Article body

1. Preliminary Considerations

“Proper names are never translated” seems to be a rule deeply rooted in many people’s minds. Yet looking at translated texts we find that translators do all sorts of things with proper names: non-translation (en[1]. Ada > de., es., fr., it. Ada), non-translation that leads to a different pronunciation in the target language (en. Alice > de., fr. Alice [A’li:s], it. Alice [a’litche]), transcription or transliteration from non-Latin alphabets (es. Chaikovski vs. de. Tschaikowsky or Čaikowskij), morphological adaptation to the target language (en. Alice > es. Alicia), cultural adaptation (en. Alice > fi. Liisa), substitution (en. Ada > br. Marina, en. Bill > de. Egon) and so on. It is interesting to note, moreover, that translators do not always use the same techniques with all the proper names of a particular text they are translating.

Translations of fiction and of non-fiction seem to differ only in that there are no substitutions in the latter, unless we consider the “translation” es. Carlos I (of Spain) > de. Karl V. (of Germany) as a cultural substitution. All the other procedures are found not only in fiction, but also in non-fictional texts, where proper names refer to real-life historical persons: es. el rey Juan Carlos > de. König Juan Carlos, de. Johann Wolfgang von Goethe > es. Juan Wolfgango de Goethe, en. Prince Charles > de. Prinz Charles, es. el príncipe Carlos, en. Queen Elizabeth II. > de. Königin Elisabeth II., es. la reina Isabel II, but, illogically, es. Isabel I la Católica > de. Isabella I., die Katholische, and not *Elisabeth I., die Katholische. It is obvious that proper names are indeed translated, if we regard “translation” as a process of linguistic and/or cultural transfer.

In fictional texts, like novels or children’s books, proper names do not refer to real, existing people in a factual way. They may, however, refer to real persons indirectly, like in Alice in Wonderland. But still, the Alice of the book is a fictional character, and no reader would expect her to be a true reproduction of the “real” Alice Liddell for whom Lewis Carroll wrote the story.

To find a name for their fictional characters, authors can draw on the whole repertoire of names existing in their culture, and they can invent new, fantastic, absurd or descriptive names for the characters they create. We may safely assume, therefore, that there is no name in fiction without some kind of auctorial intention behind it, although, of course, this intention may be more obvious to the readers in one case than in another.

In the following paper, I would like to analyse the forms and functions of proper names in Lewis Carroll’s Alice in Wonderland and the way they have been translated into German, Spanish, French, Italian, and Brazilian Portuguese.

2. Forms and Functions of Proper Names

Unlike generic nouns, proper names are mono-referential, but they are by no means mono-functional. Their main function is to identify an individual referent. It has often been claimed that proper names lack descriptive meaning:

An ordinary personal name is, roughly, a word, used referringly, of which the use is not dictated by any descriptive meaning the word may have.

Strawson 1971: 23

In the real world, proper names may be non-descriptive, but they are obviously not non-informative: If we are familiar with the culture in question, a proper name can tell us whether the referent is a female or male person (Alice – Bill), maybe even about their age (some people name their new-born child after a pop star or a character of a film that happens to be en vogue) or their geographical origin within the same language community (e.g., surnames like McPherson or O’Connor, a first name like Pat) or from another country, a pet (there are “typical” names for dogs, cats, horses, canaries, etc., like Pussy or Fury), a place (Mount Everest), etc. Such indicators may lead us astray in real life, but they can be assumed to be intentional in fiction.

Titles and forms of address can also be problematic in translation. The protagonist of the Spanish children’s book El muñeco de don Bepo, by Carmen Vázquez-Vigo, is don Bepo, a circus ventriloquist. In the German translation, he is called Herr Beppo. The Spanish honorific title don is always combined with a first name, whereas Herr can only be used with a surname. Since German circus performers “typically” have Italian stage names, it would have been more adequate to translate don Bepo by Don Beppo.

In certain cases (like in Mount Everest or Lake Placid), a generic noun indicating the referent forms part of the name. Unless it is an internationalism (like King’s College – cf. es. Colegio – fr. Collège – de. Kolleg), the reference may be incomprehensible to someone who does not know the language, which then causes a translation problem.

