This paper sets out to investigate quality in the translation of tourist discourse on the Web. It is divided into two parts. The first discusses theoretical issues: localisation and translation; which theoretical framework for Web translation; the criteria to assess translation quality. The second part is an investigation into a corpus of UK and Italian tourism Web sites: the criteria outlined are employed to explore transfer strategies, and translation problems and errors. The focus is on the tourist language regarding accommodation; the data considered are taken from the accommodation page of tourist board sites, and hotel sites.
Translation quality, Web translation, localisation, localization, Web advertising, tourist discourse.
With the digitisation of information, the promotion and the marketing of products within global markets is increasingly prominent on the Internet. When compared to traditional media (press, radio, etc.), which are purely promotional tools, the Web provides further possibilities: Web pages can be either promotional or commercial, or both promotional and commercial. The tourism industry, now a global enterprise promoting tourist products both in domestic and international markets, has captured the relevance of this new mass medium for contacting potential receivers all over the world (Antonioli Corigliano & Baggio 2002: 27-96). A hypertext can be produced with ease and periodically updated; it is not expensive; it is fast in spreading the message; it can be received throughout the world. Due to its interactive nature, it combines the characteristics of mass communication and interpersonal communication: the message addresses a mass audience (potentially, a heterogeneous audience in terms of nationality, social class, age, education, etc., dislocated all over the world), but allows for a reciprocal exchange between sender and receiver via e-mail and newsgroups.
Tourism is a fast-growing market sector on the Web (Marcussen, 2006), where many site types are found (Antonioli Corigliano & Baggio 2002: 62-63; Giordana 2004: 186-188). They can be grouped into:
Tourist communication on the Web tends to occur simultaneously in a variety of languages, leading to different transfer practices we refer to with the term Web Translation, conceived as a type of Multimedia Translation (Gambier & Gottlieb, 2001). The issues involved are manifold, concerning Translation Studies as well as Media Studies, semiotics, ICT, intercultural communication (tourist communication implies establishing a contact between cultures which leads to an experience of the other), tourism studies, and economics in its various aspects (advertising, management, marketing).
Much has been written about advertising texts in traditional media, and the issues involved in their translation (e.g. Adab & Valdés, 2004). Not much, on the other hand, has been written about Web advertising (e.g. Janoschka, 2004) and the complexities involved in the transfer of promotional and commercial Web pages (e.g. Cheng, 2000; Chiaro, 2004). This paper, linking translation, Web communication and tourism, investigates quality in the translation of tourist discourse on the Web. It is divided into two parts. The first discusses theoretical issues: localisation and translation; which theoretical framework for Web translation; the criteria to assess translation quality, crucial in texts which need to communicate effectively in order to be successfully promotional, and to convey a positive image of the product/service and its provider. The second part is an investigation into a corpus of UK and Italian tourism Web sites: the criteria outlined are employed to explore transfer strategies, and translation problems and errors. The focus is on tourist discourse regarding accommodation; the data are taken from the accommodation page of tourist board sites, and hotel sites, visited over the period from April to October 2006. Some tourist board sites are both promotional and commercial: they promote a place, while marketing its products. For example, users can book accommodation directly, and purchase special offers expressly made for Web visitors (e.g. <http://www.visitlondon.com>; <http://www.firenzeturismo.it>). The sites visited are bilingual or multilingual: users can choose their preferred version by clicking on flag icons (i.e. image links) on the homepage.
When discussing international marketing, it is necessary to introduce three terms. The first is globalisation, i.e. the business process of making a product available for sale in many countries. Once a company has decided which countries to enter, two processes take place ¾ internationalisation and localisation. The former is the engineering process of making a product easier to use or adapting for use in other countries. Localisation, commonly associated with high-tech and digital products (e.g. software programmes, video games and Web sites), is concerned with any good or service. It is defined by Esselink (2000:1, 3) as follows:
Localisation is the translation and adaptation of a web product, which includes all related product documentation. The term 'localisation' is derived from the word 'locale', which traditionally means a small area or vicinity. Today, locale is mostly used in a technical context, where it represents a specific combination of language, region and character encoding. For example, the French spoken in Canada is a different locale to the French spoken in France. [...] Localisation involves taking a product and making it linguistically and culturally appropriate to the target locale (country/region and language) where it will be used and sold.
