Introduction
In an influential article entitled ‘The Translation Turn in Cultural Studies’ Susan Bassnett suggests that the study of translation could help us to understand

‘how complex manipulative textual processes take place; how a text is selected for translation, what role the translator plays in that selection, what role an editor, publisher or patron plays, what criteria determine the strategies that will be employed by the translator, how a text might be received in the target system.’ (1998:123)

Constraints of time and space make such an ambitious programme impossible here, but it sets out the parameters for this study

Aims and content
This study aims to compare an English novel and its French and Spanish translations and to set them in the context of studies of literary and general translation, and current directions- the ‘cultural turn’ –in Translation Studies. The novel’s style is discussed in terms of the registers used. The novel and its translations are contextualised within the publishing industry’s promotional strategies and target audience(s). The body of the research consists of detailed comparisons of a chapter of the Spanish and French translations with each other and with the English original, undertaken to establish to what extent and in what ways the source text and the target texts differ, and to explain these differences. Some statistical data on elements of the text are also presented. Finally, some tentative conclusions are drawn, and directions for further research suggested.

The Novel
Tulip Fever recounts a love affair between a young woman, Sophia, and Jan, the artist commissioned by her elderly husband Cornelis to paint their portraits. The setting is Holland in the 17th century, the golden age of Dutch painting and a time of financial speculation in tulip bulbs. It belongs to the genre of popular fiction, and the sub- genres of historical and romantic fiction. However the pace of the narrative and the complex plot, involving hidden pregnancy, feigned death and financial speculation, are perhaps more typical of the thriller. The author, Deborah Moggach, is a best-selling novelist, author of over a dozen books, usually dramas centred on romantic and familial relationships within closely observed, sometimes historical, settings .

The Publishers
The market for popular literature is large, although the sector is dominated by a few major ‘blockbusters’, dictated by the concentration of the publishing sector in the hands of major publishing firms with interests in other areas of ‘leisure’ and media activity as a result of globalisation. Tulip Fever is a case in point. It has been translated into all the major European languages, as well as Turkish, Finnish and Arabic, and is a global bestseller. It was published in 2000 as a Vintage Paperback, a Random House imprint for quality popular fiction. Random House was recently taken over by the German-based media conglomerate Bertelsmann, one of the largest media conglomerates in the world (McNamara 2005), which also owns the publishers of the Spanish and French translations, so the profits are held in-house.

The Spanish translation was published by Plaza and Janés Editores, Barcelona, as Por amor a Sofia, in 2000. Plaza and Janés is a major Spanish publishing house specialising in both popular and quality fiction, both in Spanish and in translation. Their wide range of authors includes Mario Vargas Llosa, Isabel Allende and the English crime novelist Ruth Rendell. Set up in 1959, they owed their initial success to the publication of major international bestsellers. Their products are aimed at a wide cross-section of readers.

Club France Loisirs is a book club with 3.8 million subscribers in France and another million in other Francophone countries, and sells 24 million books a year. It has branched into other leisure retailing: theatre tickets, holidays, music videos and DVDs. Its overall bestsellers in 2004 was a novel translated from English, Colleen McCullough’s Le temps de L’amour’ (A Time for Love). Thus Tulip Fever, published under the title Le peintre des vanités, (The Painter of Vanities), fits well into their product range.

the marketing
The physical appearance of the books reflects marketing decisions. The English paperback’s theme, appearance- it has a woman’s face on the cover- and publicity material indicate it is aimed at a middlebrow female readership . Its inclusion of colour reproductions of 17th century Dutch paintings and a list of sources suggests a more academic readership, as does the title, which refers to the wild financial speculation in rare tulip bulbs which swept Holland in the early 17th century, an important component of the novel’s plot. However the cover, with the novel’s name embossed in scarlet letters, points to a popular rather than ‘literary’ novel.

The Spanish version is a good-quality hardback, relatively expensive at 24.96 euros. The full-colour gloss slipcover features a detail from Vermeer’s Woman in Blue Reading a Letter . The ‘blurb’ on the back cover emphasises the book’s historical setting and art-historical references. Like the French version, though unlike the Italian , it omits the reproductions of Dutch Old Masters, possibly for reasons of cost, but includes the original’s list of sources. This presentation suggests a slightly more affluent and discerning readership than the English original, or the same imprint’s paperback crime fiction. The book’s title, Por amor a Sofia, suggests it is aimed at a female market.

