhibiting fewer target-specific features. The difference in terms of simple pronoun frequencies was also more pronounced in the domain of popular texts than among prestige texts. There was thus little support for Robinson’s hypothesis that popular texts are translated in a domesticating manner. Instead, the findings seem to suggest that translations comply with the tendency of popular usage playing a larger role than prestige language in changing linguistic practices.
It seems that although the genre of self-help books has been adopted to the Finnish culture from the U.S. - essentially via translations, which is how genres typically travel - the genre itself has been adapted. Despite their differences, the translations and originals also showed affinities which distinguished them from other popular non-fiction. The translations had probably helped shape the genre in the target culture, even though Finnish self-help writers have obviously also drawn on untranslated American sources. The Finnish self-help book has apparently resulted from combining cultural writing practices (how you write in Finland / for Finns) with transferred generic practices (how you write in this genre). The reproduction of the genre in its new cultural environment is thus an adaptive response to its conditions of production, with its diverse social and linguistic determinants.
So where IS cultural adaptation? On the basis of this study I would say it is in the genre. Not in the translations, despite the express ideal of domestication. Translations, at least in the genre now investigated, seem largely to transmit source culture linguistic and pragmatic practices, ignore target-specific usage, change target culture practices and only partially adapt to it. Whether such influences on the target language and culture ought to be discouraged is another matter - one of prioritising norms.
Aijmer, Karin, Bengt Altenberg and Mats Johansson (eds) (1996) Languages in Contrast, Lund University Press
Baker, Mona (1993) “Corpus Linguistics and Translation Studies: Implications and Applications”, in Mona Baker, Gill Francis and Elena Tognini-Bonelli (eds.) Text and Technology: In Honour of John Sinclair, Amsterdam and Philadelphia: John Benjamins, 233-250.
———(1996) “Corpus-based Translation Studies: The Challenges that Lie Ahead”, in Harold Somers (ed.) Terminology, LSP and Translation: Studies in Language Engineering in Honour of Juan C. Sager, Amsterdam and Philadelphia: John Benjamins, 175-186.
——- (1999) “The Role of Corpora in Investigating the Linguistic Behaviour of Professional Translators”, International Journal of Corpus Linguistics 4(2): 1-18.
Biber, Douglas, Stig Johansson, Geoffrey Leech, Susan Conrad and Edward Finegan (1999) Longman Grammar of Spoken and Written English, London: Longman.
Frawley, William (1984) Prolegomenon to a Theory of Translation. In Frawley, W. (ed.) Translation: Literary, Linguistic and Philosophical Perspectives. London & Toronto: Associated University Presses, 159-175.
Gellerstam, Martin (1986) Translationese in Swedish novels translated from English.” In Lars Wollin and Hans Lindqvist (eds) Translation Studies in Scandinavia. Lund: CWK Gleerup, 88-95.
——- (1996) “Translation as a source for cross-linguistic studies”, in Karin Aijmer, Bengt Altenberg and Mats Johansson (eds.) Languages in Contrast, Lund: Lund University Press, 53-62.
Karhu, Anna (1997) “Suomennosten nominaalilausekkeiden piilopoikkeamat” [Covert errors in noun phrase translations] In Anna Mauranen and Tiina Puurtinen (eds) 1997. Translation, Acquisition, Use. AFinLA Yearbook 1997. University of Jyväskylä, 71-88.
Klingberg, Göte (1986) Children’s Fiction in the Hands of Translators. Lund: Gleerup
Kujamäki, Pekka (1998) Deutsche Stimmen der Sieben Brüder. Frankfurt a.M: Peter Lang.
Laviosa-Braithwaite, Sara (1996) The English Comparable Corpus (ECC): A Resource and a Methodology for the Empirical Study of Translation, UMIST: Unpublished PhD thesis.
Luukka, Minna-Riitta (1995) Puhuttua ja kirjoitettua tiedettä [Spoken and written science] University of Jyväskylä.
Mauranen, Anna (1998) Käännössuomi ja kääntämisen universaalit: tutkimus korpusaineistolla [Translated Finnish and Translation Universals: A Corpus Study], Project Research Plan, MS, Savonlinna School of Translation Studies.
——- (2000) “Strange Strings in Translated Language: A Study on Corpora”. In Maeve Olohan (ed.) Intercultural Faultlines. Research Models in Translation Studies 1: Textual and Cognitive Aspects. Manchester: St. Jerome Publishing, 119 -141.
Puurtinen, Tiina (1995) Linguistic Acceptability in Translated Children’s Literature, Joensuu: University of Joensuu Publications in the Humanities, 15.
Robinson, Douglas (1997) Translation and Empire: Postcolonial Approaches Explained, Manchester: St. Jerome Publishing.
Tirkkonen-Condit, Sonja (2000) “In Search of Translation Universals: Non-equivalence or “Unique” Items in a Corpus Test”, paper presented at the UMIST/UCL Research Models in Translation Studies Conference, Manchester, 28-30 April 2000.
Venuti, Lawrence (1995) The Translator’s Invisibility. A History of Translation. London: Routledge.