Localisation is a complex process taking into account not only the socio-cultural specificities of the target locale (e.g. behaviour patterns and values of the target culture), but also the politico-legal ones (e.g. regulations related to information and to certain products, such as tobacco) (Guidére, 2003: 3-4).
Within localisation theory, translation is viewed as
the process of converting written text or spoken words to another language. It requires that the full meaning of the source material be accurately rendered into the target language, with special attention paid to cultural nuance and style. [my emphasis] (Esselink 2000:4)
The localisation of Web sites involves the translation of all pages, including all purely linguistic elements ¾ lexias (graphically separated blocks of writing), alt(ernative) texts, links, etc., and support documents. When needed, it also includes the adaptation of the source text (ST) to ensure the acceptance and the success of the product in the local market: e.g. the adaptation of dates, weights, measures and currencies; the modification of content, by omitting information not relevant to target receivers (TR), adding extra information, or practising a summary translation of the ST; a change in style, by employing the writing style conventionally used in the target culture (TC).
The situation on the Web is varied, and localisation may vary by degree (full translation, partial translation or adaptation, implying revision or recreation), and by communication strategy, e.g. a network of local monolingual Web sites, targeting different countries, or one single bilingual (with English as one of the two languages), or multilingual Web site. With regard to tourist communication in Europe, national tourist board sites tend to be multilingual (Pierini, 2006).
The relationship between localisation and translation is an issue discussed at length in Pym (2004). The localisation industry pays increasing attention to translation and translation quality (Bass, 2006). This is reflected in the naming of a new composite sector bringing together globalisation, internationalisation, localisation and translation under the acronym GILT (Dunne 2006:4).
Localisation and translation are distinct areas, but it is more fruitful to consider the commonalities existing between them (Munday 2004: 202), i.e. common concepts, common problems, common solutions. From the two passages quoted in this section, we can identify two points shared by the two fields: the concern for culture and the need for adaptation.
Over the last years, Translation Studies have evolved from the concern for the linguistic aspects of traditional paperwork genres to the concern for the multisemiotic aspects of multimedia texts (Kress & van Leeuwen, 2001). Within the field of Multimedia Translation, Web translation is in search of a theoretical framework.
In Web translation, we are confronted with the Web page, forcing us to rethink the concept of 'text', 'genre' and 'meaning'. The Web page as text (Garzone, 2002) is a multisemiotic entity resulting from the interplay of various resources ¾ written language (and sometimes spoken language), pictorial (images, icons), graphic (layout, typographic features) and acoustic elements (sounds, music). So, meaning is generated by the totality of verbal utterances and non-verbal signs. The Web page as genre can be profiled as a class of communicative events characterised by the interaction of verbal and non-verbal features, fulfilling a set of communicative purposes (e.g. informing, influencing the receiver's behaviour), which can be divided into subgenres, e.g. the Web advert and the blog.
The basic features of the promotional Web pages under examination can be outlined as follows: their communicative purpose is to persuade; the message sender is a body (board, firm); the target audience is the general public, in the case of tourist boards, and a specific market segment, defined by the product (e.g. a 2-star hotel or a resort hotel) and its presentation, in the case of hotels. Users may be characterised as people interested in gathering information, planning their journey or buying special offers available to Web visitors.