Apart from names typically denoting a particular kind of referent, like pet names, authors sometimes use names which explicitly describe the referent in question (“descriptive names”). If, in a Spanish novel, a protagonist is called Don Modesto or Doña Perfecta, the readers will understand the name as a description of the character. In the case of the White Rabbit or the blue Caterpillar in Alice in Wonderland, the author proceeds in the opposite direction, using capital letters in order to turn the descriptive denomination into a proper name, which is bound to cause a translation problem as soon as W. Rabbit appears on the nameplate at the white rabbit’s house (see below).

In some cultures, there is the convention that fictional proper names can serve as “culture markers,” i.e., they implicitly indicate to which culture the character belongs. In German literature, for example, if a woman called Joséphine appears in a story with a plot set in Germany, she will automatically be assumed to be French. On the contrary, in Spanish literature, proper names are more generally adapted to Spanish morphology. A doctor named don Federico appearing in a Spanish setting (in the novel La Gaviota by Fernán Caballero) could be Spanish or German or French, and if the author wants him to be recognized as a German, she has to make this explicit in the context. This is a literary convention that might have to be taken into account in the analysis and translation of personal proper names in fictional texts.

Geographical names often have specific forms in other languages (exonyms), which may differ not only in pronunciation (e.g., de. Berlin > en. Berlin) or spelling (en. Pennsylvania > es. Pensilvania), but also with respect to morphology (es. Andalucía > de. Andalusien, en. Andalusia; es. La Habana > en., de. Havanna) or seem to be different lexical entities, as in de. München > es., en. Munich, it. Monaco [di Baviera]. Some are literal translations, like de. Niedersachsen > en. Lower Saxony, es. Baja Sajonia, and others go back to ancient Latin forms, like de. Aachen > es. Aquisgrán, de. Köln > en. Cologne, es. Colonia.

3. Some Translation Problems Connected with Proper Names

In spite of the “translation rule” quoted above, there are no rules for the translation of proper names. In non-fictional texts, it seems to be a convention to use the target-culture exonym of a source-culture name, if there is one, but if a translator prefers to use the source-culture form, nobody will mind as long as it is clear what place the name refers to. Perhaps the audience will think that the translator is showing off her knowledge too much. Wherever the function of the proper name is limited to identifying an individual referent, the main criterion for translation will be to make this identifying function work for the target audience.

In fiction, things are not quite as simple as that. We have assumed that in fictional texts there is no name that has no informative function at all, however subtle it may be. If this information is explicit, as in a descriptive name, it can be translated – although a translation may interfere with the function of culture marker. If the information is implicit, however, or if the marker function has priority over the informative function of the proper name, this aspect will be lost in the translation, unless the translator decides to compensate for the loss by providing the information in the context.

Of course, there are proper names that exist in the same form both in the source and the target culture. But this causes other problems: The character changes “nationality” just because the name is pronounced in a different way. An English Richard thus turns into a German Richard, and a French Robert into an English Robert – which may interfere with the homogeneity of the setting if some names are “bicultural” and others are not. For example: In a little comic strip I translated with my students in the Spanish-German translation class, the two characters, brothers, are called Miguelito and Hugo (cf. Nord 2001: 58ff.). If we leave the names as they are, Miguelito will be clearly recognizable as a Spanish boy in the translation, whereas Hugo may be identified as a German. In order to avoid the impression that this is a bicultural setting, the translator would have to either substitute Miguelito by a clearly German name or replace Hugo by a typical Spanish name, depending on whether the text is intended to appeal to the audience as “exotic” or “familiar.”

This is a very common problem in the translation of children’s books, especially if there is a pedagogical message underlying the plot. A story set in the receiver’s own cultural world allows for identification, whereas a story set in a strange, possibly exotic world may induce the reader to stay “at a distance.” This can be clearly shown by an analysis of the Brazilian translation of Alice in Wonderland, where all the culture markers, including the proper names, are consistently adapted to the target culture (cf. Nord 1994).

Different name conventions in literature can also lead to translation problems. If, as stated above, names are adapted to the Spanish language and culture in Spanish literature and, on the contrary, serve as culture-markers in German literature, the translator should take this into account. In a Spanish play set in France (Max Aub, El puerto), the characters are called Claudio, señora Bernard, Josefina, Andrés, Marcela, Julio and Matilde (all of them French) and Estanislao Garin Bolchenko (Polish) in the original. In the German translation (Max Aub, Der Hafen), all the French characters have French names (Claude, Madame Bernard, Joséphine, André, Marcelle, Jules, Mathilde) and the Pole is called Stanislas Garin Bolschenko (in German transcription).