The French book is a smaller format hardback, with lower quality print and paper. Its full-colour slipcover features a detail of a study of a tulip by the French painter, Antoine Berjon . The title, Le peintre des vanités (The Painter of Vanities) draws attention to the artistic strand of the novel: Jan is a professional artist. However, the cover blurb focuses on the love interest: ‘Amsterdam, 1636. Riche marchant marié à la jeune et belle Sophia, Cornelis Sandvoort est un homme comblé mais vieillissant...’ (Cornelis Sandvoort, a rich merchant married to the young and beautiful Sophia, is a fortunate man, but an ageing one...’). The book is probably aimed at a loyal female readership- the publisher is a book club- who will build up a collection of romantic novels.

the translators
The choice of translator also reflects marketing strategies. The Spanish publisher’s decision to engage a prize-winning novelist as translator suggests a desire to appeal to a slightly more highbrow readership than that aimed at by the French publishers. The targeted readership affects how the translators view their task.

The Spanish translator, Eduardo Iriarte Goñi, has a high literary pedigree: a published author in his own right, author of Similacros de vida published in 2002 and winner of the Spanish XXIX Gabriel Sijé Prize for a short novel for his Sombras lentas que caen in 2004. He is an independent publisher and translator who has translated authors including Gore Vidal and Tom Wolfe. He is accustomed to translating literary works, and would probably have approached this novel as such.

The French translator, Martine C. Desoille, is an extremely prolific and versatile translator who published a dozen translations between 1990 and 2004. These include several books by the romantic novelists Maeve Binchy and Colleen McCulloch, a historical novel, Samurai, by Takashi Matsuoko (written in French) set in 19th century Japan, and three ‘noir’ police thrillers by Susan Isaacs. Her experience lies in translating various genres of popular fiction and this will have influenced her approach to this novel.

style
A popular novel, by definition, aims at a mass readership, so the style is likely to be more accessible and less idiosyncratic than that of ‘literary’ or experimental writing. This is certainly true of Tulip Fever, whose style is characterised by the use- some might say overuse- of very short, mainly active, sentences, with few descriptive or reflective passages: see the section on ‘Readability’ below. This style gives a somewhat staccato effect, but has the advantage of moving the narrative along rapidly. The narrative itself is conventional, unfolding in strictly chronological order, and is taken up by each character in turn, in short chapters headed by his/her name. The novel uses the present tense throughout, as do both translations, which also reinforces the immediacy of events

register
The novel’s style features a range of registers. Finding equivalent registers in the target text (TT) poses certain problems for the translator. Firstly, there is the difference in register between the language of the various characters, which correlates largely with their age and social class. Cornelis, the aging merchant, uses pompous and orotund language that sounds like lecturing to his young wife. The lovers, Sofia and Jan, use a more natural conversational register, although Sofia’s first-person narrative tends to the poetic and effusive. The servant, Maria and her fishmonger boyfriend Willem use a more popular and earthy style, as presented in table 1.

table 1

CHARACTER ENGLISH FRENCH SPANISH

CORNELIS In this transitory life do we not all crave immortality? p1 Dans cette vie éphémère, ne ne sommes-nous pas tous à l à la recherche de l’immor-talitalité ? p12 En esta vida pere-cedera, ¿acaso no ansiamos todos la inmortalidad? p13
WILLEM Know what I started with,what I scrapedtogether? ... Nearly ruined me too p73 Il m’a coûté un os, ouais. Toute ma bourse y est passée. Une vraie ruine. p92 ¿Sabes con qué em-pecé, qué es lo que conseguí reunir de aquí y de allá? p79

In rendering Cornelis’ speech the French TT adopts a formal and slightly archaic form of question not found in the modern spoken language, while the Spanish TT produces the same effect by using the rare adjective ‘perecedera’ and the uncommon construction with ‘ansiamos’. Willem’s truncated sentences and use of slang are in stark contrast to Cornelis’ verbosity. The French version reproduces these features, using incomplete sentences, slang- ‘Il m’a coûté un os’ and the substandard form ‘ouais’ for ‘oui’. The Spanish TT has fewer obviously slangy features, though ‘¿sabes qué?’ is a popular conversational form, as is ‘de aquí y de allá’.