As to the non-verbal component, the feature of this genre is the important role played by attractive photos. As to the verbal component, the language of tourism is a specialised discourse that, when used in the interaction between experts (tourist professionals) and non-experts (varied audiences), as in Web adverts, tends to be more similar to general language (Gotti 2006: 20-21). English tourist discourse in tourist board sites exhibits the following features (Pierini forthcoming), also observed in hotel sites:
Some of the features identified seem to characterise promotional tourist discourse across genres and media: for example, the imperative, figures of speech, clichés and the direct appeal to the individual are also found in tourist brochures and print adverts (Dann 1996: 79 fll, 171 fll, 186-187, 194-195). Others are general features of Web advertising (Janoschka 2004: 121-158), such as virtual orality, or features of Web communication, such as light style.
The multisemiotic nature of the Web page requires specific methodological guidelines for its transfer. One of the first scholars to discuss this issue is Cattrysse (2001). The first point of his proposal is to integrate the translation of the verbal component with the visual components of the message (images, icons, etc.). Also Cheng (2000: 38) and Bass (2006: 78-81) point out that an understanding of the semiotic context is of primary importance for the translation of the verbal component. The second point is that, if language is the most obvious element to change in order to reach a new audience, it is not the only one. Although images and icons usually cross cultural borders, not all visuals function in a universal way: e.g. colour may convey values which vary across cultures. In some cases, the transfer of the ST may require the adaptation of non-verbal material, for example, when the visual may be offensive for TR, or not well-understood. Obviously, this may have consequences on the verbal component.
It follows that there are four possible strategies: 1) the verbal component is translated without any adaptation, and the non-verbal one is left unmodified; 2) only small (verbal and/or non-verbal) parts of the ST are modified; 3) the verbal component is completely rewritten, leaving unchanged the non-verbal one; 4) both components are modified.
Since our concern is for Web adverts, let us now consider the transfer of advertising texts in traditional media. The studies on this topic (e.g. Guidère 2001; Munday 2004) have highlighted that the translation of the ST involves the production of a TT achieving the same effect (persuasion) as the ST. This may require the strategy of adaptation: the TT may be modified verbally and/or visually, in order to suit the message to the TC. Studies on the transfer of tourist brochures (Snell-Hornby, 1999; Sumberg, 2004) argue in favour of a similar approach: the tourist brochure is a text whose extralinguistic effect has to take precedence over its content and form, thus requiring an adaptation to the norms and expectations of TR.
The translation of Web pages can be accounted for by a theoretical framework re-orienting traditional models: on one hand, approaches which view translation as a process centred on text-in-context, aimed at the transfer of the semantic, pragmatic and cultural content of the ST to the TC for TR (Neubert & Shreve 1992; Snell-Hornby 1995, with different focus); on the other hand, functionalist approaches, considering translation as a purposeful activity, where text function is central (Nord 1997). The relation between the ST and the TT is held as a functional (pragmatic) equivalence, considering text as the unit for equivalence. Both types of models, originally designed for mono- or bimodal texts, can be semiotically re-oriented, to include multimedia texts.
In short, a theoretical approach to Web translation should integrate the following points in a comprehensive framework: a) the concept of translation as cross-cultural transfer; b) the centring of the translation process on the Web page as text; c) a semiotic perspective; d) the focus on the target audience; e) the concern for the communicative purpose of the TT (persuasive effect must remain a central concern for translators of promotional/commercial sites); f) the need for adaptation; g) an in-context approach (verbal elements should be viewed in the context of the original Web pages).
It is well-known that the translation quality of tourist discourse in written material is not always good. This 'tendency', which is not language-pair specific, is recognised and discussed in theoretical studies (Neubert & Shreve 1992:104-109; Chesterman 1997: 129-130) as well as empirical investigations (e.g. Sumberg 2004: 344).
Translation quality assessment, which can vary in scope and aim, is an issue extensively discussed in the literature from different perspectives, e.g. Kussmaul (1995: 127-148), Hatim & Mason (1997: 164-178, 197-212), Chesterman (1997: 121-145), House (1997), and Scarpa (2001: 171-202). If quality is a thorny issue in the translation of traditional genres, it is even more complex in Web translation. Linguistic criteria, which still remain relevant, are no longer the only parameters on which assessment can rely. Other criteria are needed, related to the genre (the advertising text) and the medium (the Web).