In the following section, we will apply these considerations to the translation of proper names in Alice in Wonderland. We will first look at the forms and functions of the proper names appearing in the book and then discuss the translation procedures found in the various translations which constitute our corpus and their possible effects for the reception of the book. Thus, our approach is both functional and descriptive.

4. The translation of proper names in Alice in Wonderland

As is well known, Lewis Carroll, alias Charles Dodgson, wrote Alice in Wonderland in 1862 for his little friend Alice Liddell, 10, and her two sisters Lorina Charlotte, 13, and Edith Liddell, 8 years old. Certain characters or figures of the story are explicitly or implicitly taken from the girls’ real situation, and we may assume that this must have amused them very much. Others are pure fiction.

In Alice in Wonderland, we find the following forms of proper names:

  1. names explicitly referring to the real world of author and original addressees,

  2. names implicitly alluding to the real world of author and original addressees by means of wordplay,

  3. names referring to fictitious characters.

4.1. Proper names referring to the real world of author and original addressees

The “real” world is England in the second half of the 19th century, including historical facts presupposed to be known by the first addressees of the story, namely Alice Liddell and her sisters. Apart from the first addressees A1, as soon as it is published, the book is directed at a broader audience A2, namely children and/or adults, probably sharing the same real-world knowledge. The names explicitly referring to the real world of author and A1 can be assumed to fulfil their identifying function also for A2.

In the story or in the poems quoted in the story, we find several names of persons belonging to the “real world” of the author and the audience A1 (Alice, her nurse Mary Ann, her school mates Ada and Mabel), of places (New Zealand, Australia, London, Rome, Paris, the Nile), and historical personalities (Shakespeare, Edwin, the Earl of Mercia, Morcar, the Earl of Northumbria, Stigand, the Archbishop of Canterbury, Edgar Atheling, William the Conqueror). These names are primarily identifying, and this function relies on the receiver’s previous knowledge. For the addressees A1, it is guaranteed, since the historical allusions are indirect quotations or travesties of schoolbook texts. This is probably true also of the reference to “Shakespeare in the pictures of him,” which is used to illustrate how the Dodo is sitting “for a long time with one finger pressed upon its forehead.” The name of the game mentioned in the third chapter, the Caucus race, too, is probably formed using a buzzword of the time – it sounds somehow funny, but it is probably incomprehensible to children.

For the audience A2, the geographical and historical references will also be clear. However, Alice, Ada and Mabel and Mary Ann will be fictitious characters for them. This would not cause a comprehension problem, since the characters are introduced in contexts where their identity is made clear. Nevertheless, the appellative function of being amused when the receivers read about themselves and their own situation, which characterizes the reception of A1, does not work for A2.

For modern readers of the translations, whom we might call audience A3, the situation is more or less the same as for A2, as far as the references to living persons and to places is concerned. They may not know the pictures of Shakespeare, but since the position is described explicitly, there will be no comprehension problem, but probably no appellative recognition of something known either.

-> See the list of tables

The examples show that different techniques are used to render the names of the persons alive at the time of text production. It may not be surprising that the name of the protagonist, Alice, is left as it is by almost all translators. The Spanish translator follows the convention of adapting the form of the name to Spanish morphology. German, French, Italian and Brazilian readers will pronounce the name according to their respective phonologies. The Brazilian translator consistently adapted all the names, so did Enzensberger (DE-ENZ), with the exception of Ada and Mabel. Ada is not a German name, nor is Mabel, which sounds rather weird in German pronunciation. In an adaptation, we might also expect “typical” names for pets (like Mimi in BR, which would be Mieze for a German cat) or housemaids.

As we have mentioned before, for A3, like for A2, the characters are fictitious anyway. Therefore, an adaptation of the names allows for easier pronunciation and does not interfere with the identifying function. However, target-culture proper names mark the setting as belonging to the target addressee’s own real world, and the translator should make sure to keep up this strategy throughout the story, in order not to produce culturally incoherent scenes.