the quotations
An extract from a contemporary (i.e. 17th century) writer introduces each chapter. In general these quotations are in a more formal and old-fashioned register than the body of the text. Some are short pithy sayings providing a commentary or moral to the chapter, such as ‘where the knot is loose, the string slippeth’ (p202). This obsolete verb ending cannot be rendered in either French or Spanish, so there is some translation loss. The French TT has ‘Quand le nœud se défait, le fil se relâche’ which is less archaic than the English, though it maintains the form of the proverb, while using a reflexive verb rather than the intransitive verb used in the ST. The Spanish TT renders the expression ‘’Donde el nudo anda flojo, salta la cuerda’ which uses the archaic ‘anda flojo’ (‘becomes loose’) and maintains the proverb form. However, reversing the subject-verb order in the second part does not reproduce the repetitive structure of the ST, entailing some translation loss. These quotations add historical colour and depth to the narrative, and the translations endeavour to respect their stylistic differences.

the texts
The French and Spanish translations are dealt with separately. For clarity, long sentences, or comparisons involving three or more examples, are presented in table form.

analytical framework
Clearly, detailed comparison between texts presupposes some kind of prior analytical framework. Hervey and Higgins (1992) on French/English translation and Hervey, Higgins and Haywood (1995) on Spanish/English provide a list of categories from the fields of linguistic and translation theory which I have partially, though not uncritically, adopted. These include: cultural compensation, grammatical and
lexical issues, sentence and discourse, connotative meaning, code and register.

Comparison 1: english/french

code and register
Deliberately archaic words and expressions are used, as in the ST, to evoke the novel’s setting- 17th century Holland. French examples, with their modern equivalents in brackets, are given below.

assidûment (avec zèle)
en particulier (notamment)
vous vous êtes engagé (vous avez promi)
n’aie crainte (n’aie pas peur)

cultural compensation
Hervey and Higgins include under this heading replacement of common ST expressions or constructions by a different but similar one in the TT to compensate for translation loss. This is a somewhat elastic category, further subdivided into compensation in place/by merging/by splitting. Alternatively, one could classify the examples below as functional or dynamic equivalence, to use Nida’s terminology .

table 2
ENGLISH FRENCH

he knows that he is going to go far (p158) sa carrière est déjà toute tracée (p187)
His role model is not Jan... (p159) Il ne veut pas marcher dans les traces de Jan... (P188)
leaves him cold. (p159) le laisse de marbre. (P188)
Some days he doesn’t go near his easel at all. (p160) Parfois, il se passe une journée entière sans qu’il touche un seul pinceau. (p189)

compensation in place
Loss of meaning in one area of the TT paragraph or sentence can be compensated elsewhere. In the TT the first verb does not convey the intentionality of the English ‘plans to’, but the second verb reintroduces it.
.
‘A vingt-cinq ans, il s’établira (he plans to) à son compte et ouvrira son propre atelier. Il veut (will) se spécialiser dans l’art du portrait...’ (p187)

discourse cohesion
‘Dou’s is the style to which Jacob aspires. Neatness and order, not the...’ (p159)
‘C’est ainsi qu’il voulait peindre, car il aime l’ordre et la minutie et non...’ (p188)

In the ST, ‘style’ refers back to, and avoids the repetition of, Dou’s ‘fine detailing’, while the TT replaces this phrase entirely by ‘ainsi’. It also amalgamates the two ST sentences by interposing ‘car il aime’, making it clear why Jacob likes Dou’s style.

‘...it is more common in French than in English for texts to be explicitly structured by the use of connectives ...that signpost the logical relationships between sentences.’ (Hervey and Higgins 1992:49)

In English, cohesion is often accomplished by the repetition of names or anaphoric pronouns, but French avoids repetitions, particularly of names and places, preferring to find alternative expressions , a tendency which Fuller (1984:35) terms ‘elegant variation’. In the paragraph beginning ‘Nowadays, however, Jan...’ (160-1) ‘Jacob’ occurs four times but the TT replaces the second ‘Jacob’ by ‘son apprenti’, and simply omits the third. Other examples of TT cohesion are shown below.

table 3
ENGLISH FRENCH

‘His other idol is Gerrit Dou, a past pupil of Rembrandt van Rijn. How different is Dou from his erratic and temperamental master!’ (p159) ‘Son autre idole est Gerrit Dou, ancien élève de Rembrandt van Rijn, dont il n’a pas l’humeur fantasque et capricieuse.’ (p188)
- Where does the man sleep ? The gutter? A le voir aussi loqueteux, on croirait qu’il dormait dans le ruisseau. (p189)