The first criterion we are going to discuss is the relation between the ST and the TT, which is one of pragmatic intertextual equivalence. The relation can be broken down into three aspects, each assessed separately ¾ the semantic, the pragmatic and the cultural. According to the approach outlined in Section 2.2.3, the semantic and the cultural content may be modified in order to adapt the message for TR. What is central is the pragmatic relation: the TT should achieve the same effect as the ST (the successful promotion of the tourist product), using appropriate means for the TC.
The second criterion is centred on the TT: it should meet the standards required for any text produced in the TC ¾ adequacy and acceptability ¾ established with reference to the norms operating in the TC, which can be distinguished into: a) norms of the TL system; b) general usage norms; c) genre-specific norms, with respect to the various genres produced in the TC; they are also called conventions because they only express preferences and are culturally determined. Adequacy is related to the language system norms, and refers to morpho-syntactic correctness and lexical accuracy. Acceptability is related to usage norms and genre-specific conventions, and refers to the adherence of the TT to both. The TT should be indistinguishable from native, non-translated texts belonging to the same genre, and the adjustments required by 'normalisation' should have the effect of writing a TT familiar to TR. An approach based on adequacy and acceptability entails a notion of 'translation error' as a deviation from any of the three types of norms, and the classification of errors into two broad categories: a) linguistic errors, resulting from breaches of the TL system, which are 'binary' (choices are right or wrong); b) pragmatic and cultural errors, resulting from deviations from preferred usage in relation to discourse and genre, and a lack of cultural filtering; they are non-binary, covering a range of gradation from 'preferred usage' through 'acceptable' to 'deviant' and 'definitely unacceptable'.
The criterion related to the genre in point is effectiveness. To be successfully persuasive, the TT should be modified, when needed, to fit the tastes of the target market by adopting various strategies. Giordana (2003: 90-91) argues that the message should meet the classical requirements of advertising discourse: AIDA ¾ capture Attention, create Interest, increase Desire and motivate to Action. (To catch and hold the receiver's attention is particularly important on the Web, since most users tend to surf the Net). Dann (1996: 174) observes that the message should be short, clear, active and include key words (e.g. away, escape, adventure, dream, pleasure, imagination, experience), which correspond to the socio-psychological needs of the potential tourist. Janoschka (2004: 132-149) remarks that, besides emotionally appealing strategies, Web adverts need, more than traditional advertising texts, to interact with users: they should create a conversational relationship with users, imitating interpersonal communication. Sumberg (2004: 336-337) and Cheng (2000:42) point out that translators should tailor the message under the guidance of a targeted marketing strategy.
The last criterion is related to Web communication, where the quality of Web sites is measured in terms of usability (Nielsen 1999; Fiormonte 2004), i.e. the ease of site handling. Among the usability factors (Web site design, navigability, searchability, etc.), the factor concerning discourse is legibility. Studies on how people read on the Web have shown that screen reading is slower than print reading, and that users do not actually read: instead, they scan the Web page. Therefore, Web writing requires: light style (use of simple sentences, and discourse organisation in short lexias); conciseness (less information to process); one idea per paragraph; the inverted pyramid style (from main information to details); typographic features enhancing scanning (Carrada 2000).
My working definition of translation quality, which will be applied to the analysis of data in the next section, is as follows: the TT shows a good quality when it is a native-like text, well-written according to the genre style and Web writing style, achieves the intended effect (persuasion), and does not exhibit unintended side effects (humour, offence).
In the Web sites visited, the transfer of the ST reflects the adoption of three strategies: 1) full translation, leaving non-verbal elements unchanged; 2) partial translation, or summary translation; 3) modifications of small parts of the ST.