The Brazilian translator is very consistent in this respect and omits all references to historical figures. It was a ship-wreck, and not William the Conqueror that had brought the Mouse over, and the “dry story” about the unpronounceable Earls is turned into a “very dull old story,” which is not reproduced in detail. The German translator Teutsch (DE-TEU) is not quite as radical, but she composes a new schoolbook text with many dates and references to English history, using German exonyms for the names of persons. Enzensberger (DE-ENZ) uses a cultural substitution, referring to German historical figures. This is consistent with the general procedure he uses translating proper names, but it is not always in line with his translation of other cultural references, as I have tried to show in Nord 1994.

The names of the historical persons (apart from William the Conqueror, for whom the other languages have exonyms) are rather hard to pronounce for anybody who is not familiar with the English language. If the readers cannot be expected anyway to have heard these names before, it would not undermine the functionality of the translation to replace them by others, since they are mentioned in a paragraph where the Mouse’s recites “the driest thing she knows.” So the focus is on the “dryness” of the citation and not on the historical facts related in it. Yet the reader would probably expect the facts to be historically true or, at least, consistent. The substitution of Earl by es. duque or fr. seigneur is not necessary from the point of view of addressee orientation or cultural transfer.

-> See the list of tables

As far as the geographical references are concerned, all translators use exonyms either phonologically or morphologically adapted to the target languages. The Brazilian translator has also adapted the reference to the Antipodes, thus avoiding pragmatic incoherence. This shows clearly that she intended to “transfer” the story to Brazil.

-> See the list of tables

The reference to Shakespeare is adapted by Enzensberger (DE-ENZ) and generalized by Bublitz (DE-BUB) and Remané (DE-REM). The problem with adaptation here is that the pictures of Goethe a reader may recall do not show him in this particular posture. Therefore, the appellative function (i.e., to make the receiver remember something known) works neither with Goethe, nor with famous poets (DE-BUB) or grand philosophers (DE-REM). A translation like grand philosophers could evoke Rodin’s Penseur, but then the translator would have to substitute pictures by images. This is why Teutsch (DE-TEU) and the Brazilian translator omit the reference altogether. Comparisons are usually intended to clarify the verbal description and not to mystify it. Some researchers assume that the comparison was a private joke between the author and his first addressees (the Liddell sisters), because it is difficult to find a picture showing Shakespeare precisely in this position. Nevertheless, a translator may want to make the target text “work” for its addressees.

-> See the list of tables

In the translations of Caucus race, some translators try more or less faithfully to reproduce the meaning of the word (DE-BUB, ES, IT), others use a political term probably incomprehensible to children (DE-ENZ, DE-REM), the Brazilian translator plainly describes the “crazy race” without giving it a proper name, and Teutsch (DE-TEU) gives a pseudo-description with nice, incomprehensible latinisms. The French translator’s decision simply to use the English word seems to work just as well, since the word sounds beautiful in French pronunciation and means nothing. The example shows the importance of translating functions instead of words.

4.2. Names implicitly alluding to the real world of author and addressees by means of wordplay

An important element of the real world of author and addressees is the English language. Apart from certain proper names in Alice in Wonderland that allude to real persons in an indirect way, we find names alluding to idiomatic expressions. In both cases, the allusion will have produced a particularly appellative function for the audience A1 when detecting the hidden reference. The implicit allusions to real persons is bound to be lost for both A2 and A3, whereas the indirect reference to idioms and set phrases will probably work at least for A2. For A3, they can only fulfil an analogous function in a target-oriented translation.

The Spanish translator opts for a meta-text and adds a long list of annotations to his translation. He informs the reader that the Dodo, apart from its reference to the idiomatic expression “as dead as a Dodo,” is an allusion to Lewis Carroll’s slightly stuttering way of pronouncing his own name: Do-Do-Dodgson. The Duck refers to his friend, the reverend Duckworth, and the Lory and the Eaglet to Alice’s sisters Lorina Charlotte and Edith, respectively. The three little sisters living in the treacle well, Elsie, Lacie and Tillie, also represent the Liddell sisters: Elsie stands for Lorina Charlotte (“L.C.”!), Tillie for Edith Mathilda, and Lacie is an anagram of Alice.