In the second example there is some translation loss. The French text retains the literal meaning of the original but sacrifices its rhetorical force.

grammatical structure: nouns and verbs
French has a marked preference for nouns where English would use verbs. Verb forms like the infinitive and the past participle can be treated as nouns, as in ‘le savoir-faire’ or ‘l’assassinée’. The ST the verbal construction ‘one’s fame spreads abroad’ (p159) is rendered in French by the nominal construction ‘la renommé de leur auteur’. Verbs are also sometimes replaced by adjectives, as when ‘trembling with anger’ becomes ‘furieux’.

grammatical structure: adverbs
In general French uses fewer adverbs than English. ‘she moves differently too’ (p152) becomes ‘sa démarche aussi a changée’ (p181). An English adverbial phrase may be rendered in French by noun phrase, as when the ST ‘paint to a reliable standard’ becomes the TT ‘Ils executent... des œuvres de qualité’. This is not to say that French avoids adverbs entirely: in one case the TT introduces one not present in the ST, and ‘deliver their canvases on time’ becomes ‘respectent scrupuleusement les delais qui leur sont impartis’ .

grammatical structure : the passive
French tends to avoid passive constructions, so ‘They are commissioned....’ and ‘Jan has been offered...’ become as ‘Ils executent sur commande...’ and ‘Jan reçoit...’. This tendency is given statistical support in the ‘readability’ section below: there were no passives in the French text extract. However the ST ‘on time’ is rendered by a passive, ‘les delais qui leur sont impartis’ (p188)

Other authors classify such grammatical differences under other headings. Jones (1997:77-81) sees noun-verb and adverbial changes as examples of transposition. Some differences fit uneasily within the Hervey and Higgins schema , and could be placed in more than one category, such as cultural compensation, grammatical structure or discourse cohesion, which suggests a certain indeterminacy in their classification, which should be unambiguous and clearcut. I have found it useful to add the following categories.

expansion
‘But not yet. Not now’. (p159) is expanded to ‘mais il est encore trop tôt pour y songer...’ (p189), reflecting perhaps a dislike of the ST staccato style of the original, and/or a preference for well-formed sentences that include a main verb, while ‘at all hours’ becomes ‘à toute heure du jour et de la nuit’. ‘he speculates on which house he will buy’ is expanded to ‘il songe à la maison qu’il se fera construire’, reasonable, given previous reference to houses being built, though unnecessary.

contraction
Given the extremely short sentences of the original, there is little scope for this. One example is Jacob’s reaction to his countrymen’s tulip craze, ‘He feels nothing for it but contempt’ , shortened in the TT to ‘Il les méprise’ . This pithy verdict has more impact than the English original, as does the shortening of ‘And he has turned it down’ to ‘Il refuse’. This, combined with the use of the present tense, adds impact, reinforcing the description of this event as ‘un coup de tonnerre dans un ciel bleu’.

omission
There are a number of instances of omission from the ST. ‘Jacob likes to be in control’ (p159) is completely omitted from the TT, while ‘A boy steps into the room’ (P162) becomes simply ‘Un garçon entre’ (p192). I can give no explanation for such changes, beyond suggesting that the translator considered such information excessive or irrelevant. However, a translation should surely make some attempt to translate the text in its entirety, rather than omitting information for idiosyncratic reasons. Landers warns sternly that

‘deleting any part of the original text is the equivalent of unconditional surrender, an admission that a certain word, phrase or construction is beyond the translator’s ability to render.’ (2001:95)

miscellaneous
Some differences between the ST and TT which are difficult to classify are summarised below.

table 4
ENGLISH FRENCH

1 painting is a trade, like any other la peinture n’est-elle pas une merchandise comme une autre?
2 Where is his professionalism ? N’a-t-il donc pas aucun amour-propre?
3 ‘Why ?’ asks Jacob, his brush poised’ ‘Mais pourquoi? s’étonne Jacob.