I shall now discuss some examples concerning the verbal component of Web pages, by comparing English textual segments and their translations into Italian, and vice versa. The first is taken from the site of Scotland's National Tourism Board <www.visitscotland.com>, where the Italian version is an example of 'partial translation': not all pages are translated into Italian. In particular, the page containing the search engine, with which users can search the data bank for accommodation, and select information by entering keywords, is available only in English. For this reason, the translator made changes to the Accommodation page, as shown in (1a) and (1b).
(1a) A Warm Welcome. Guaranteed. And for something just a little out of the ordinary - perhaps a converted church, a former lighthouse, or even a medieval castle, there are plenty of opportunities available to fire your imagination. Whatever your budget or itinerary, you can find your ideal place to stay using the on-line accommodation search facility or by getting in touch with our contact centre. All establishments have been inspected under the Visit Scotland Quality Assurance Scheme so you can be certain of the very best facilities where, of course, a warm welcome and true Scottish hospitality come as standard. [paragraphs added: To book online (in English), click on 'Accommodation: Search Online' placed at the top of this page. You can choose by: accommodation type, town or area. You can also book your accommodation at a tourist office when you arrive in Scotland, or at the VisitScotland Centre in London.
For help with your search, the following translations can be useful: [....]
|
(1b) Calorosa accoglienza garantita L'insuperabile scelta di modi di fare vacanza in Scozia è pari solo alla varietà di strutture ricettive offerte dal paese. Sia per un breve soggiorno che per uno più prolungato, avete a disposizione la scelta tra lo chic dei grandi alberghi metropolitani, l'eleganza delle ville di campagna, l'accoglienza ospitale dei Bed & Breakfast, e la comodità degli appartamenti in affitto. [final sentence omitted]
[paragraph omitted]
Qualunque sia il vostro budget e il vostro itinerario, potrete trovare il vostro posto ideale per fermarvi, sia utilizzando i sistemi di ricerca degli alloggi che si trovano on-line, sia rivolgendovi al nostro centro di contatto. [paragraph omitted]
Per prenotare on-line (in inglese) fate clic sull'opzione 'Alloggi: Ricerca on-line' che si trova in alto a sinistra di questa pagina. Potrete scegliere secondo il tipo di alloggio, la città, o la regione. Altrimenti potete prenotare l'alloggio presso un ufficio del turismo quando arrivate in Scozia, o anche presso l'ufficio di VisitScotland a Londra.
|
In (1b), we observe, on one hand, the omission of two paragraphs, one about unusual types of accommodation, probably deemed of no interest to the Italian market, and one about the guarantee of accommodation quality, considered non-relevant. On the other hand, there is the addition of two paragraphs containing instructions on how to use the search engine and to book accommodation, and a bilingual list of terms denoting the various accommodation types. The TT exhibits an adequate and acceptable quality in linguistic terms, a certain degree of adaptation (a re-arrangement of content with omission and addition of information), and a good level of legibility. As a result, (1b) seems to be an effective message.
Another point concerns the search results, delivered as a list of accommodation in the form of clickable links. By activating them, users can access short informative fact sheets ¾ constituting the support documentation ¾ usually translated into the TL. This brings us to a phenomenon observed in two sites. The fact sheet in Figure 1, taken from the site of ENIT (Italy's National Tourist Board), is part in English and part in Italian, and the alternative texts, appearing when the cursor passes over the icons, are in Italian.
Figure 1 – Fact sheet from <www.enit.it>, accessed on 28.4.2006
A similar mix of languages, representing a 'localisation error', is shown in Figure 2, taken from the site of the British Tourist Authority. These cases may be explained by the fact that inserting the translated data is expensive, considering the huge amount of accommodation firms. Pym (2004: 11) maintains that the degree of localisation depends on two factors: the size of the market concerned (the bigger the locale, the more products can be sold to it, the more resources are invested in localisation); the degree of its linguistic and cultural diversity ('exotic' locales require more investments and complete localisation with major changes).