Thus, independently of whether the translator adds the annotations, these names turn into purely identifying or descriptive names for the audiences A2 and A3. The Duck, the Lory, the Eaglet, the Gryphon and the Dodo are represented in the illustrations by Tenniel, so the translators are bound to establish coherence between verbal and nonverbal text. The Duck and the Gryphon remain what they are in all translations of the corpus. The Lory is substituted by a Brachvogel, a kind of snipe, by Enzensberger (DE-ENZ). This does not seem very plausible since German readers can be more easily expected to know a Lory than a Brachvogel, which, by the way, cannot be detected in the illustration reproduced in the Enzensberger edition. Enzensberger is the only one to replace the Eaglet by a Weih, a fantasy creature, whose name he needs for an adapted wordplay in the context.

-> See the list of tables

The Dodo is an interesting case since it is known as an extinct bird in the target cultures too. Dronte (DE-BUB) is another name for the same bird, and its encyclopaedic description corresponds to the creative translation given by Teutsch (DE-TEU): a gigantic bird like a mixture of pigeon and turkey. The Brazilian translator adapts to papagaio, a word which in Brazilian Portuguese is less frequent than arara, her translation for the Lory. Although it is not clear why Enzensberger and Remané replaced the Dodo by a Marabu (DE-ENZ) or a Pelikan (DE-REM), the substitution does not cause any inconveniences either. In the Remané edition, the illustration indeed shows a Pelican.

Like the Dodo, the Hatter and the March Hare allude to idiomatic expressions “as mad as a Hatter” and “as mad as a March Hare,” which have no direct equivalents in the other languages. Remané (DE-REM) tries to preserve the connotation by translating the March Hare by Schnapphase, explaining the name in the context: because he was übergeschnappt (“mad”). The explanation is not really convincing. According to German morphology rules, a Schnapphase is understood as a hare that snatches something away. Whereas the idea of hares which are mad (for a mate) in March may be evoked in some readers, the idea of a hatter being particularly mad will probably not come to the mind of the audiences. But since the Hatter is depicted as a rather weird figure both in the context and in the illustrations, this may not be a comprehension problem.

-> See the list of tables

The Cheshire Cat is an allusion to a particular brand of Cheshire cheese which had a picture of a grinning cat on the package and seems to be the origin of the idiomatic expression “to grin like a Cheshire Cat.” This connotation does not work in the other cultures, and a substitution (e.g., by a cow in French, cf. la vache qui rit, or a honey-cake horse in German, cf. grinsen wie ein Honigkuchenpferd) would be out of place because of the illustrations which show a grinning cat. Therefore, the allusions to some kind of cheese (DE-BUB, DE-ENZ, FR) are as pointless as the literal translation (es, br, it). This is, by the way, the only case where the Brazilian translator deviates from her adaptive strategy.

The Dormouse may evoke connotations of sleepiness in English even if the reader does not know exactly what kind of animal it is. The name sounds a bit like dorm(itory), which seems to be a rather obvious association, at least for teenage children in England. In German, there are several names for the animal in question, of which Schlafmaus (DE-BUB) and, particularly, Siebenschläfer (DE-TEU) seem the most appropriate in this context because the names refer to schlafen (“to sleep”). The name Haselmaus, although zoologically correct, seems as unmotivated as the respective equivalents in Spanish, Portuguese, French and Italian.

The Mock Turtle (which introduces itself by saying that it used to be a Real Turtle once) is a particular challenge for any translator. In Germany, where Mockturtle(suppe) seems to have disappeared from the shelves of supermarkets and delicatessen shops, the referent itself will appear extremely strange to the readers, especially to children. Teutsch (DE-TEU) therefore creates an Oxtail Turtle, which preserves the reference to soup (important in the context) and makes recognition easier. A False (Soup) Turtle (DE-ENZ, DE-REM, es, br, it) is consistent with the illustrations, which show a turtle with a calf’s head, but not quite coherent for readers who do not know that mock-turtle soup is made of veal broth. Therefore, the French translator created an ingenious compound referring precisely to this aspect.

-> See the list of tables

Bill is a lizard. His name is used in a pun in the very title of the chapter: THE RABBIT SENDS IN A LITTLE BILL. The majority of the translators skip this pun, except Bublitz (DE-BUB), who plays with Bill and billig (“cheap”), and the Spanish translator, who uses the diminutive of Pepe, Pepito, which means “a little lump of meat.” Although it is not bits of meat but of cake that are sent in by the Rabbit, the pun will probably work for Spanish readers. Enzensberger’s solution (DE-ENZ) is in line with his general adapting strategy, but the name Egon seems completely unmotivated. This is a frequent problem with substitutions; once you have started, it is difficult to tell where to stop.