In Ex.1, there is a shift of emphasis from painting as an activity to the object of this activity, a reasonable decision. Ex.3 however seems bizarre, as French ‘amour-propre’ means ‘self-esteem’ or ‘pride’, not ‘professionalism’. I have no explanation for this change: perhaps the French translator thought lack of ‘amour-prppre’ would be more shocking to French readers. In Ex.3, the French version makes Jacob’s emotional state more explicit but loses the vivid image of him in mid brush-stroke, frozen by surprise. Later reference to the brush in ‘Trembling with anger, Jacob lays down his brush.’ (p162) also loses its point.

mistranslation
Finally, there is a mistranslation of ‘on the open market’ (p161) which here refers to selling paintings to any buyer rather than painting on commission for a specific buyer. The French translator takes ‘open market’ to mean ‘open-air market’, and translates it as ‘à la foire’. The Spanish translator also makes this error.

summary
The overall impression given by the French text is that the translation is rather free, and in places takes a cavalier attitude to accuracy. Some of the differences can be explained by the differing grammatical structures of French and English and their preferences for certain constructions above others. Elsewhere the TT omits what the translator considers extraneous material to produce a more readable text. In other cases the differences seem unmotivated and random. The aim of the translator seems to have been to produce a racy and readable text in line with the readers’ perceived expectations.

Comparison 2: english/spanish

analytical framework
Hervey, Higgins and Heywood’s Spanish coursebook (1995) follows the schema of the French course book though adding a category of grammatical transposition, under the heading of cultural filters. It is a moot point whether grammatical transposition is a cultural or a purely linguistic phenomenon, but the authors argue strongly that it has ‘important implications for the presentation of different world views’ (1995:215) so should be considered as cultural.

discourse cohesion
The Spanish TT produces a more cohesive discourse by combining short sentences in ST, giving longer sentences: see also the statistical analysis below.Examples include:

‘Where does the man sleep? In the gutter?’ / ¿Donde duerme este hombre, en las cunetas?
‘In two weeks. On urgent business.’ / ‘De aquí a dos semanas, por una cuestión urgente.’

More strikingly, three short sentences may be combined into one long one.

Table 4

Of course, this is a disappointment. Jacob was expecting more instruction. But it has also worked to Jacob’s advantage. (p160) Como es natural, todo ello supone una decepción para Jacob, que confiaba en obtener una mejor instrucción, pero también ha jugado en su favor.’(p158)
A boy steps into the room. For a moment, Jacob thinks: it is all a lie. Jan is getting rid of me so that he can take on another pupil. (p162) Entra en la estancia un chico y por un instante Jacob piensa que todo es un embuste y que Jan intenta librarse de él para coger otro alumno.’ (p159-60)

compensation
ST effects are recreated in a different form. The ST sentence ‘The Swedish ambassador... pays a thousand florins- a thousand- simply...’ (p159) adds emphasis through the use of italics and repetition, while the TT uses an additional phrase: ‘- Mil, nada menos- ’.

grammatical structure 1: verbs versus nouns and adjectives
One clear grammatical difference is the Spanish preference for verbs as opposed to nouns and adjectives. The English adjective in ‘...he will find a suitable girl from a good family...’ is rendered by a Sspanish verb phrase: ‘...encontrera una chica de buena familia que le convenga...’. Similarly, ‘Jan is often out’ is is translated by two verbs- ‘Jan suele ausentarse’, while ‘has not had a haircut for months’ becomes ‘Hace meses que no se corta el pelo’.

grammatical structure 2: the passive
Spanish and English differ in the use of the passive. ‘The passive with ‘ser’ is very rare in spontaneous Peninsular speech’ (Butt and Benjamin 200:382), although increasingly common in Latin American speech and in written, especially newspaper, texts, presumably under the influence of the English passive . Spanish prefers an impersonal construction:
‘They are commissioned’ / ‘se les hacen encargos’
‘new mansions are being erected’ / ‘se están erigiendo las nuevas mansiones’
‘Jan has been offered...’ / ‘A Jan le han ofrecido...’

However, there are examples of passive constructions with ‘ser’ and ‘estar’:
‘Not only is one paid’ / ‘uno...es recompensado’
‘the picture then hangs’ / ‘el cuadro es expuesto’
Unusually, the second example shows a Spanish passive when the English sentence uses an intransitive verb. Butt and Benjamin conclude that ‘the advance of this Anglicised passive seems unstoppable, and it may eventually become a pervasive feature of Spanish.’ (200:382)

grammatical structure 3: the subjunctive
English has only a residual marked subjective, as in ‘if I were you/a rich man’, preferring to express possibility or uncertainly through the conditional- ‘he might come’, ‘I may do’ . The first sentence below, which comes from a quotation from a 17th century text contains a possible subjunctive: it is certainly an archaic construction. One would expect ‘so that you complete...’, or ‘make sure you complete...’. Spanish however requires a subjunctive.