Figure 2 – Search results from <www.visitbritain.com>, accessed on 10.10.2006
The next examples concern the sites of two British hotels. The Italian version in <www.thelodgehotel.co.uk> consists of one Web page resulting from the merging and translation of two pages taken from the ten that make up the site; but it is of little use to TR, since basic information about the product (types of room and rates) is missing. Different is the case of <www.dorchesterhotel.com>, where the Italian version is a summary of the content of the whole site with selected images, in the form of a fact sheet in PDF format.
A major problem in translating tourist discourse, which describes and informs about 'other' cultures, is cultural. In (1b) above, we can observe a lexico-cultural problem in rendering three terms denoting types of accommodation: bothy, chiefly used in Scotland to denote a cottage or a hut, is not translated (it may be rendered with capanno, cottage); bunkhouse is rendered with rifugio by approximation; as to lodge, we may consider the Italian locandaas rendering both lodge and inn by approximation. These terms constitute problems because referring to entities of Anglo-Saxon 'material culture' (Newmark 2001: 97-98), which have no equivalents in Italian culture.
Another problem is posed by different stylistic conventions. In (2), we can see a segment in Italian and its translation into English, taken from <www.enit.it>:
(2a) Dove dormire in Italia? Esiste solo l'imbarazzo della scelta. Agli oltre trentatremila alberghi, disseminati in ogni località, si aggiungono altri trentacinquemila indirizzi di campeggi, alloggi agrituristici, bed & breakfast, ostelli per la gioventù, alloggi privati e così via. Un'offerta vasta e differenziata, che comprende le soluzioni economiche e quelle più lussuose, le più semplici e quelle più raffinate: tutte, comunque, in linea con la tradizionale ospitalità italiana, fatta di attenzione, cortesia e premura. Alle nostre banche dati nazionali, che per ora comprendono alberghi e campeggi, si aggiungeranno presto le informazioni riguardanti gli altri tipi di alloggio. |
(2b) Where to sleep in Italy? There is only an embarrassment of choice. In addition to more than thirty thousand hotels located in every part of the country, there are an additional thirty five thousand addresses of campsites, country farmhouses, bed and breakfasts, youth hostels, private accommodation and so forth. A vast and diverse range of solutions which includes the cheapest and the most luxurious, the most simple and the most sophisticated – all, however, in keeping with the tradition of Italian hospitality made up of attention and care. Information about other types of accommodation will soon be added to our national databanks which currently include hotels and campsites. |
The TT (2b) is a long, not very effective lexia, bearing the signs of a strong interference from Italian (2a). Interference of the ST is a well-known and pervasive phenomenon: Toury (1995: 274 fll) talks of a "law of interference" and states that translators ¾ including professionals, to some extent ¾ universally tend to be influenced by the language of the ST in a variety of ways.
The translator transferred Italian discursive and linguistic features into the TT: a) non-personalisation, i.e. lack of first and second person pronouns; b) 'heavy' style, with long and complex sentences, and presence of less-important information (e.g. more than thirty thousand), thus deviating from light style and conciseness characterising Web writing; c) syntactic 'oddities' (e.g. There is only an embarrassment of choice); d) lexical choices: dove dormire in Italia® where to sleep, alloggi privati ® private accommodation). In short, (2b) exhibits 'clumsy' language with signs of non-nativeness due to interference and non-adherence to the stylistic conventions established for tourism Web adverts in English, as outlined in Section 2.2.1.
As an example of a 'native-like' TT, in line with English conventions, I would like to suggest a possible rewriting in (2c), which adopts a promotional tourist style and tries to balance between keeping useful information and making the paragraph concise and easy to read:
(2c) Where to stay in Italy? There is a whole range of accommodation options to suit any taste or budget. You can choose from basic or luxurious hotels, well-equipped campsites and quality hostels, cosy guest houses and comfortable B & Bs, self-catering accommodation, and farmhouses for contact with the countryside. Whatever your destination, you will enjoy the warmth of traditional Italian hospitality.[...]