-> See the list of tables

Type names, i.e., typical names for certain classes of objects, also refer to the real world of author and addressee because they are culture-specific. The only name of this kind in Alice in Wonderland is Fury. Since this name appears in the Mouse’s tale where she justifies why she hates cats and dogs, a literal translation rendering the proper name by a generic noun (as in DE-ENZ, ES and IT) destroys the coherence between the context and the tale. Bublitz (DE-BUB) uses a typical German name for a big and dangerous dog. For modern German children, Fury would refer to a famous TV horse.

There are two more instances referring to “typical” proper names in the book. One is the address of the Christmas parcel Alice thinks of sending to her foot when she has grown very tall, and the other refers to the White Rabbit’s doorplate. In both cases, it is the culture-specific form of the name which seems to cause translation problems.

In English, the author uses the conventional form of address with the addressee’s name and residence. Hearthrug, near the Fender imitates the name of a little village in the neighbourhood of a bigger town. In spite of the fact that Esq. used to be no more than a politeness marker which does not imply that the addressee belongs to aristocracy, almost all translators use some kind of very formal treatment. The Brazilian translator skips over the problem by avoiding the address form altogether. Teutsch, again, is the one who adapts the forms of both the name and the address in the most consistent way. She uses the abbreviation “z.Z.” (= “zur Zeit”), which is only found in letter heads or on envelopes to mark a temporary address.

-> See the list of tables

In an illustration, we see a house, and beside the door a doorplate with W. Rabbit on it. In English, this combination of initial and surname seems quite natural, because White does not sound very different from Walt or William. In German, the adjective weiß (“white”) must be declined, and Weißes (DE-BUB, DE-ENZ, DE-REM) does not at all sound like a first name. Changing Kaninchen into Kanin by omitting the diminutive suffix (DE-TEU) is not very logical either. On the other hand, Weiß is a usual surname in Germany, so Weiß, K. might be an acceptable solution since people often put their first name after the surname on doorplates. In the Spanish translation, B. Conejo is not coherent with Conejo Blanco. Spaniards use two surnames, the father’s name and the mother’s name, so the whole expression Conejo Blanco would be a perfect surname, perhaps together with any initial for a first name, e.g., F. Conejo Blanco. This is the solution the Brazilian translator has chosen, and she is lucky because Coelho Branco is, indeed, a usual surname.

4.3. Names referring to fictitious characters

It is a specific characteristic of Alice in Wonderland that, with very few exceptions (like Alice, Pat, Bill), the fictitious characters have no names in the conventional sense of the word. Characters, mostly animals or fantasy creatures, are usually introduced by a description which is afterwards used as a proper name just by writing it with a capital letter. For example, at the end of chapter IV we read: “She stretched herself up on tiptoe, and peeped over the edge of the mushroom, and her eyes immediately met those of a large blue caterpillar, that was sitting on the top with its arms folded, quietly smoking a long hookah, and taking not the smallest notice of her or of anything else.” The next chapter begins as follows: “The Caterpillar and Alice looked at each other for some time in silence…” (emphasis C.N.).

White Rabbit is one of the numerous generic nouns turned into a proper name: The White Rabbit, the Mouse, the Duchess, the Gryphon. Translating into Romance languages it is easy just to follow the author’s model capitalizing the generic nouns. In German, however, all nouns are written with a capital first letter, therefore, capitalization cannot be used as a means to mark them as proper names. The only way out of the dilemma would have been to use the nouns without the definite article, but this procedure cannot be found in the German translations of our corpus.

Apart from this group of names we find personifications of playing cards: Five, Two or Three, together with the King, the Queen and the Knaves. There are no translation problems here, and the illustrations support the text, especially for readers in cultures where other kinds of playing cards are used.

-> See the list of tables

The last proper name I would like to mention is Pat. Pat appears in the same scene with Bill, the Lizard, and it does not become clear what kind of creature he is, perhaps one of the guinea pigs that are mentioned in the context. By his way of speaking, Pat is characterized as an uneducated person, perhaps a farmer. Judging by his name, he could be an Irishman. The author comments on his pronunciation of the word arm: He pronounced it “arrum.” This could also be a North England or Scottish accent.