‘Be tenacious so you complete the paintings you have started’ (P156)
‘Muéstrate tenaz de modo que acabes los cuadros que has empezado’ (p158)

‘when he has established himself’ (p159)
‘cuando se esté establecido’ (p157)

‘a suitable girl’, i.e. a girl who would be suitable (p159)
‘una chica... que le convenga’ (p157)

Omission
There are few omissions in the Spanish text. However, it omits the phrase ‘head-and-shoulders so much, full-length more’, referring to how payment for group portraits was calculated, perhaps considering this detail irrelevant. ‘Then there is this tulip business’ (p160) is also omitted. The same strictures against omission apply here, though the Spanish text is less cavalier than the French in this respect.

mistranslations
In the ‘Letter to Gerard Ter Borch’. the contents of the parcel include ‘beautiful paints’. The French TT has ‘jolies couleurs’, i.e. artist’s colours, but the Spanish text has ‘hermosos cuadros’, beautiful paintings, rather than paints. It is unlikely that a father would be sending finished paintings to his artist son. Another mistranslation is that of ‘mercado al aire libre’, open-air market, for ‘open market, a mistake shared by the French TT

summary
To summarize, the main divergences between the English and Spanish texts lie in discourse cohesion and in those grammatical structures where English and Spanish differ. Compared to the French TT, the Spanish TT is markedly closer to the ST, though like the French, it displays a preference for combining very short phrases and sentences into longer units.

readability analysis
I carried out a grammatical and readability analysis of all three versions using the readability software package provided by Word for Windows, 2000 edition, in order to supplement the previous qualitative analysis by statistical data. The readability check provides measures of the average complexity of the sentence structure, average sentence length – a long sentence is defined as over 35 words- and a readability score. Unfortunately the data are not directly comparable. The English readability indices and grammatical parsing tools are different from the Spanish and French ones and analyse fewer features, while the Spanish and French categories do not include a reading grade level.

Table 2 below shows the scores for average sentence length, complexity and readability for the three texts and the English quotation, which was analysed separately because of its different style and longer sentences.

TABLE 2

FEATURE SPANISH FRENCH ENGLISH LETTER

words 1409 1488 1377 240
sentences 124 127 122 12
paragraphs 37 41 41 1
short sentences 62 50 n/a n/a
long sentences 3 3 n/a n/a
simple sentences 57 18 n/a n/a
long words 281 190 n/a n/a
words per sentence 11.36 11.71 10.6 17.7
reading grade n/a n/a 5.0 7.4
Flesch reading ease 34 58 76.9 69.7
% passive sentences 2 0 4 8
sentence complexity score 11 20 n/a n/a
word complexity score 46 22 n/a n/a


text and sentence length
The intuition that the Spanish version uses longer sentences is confirmed: 11.36 words in Spanish compared to 10.6 in English. However, the French version also uses longer sentences than the English, and in fact has the longest average sentence length. The Spanish text is slightly longer than the English, although not as long as the French version. Romance languages tend to be more prolix than English so a text translated from Spanish, French, Italian or Portuguese into English will normally have fewer words than the original (Mott 1993:53).

reading grade score
The reading grade is based on average reading comprehension of pupils in the American high school system. The ‘average’ text (whatever that is) should aim at approximately reading grade 8, or pupils aged 12, according to the Word guide to readability scores. The English text is grade level 5, that is, comprehensible to nine-year-olds, suggesting that is a rather undemanding text.

reading ease score
The English TT scores 76.9, slightly easier than the average score of 60-70. On the Flesch scale, rather counter-intuitively, the higher the score the easier the text. However, both translations have lower scores: the French text is assessed as being ‘assez difficile’- quite difficult, while the Spanish score of 34 earns it the description ‘difícil’. This confirms my intuition that the Spanish text is less easy to read than the English original, even taking into account its slightly greater difficulty for a native English speaking reader, but I did not consider the French text difficult.

sentence and vocabulary complexity
The lower the score out of 100 the easier the text. The Spanish score indicates a very simple text, which seems odd given its low readability score and relatively high word complexity score. The Spanish vocabulary is judged to be of average difficulty, and the French less so. The most complex vocabulary scores 100. There is no comparable score of complexity for the English vocabulary . There are also fewer passive sentences in the Spanish version, in line with the lower prevalence of passive locutions in Spanish mentioned above. The passive is clumsy in French and the translator avoids it here, preferring an active and racy style.

quotation scores
The long passage translated from the 17th century Dutch text introduced a distorting element into the comparison, which was why it was analysed separately. On all these measures the quotation scored differently from the text as a whole. Not surprisingly, given the formality and archaism of the language, the sentence length was greater, the reading ease was lower, and the grade level needed for adequate comprehension higher. The reading ease score is based partly on the percentage of passive sentences, on the grounds that these take more linguistic processing, so it is not surprising to find that in this extract a high percentage of passive sentences accompanies a lower reading ease score.