Pragmatic problems are also related to differences in lexical choice and collocation between Italian and English. Consider (3), taken from <www.thelodgehotel.co.uk>:
(3a) The Lodge Hotel Edinburgh is owned and managed by Jean and Steve de Roeck with the help of a small, dedicated staff.
(3b) Il Lodge Hotel di Edimburgo è gestito da Jean e Steve de Roeck con l'assistenza di un ristretto numero di collaboratori fedeli e cordiali.
Dedicated staff, translated in (3b) with a sequence of words not occurring in Italian for tourism,can be rendered with a more 'natural' uno staff attento alle esigenze dei clienti. In this sense, specialised electronic corpora can provide interesting insights into genre- and domain-specific usage (Tognini Bonelli & Manca, 2002; Wilkinson, 2005).
In order to illustrate the range of errors made in translation practice, let us now examine two segments in Italian and their translations into English, taken from <www.abbaziahotel.com>:
(4a) L'Hotel Abbazia è situato in pieno centro storico, a soli 150 metri dalla stazione ferroviaria "Santa Lucia" e a circa dieci minuti di cammino dal terminal automobilistico di Piazzale Roma.
(4b) The Hotel Abbazia is situated just in the historical centre, only 150 metres from the railway station "Santa Lucia" and walking about eight minutes from the car terminal "Piazzale Roma".
(5a) Il recente restauro ha conservato il fascino dell'antica abbazia, nella quale sono state ricavate 50 camere, parti delle quali con vista sul giardino interno, accessibile a tutti i clienti. Tutte le camere sono dotate di servizi privati, asciugacapelli, aria condizionata / riscaldamento, telefono (con presa per accesso alla rete), cassaforte, TV sat. e minibar.
(5b) The recent restoration wanted to preserve the charm of the ancient Abbey, where there have been opened 50 rooms, a part of which looking a private inner garden. All the rooms are supplied with bathroom, air conditioner (in summer) or central heating, direct dialling telephone, TV sat. and mini-bar.
(4b) and (5b) are not adequate, showing obvious lexical and syntactic errors, and deviant occurrences, unacceptable in some points, because of pragmatic errors. In addition, in (5b) some lexical items were skipped by the translator (accessibile atutti i clienti, asciugacapelli, cassaforte). Doubtless, this text was not translated by a professional translator, thus pointing to one cause of poor translation of tourist language ¾ translations are sometimes commissioned to non-professional translators. Below I offer alternative solutions and some comments:
(4c) The Abbazia Hotel is located/situated in the heart of Venice /in the city centre, very close to the "Santa Lucia" railway station, and within (easy) walking distance of the car park in Piazzale Roma.
(5c) The building has recently undergone a substantial refurbishment which has preserved the charm of the ancient Abbey. The hotel has 50 rooms, some overlooking a secret garden where guests are welcome to sit and relax. All (of our) rooms have/are equipped with en-suite /private bathroom, air conditioning, hairdryer, direct-dial telephone, safe, satellite TV and minibar.
I have corrected the errors and reformulated the segments with typical lexical items and collocations occurring in English for tourism, selected on the basis of concordances extracted from a small electronic corpus of British hotel Web pages built ad hoc, which now provisionally stands at 66,000 words.
In (5c), the first sentence has been split into two sentences, in order to adhere to English light style and make the segment more readable. I have reformulated deviant (there have been opened 50 rooms)and wrong constructions (a part of which looking). The Italian sequence giardino interno accessibile a tutti i clienti in (5a) has posed a translation problem. A secluded, inner garden is referred to in English by the culturally-biased adjective secret. The postmodifier accessibile a tutti i clienti can be interpreted as a permission or an invitation: hotel guests can go into the garden, and do what? Some help for the translator is provided by a photograph showing a garden where we can see small tables and chairs. Further relevant help comes from the study carried out on two comparable corpora in English and Italian by Tognini Bonelli & Manca (2002: 328-329): the unit guests + be welcome + to is functionally equivalent (i.e. equivalent at a pragmatic intertextual level) to the Italian sequences gliospiti possono + infinitive, and a disposizione degli ospiti c'è / ci sonox. For this reason I have opted in favour of the rendering guests are welcome to sit and relax. The alternative solutions in the final sentence, have /are equipped with vs. other potential renderings (offer) and en-suite / private (bathroom), are based on corpus evidence.