The annotations do not say anything about Pat’s identity, but it seems to me that there would be some kind of model for Pat in the real world of author and original addressees. All translators try to give the pronunciation some kind of dialect touch. In German, Aam or Ahm (DE-BUB, DE-ENZ, DE-TEU) points to someone from North Germany, the regional connotation of Arrem (DE-REM) is not obvious to me. In Spanish, the pronunciation brasso characterizes an Andalusian or Latin American speaker. The French brraa could indicate a person from Corsica. However, according to my sources, bracco for braccio is not possible in any Italian dialect. Anyway, in none of the translations do we find a correspondence between the name and the way of speaking.

5. Conclusions

We have seen that there are various strategies for dealing with proper names in translation. It would be interesting to see whether a particular strategy correlates with addressee-orientation. Since experts are still debating whether the original Alice in Wonderland is a book for children or for adults, we have to look at the form of publication to find out whether a translation is directed at children or adults. As far as the German translations are concerned, BUB, REM and TEU are definitely books for children, they have been published in children’s books series. REM and BUB have new illustrations, the ones in BUB following Tenniel’s footsteps, the ones in REM showing a rather modern Alice in mini-skirt and ponytail. TEU reproduces the original illustrations by John Tenniel. ENZ also uses the Tenniel pictures, but the publication of the book in the prestigious Insel Verlag points to an adult audience.

-> See the list of tables

A quantitative analysis of translation procedures shows the following results. The reproduction of source-language names without any changes in the form (repro), although usually with an adaptation of the pronunciation to target-language norms, is the most frequently used technique in DE-BUB (38%), DE-REM and FR (both 48%) as well as IT (55%). The use of adaptation of source-cultural names to target-language morphology (adapt) and of exonyms (exonym) is most frequent in the Spanish translation (ES, together 58.6%), which confirms the assumption that the adaptation of proper names is conventional in Spanish literature. The substitution of source-culture names by target culture names (subst) is the favourite procedure in DE-ENZ (44.8%). Together with the proper names rendered as generic nouns, which also has an adapting effect, substitutions sum up to 65.5% in DE-ENZ. As we have seen before, the Brazilian translator has left out a large number (38%) of the proper names and substituted another 31%, which makes her translation the most target-oriented of our corpus. Actually, on the front page of the book, the text is characterized as an “adaptação.” The only translator who has no real “favourite” technique is DE-TEU, but adaptations, substitutions and translation by generic nouns represent 62% of her procedures. Therefore, her translation also has a strong target-orientation.

Neutralizations (neutr) are cases where a culture-specific name is rendered by a culture-unspecific or “transcultural” reference. They mark neither the source nor the target culture, but it can be empirically proved that readers tend to “domesticate” such references. Calques are literal target-language translations of source-language names. As such, they preserve their semantic strangeness but lose their foreign look. Both neutralizations and calques are not found very often in the corpus.

-> See the list of tables

Coming back to the question whether there is a correlation between the number of adaptive procedures and addressee orientation we have to state that the analysis of the corpus does not confirm the assumption that adaptive strategies would generally be more frequent in children’s books than in translations for adults. Both DE-REM and FR use mainly reproductive techniques although they are translating for children, and DE-ENZ uses adaptive strategies although he is translating for adults.

In several cases, adaptation was impossible because the illustrations showed the source-text referents (cf. Cheshire Cat). The W. Rabbit example shows that the pictures sometimes can cause additional translation problems where the text itself would not be difficult to translate. If the translator had an influence on the pictures, certain translation problems would be easier to solve.

The last aspect I would like to mention is annotation. Notes are meta-texts, and meta-texts are usually referential. In Alice in Wonderland, the notes inform the reader about the appellative function(s) of the original. The problem with the explanation of puns or jokes is that it kills them: a joke that has to be explained is as dead as a Dodo. Moreover, the reader receives two texts, i.e., a text where the names seem to be purely identifying or referential, and another text that explains why these names are not purely referential. This procedure will necessarily change the whole communicative effect of a text. To my view, the decision for, or against, annotations must be guided by addressee-orientation. For an adult readership, it may be interesting to read the two texts, either “side by side” or one after the other. For children, one text will probably be sufficient. Consequently, in our corpus, annotations and translators’ commentaries are only found in the translations for adults (DE-ENZ and ES).