Summary results
Statistics can back up or give the lie to subjective impressions. Gratifyingly, most of these results confirmed my own initial impressions of the texts. Both the Spanish and French sentences were longer than the English. There was a clear decrease in reading ease from the English to the French and then the Spanish version, and a similar increase in complexity of vocabulary. However, I was surprised that the Spanish text scored lower than the French on the measure of sentence complexity, although longer sentences are not necessarily more complex. Although crude, this quantitative analysis is a useful adjunct to a qualitative approach.

conclusion
A comparison of the novels suggests that their publishers adopted different marketing strategies which are reflected in the selection of translators and ultimately in the linguistic choices they make. The French translation is freer and the style more popular than the Spanish, which consistently uses longer sentences and more formal language. However, these results are only preliminary. Ideally, interviews with the people involved would have established their perceptions of the texts and the reasons for their editorial and marketing decisions. Similarly, it is impossible to establish how the target audience perceived the text without carrying out major audience research. More, and more systematic, textual analysis could have been carried out: some of the analysis here is crude and undeveloped. However I have at least sketched the outlines of a more ambitious research proposal, and have carried out some of the preliminary work for a more extensive study.


BIBLIOGRAPHY

Bassnett, Susan and Lefevere, André (1998) Constructing Cultures: Essays on Literary Translation. Clevedon: Multilingual Matters.

Boase-Beir, J & Holman, M (1998) The Practices of Literary Translation: constraints and creativity Manchester, UK: St Jerome Publishing

Fernández, Victór ‘Cataluña, una invitación editorial’ in La Razón Digit@l 5/06/05
http://www.larazon.es/publicidad/noticias/noti_cul57573.htm

Fuller, F (1984) The Translator’s Handbook Gerrard’s Cross, UK: Colin Smythe

Guix, J G L, & Wilkinson, J M (1997) Manual de traducción: Inglés-Castellano Barcelona: Editorial Gedisa

Hervey, S & Higgins, I (1992) Thinking Translation: A course in translation method: French to English London and New York: Routledge

Hervey, S, Higgins, I, & Hayward, L (1995) Thinking Spanish Translation: A course in translation method: Spanish to English London and New York: Routledge

Jones, Michèle, (1997) The Beginning Translator’s Handbook Lanham, Maryland, University Press of America

Iriarte, Eduardo (2002) Similacros de vida Barcelona, Laia Libros SL
http://www.barcelonareview.com/48/s_eig.htm

Landers, C E (2001) Literary Translation: A practical guide Cleveland, UK: Multilingual Matters Ltd.

McNamara, Katherine ‘Fantastic Design, with Nooses’, Endnotes, Archipelago 2.1
http://www.archipelago.org/vol2-1/endnotes.htm

Montgomery, Martin (1986) An Introduction to Language and Society London and New York, Methuen

Mott, Brian (1993) A Course in Semantics and Translation for Spanish Learners of English Barcelona: PPU

Nida, E A, and Taber, C R (1969) The Theory and Practice of Translation Leiden: E J Brill

Penny, Ralph (1991) A History of the Spanish Language Cambridge UK, Cambridge University Press

Smith, David (2005) ‘Women are still a closed book to men’, The Observer, May 29.

Stark, S (1999) “Behind Inverted Commas” Translation and Anglo-German cultural relations in the nineteenth century. Cleveland, UK: Multilingual Matters Ltd.



PAINTINGS

Antoine Berjon (1745-1843) Etude de Tulipe, Musée d’Orsay Paris

Parmegianino Portrait of a lady known as Antea, Naples 1535 Capodimonte Museum Italy
http://www.antea.com/antea/antea-parmigianino.htm
http://www.sorrento.cc/popup_place.php?fotoID=134

Johannes Vermeer Woman in Blue reading a Letter, 1662-5 Rijksmuseum, Amsterdam
http://www.aristos.org/aris-04/loveltrsim.htm