In our discussion of quality in Web translation, we have suggested to integrate the general linguistic criteria of adequacy and acceptability with criteria related to the genre (effectiveness) and the medium (legibility). The objective to pursue is to produce a 'natural' TT without errors: even if the TT as a whole makes sense within the context, any error of any type weakens the persuasive effect. As users, we are accustomed to faux-pas in tourist texts, we may also find them amusing, but they do not communicate effectively. Solid, well-written, grammatically correct content is crucial to gaining the user's trust.
Considering that the data examined are limited, and only partially reflect Web reality, four points emerge from our investigation. The first is a preference for the full translation of the verbal message, with a very low or zero degree of adaptation, while non-verbal elements are left unchanged. The second point is that we have observed cases of mixes of languages, partial translation, and summary translation. Such practices, accompanied by a lack of adaptation, which implies the lack of a targeted marketing strategy, hinder the effectiveness of the message.
With regard to legibility, discourse is usually legible in linguistic and typographic terms, even if some site publishers do not realise that Web communication is not print on screen. The fourth point is that the translation quality of tourist discourse is not always good. Overall, it is rather good in tourist board sites, since their publishers have the means and the expertise, while it is sometimes poor in hotel sites. Hotels, which are usually small companies, may not have the budget or the know-how, and do not fully exploit the possibilities offered by the Internet. At present, the exploitation of Web resources can be considered as a strategic asset with respect to competition, in that it enables the company to stand out in the highly competitive tourism market.
As is well-known, many factors may hinder translation quality, e.g. time pressure, budget constraints, low quality of source materials, technological problems (Bass 2006: 70-76, 84 fll). We would like to focus on three factors ¾ one internal and the other two external to texts. The complexity of promotional tourist discourse is underestimated by clients and translators: it may appear to be deceptively easy to translate with its extensive use of general language; yet, it is a specialised discourse with specific linguistic/cultural features. For this reason, specialised electronic corpora in the TL can be of great help in producing a 'natural' TT.
The other reasons have to do with the translation market, where anyone can call themselves a translator, and with the endless amount of material translated for the Web. Clients commission translations also to non-professionals, sometimes with a limited knowledge of the TL and no specific training in translation or domain-specific languages, who accept low fees and produce quick (dirty) translations. Or, they use machine translation because cheaper, producing texts which often make people smile.
www.visitbritain.com
www.visitscotland.com
www.visitwales.com
www.visitlondon.com
www.visitmanchester.com
www.visitbrighton.com
www.enit.it
www.visitsanmarino.com
www.romaturismo.it
www.firenzeturismo.it
www.triesteturismo.com
www.thelodgehotel.co.uk
www.dorchesterhotel.com
www.thebalmoralhotel.com
www.thelowryhotel.com
www.ibishotel.com/ibis/fichehotel/gb/ibi/2936/fiche
www.aldrovandi.com
www.abbaziahotel.com
www.hotelcolombaflorence.com
www.bellevuecortina.com
www.hotelbalocco.it
Patrizia Pierini is Associate Professor of English Linguistics at the Faculty of Letters, University of 'Roma Tre'. She has published extensively in the fields of contrastive linguistics, translation studies, discourse analysis and ESP. Among her publications: the edited volumes L'atto del tradurre. Aspetti teorici e pratici della traduzione (1999), and Lo sviluppo della competenza traduttiva. Orientamenti, problemi e proposte (2001).
Contact: pierini@uniroma3.it