Is translation an art or a craft? If we consider the creative activity that takes place when understanding a text, reshaping it in another language, and thus creating a new text, then translation is an art. Yet, if we think of the many strategies that the translator has to apply during his or her work, then translation is a craft, a very complex skill, which requires much more than simply the knowledge of two languages.

The key asset of the translator’s competence is the so called translation competence (Klaudy, 1997a). Having translation competence means that the translator is able to use two languages simultaneously (Klaudy, 1994). In order to be able to ‘transform’ the source text written in the source language into a new text written in the target language, the translator has to acquire certain transformational strategies with the help of which he will fulfil the task. These transformational strategies appear at the word level, and they can be found at higher levels: the phrase, clause,  sentence and discourse levels. 

Transformations can be learnt, because all the languages have a so called ‘translation behaviour’ in relation to the other languages, and this behaviour is different in each language (Klaudy 1997a, p. 147). All languages have certain features that come out only when they are translated into another language. If such translation behaviour of languages exists, then the translator will have to acquire these language-pair specific transformational strategies that can be referred to as the transformational routine of the translator.

The crucial role of transformational strategies in the translation process makes it important to analyse them. In the research conducted for my thesis, I intended to identify and investigate the different transformational strategies by analysing translations from English into Hungarian written by student translators and professionals. I investigated whether the translations show differences and similarities on the various levels of language (word, phrase and clause, sentence and discourse) between the transformational strategies used by student translators and professionals. However, the most interesting part of the study for the language teacher may be the exact differences and similarities that I have found on the various levels of language between the strategies used by student translators and professionals, so I am going to focus on that area.

Method

For the purpose of the research a group of professionals and a group of students were asked to translate a carefully selected text from English into Hungarian. The translations were analysed by using the method of parallel comparison. This means that the translations were analysed by examining a unit of translation for all the translations.

Previous studies in the field (Király, 1995; Krings, 1986) differed from the present one in their procedure of analysis, as they used the think-aloud protocol as their method of investigation instead of the parallel comparison used in this study. Consequently, they cannot be compared directly with the results of this study, only indirect comparisons and implications can be made.

The source text

The source text was an article from the Times Literary Supplement  entitled “How computers can damage your prose”, written by Edward Mendelson [(see Appendix The source text, A professional translation (first paragraph), A student translation (first paragraph)]. This is a stylistically sophisticated text, full of irony and metaphors, which might cause problems for the translator. Consequently, although only the first three paragraphs had to be translated, the translators were given the whole article and were asked to read it in its entirety in order to gain a deeper understanding of the author’s intention, opinion and language style. 

The text can be found in Angol fordítóiskola  (Bart, Klaudy, & Szöllõsy, 1996), which is a coursebook on teaching English translation. In the book, the text is followed by exercises which prepare the translator for the task by examining the key concepts and the problematic words and expressions. Then, a suggested translation can be found, followed by annotations regarding various translation problems, transformational and other strategies. These explanations are very useful from the point of view of the present analysis, as they served as the basis for detecting the problematic points in this particular text.

Participants

The text was translated into Hungarian by altogether sixteen translators, forming two main groups. The first group comprised six professional translators, five of whom were trained at the Training Centre for Translators and Interpreters at Eötvös Loránd University, Budapest in the past five years. They were all trained professionals who - except for one person - earned money by translating and interpreting on a regular basis.

The second group consisted of ten university students who studied English language and literature at Eötvös Loránd University, Budapest. They were either in their fourth or in their final year at the university. They did not have any previous training in translation, although some of them were interested in translation and sometimes did translations.

Procedure

The translations were not done on the spot, the translators took the source text home, and they were given two weeks to do the work. They were allowed to use dictionaries, but were asked to work on their own.

Analysis

In order to understand the procedure of text analysis, the commonly used transformational strategies need to be presented. In her framework, Klaudy (1997b) distinguishes various transformational strategies and classifies them from three different aspects. From one aspect, obligatory and optional transformational strategies  may be distinguished. Obligatory transformational strategies have to be used due to the differences in the lexis and the syntax of the languages. If these strategies are not applied, semantically incoherent or ungrammatical sentences will be formed. For example, when translating the English sentence ‘He went to school’, the translator has to apply grammatical change, because the complements being to the right of the predicate in English are to the left of the predicate in Hungarian. Thus, the correct translation is ‘Iskolába ment’, and not ‘Ment iskolába’. Optional transformational strategies, on the other hand, are carried out after the obligatory ones. Antonym translation (i.e. translating a word by its antonym) is for example an optional transformational strategy (e.g. ‘My uncle died’ - ‘A nagybátyám meghalt’ or ‘A nagybátyám nem él’).

According to the level of language on which the transformational strategies need to be applied, word, phrase and clause (word string), sentence and discourse level transformational strategies can be distinguished. A transformational strategy on the word level is, for example, the replacement of the source word for its target equivalent which is more general in meaning (e.g. ‘foot’ - ‘láb’). A strategy on the word string level is, for example, changing the order of words in an adjectival construction (e.g. ‘the girl in the red coat’ - ‘a piros kabátos lány’). A strategy on the sentence level is, for instance, turning a passive construction into an active one (e.g. ‘he was left alone’ - ‘egyedül hagyták’). Finally, a strategy on the discourse level can be the standardisation of the subject within a paragraph. 

Thirdly, depending on whether their domain is the lexis or the grammar of the language, transformational strategies can be classified into two further groups: lexical  and grammatical transformational strategies. These strategies are applied in both directions; however, due to the scope of the present study, the focus will be put on transformational strategies that the translator needs to make use of when translating from English into Hungarian. Some lexical transformational strategies are: concretisation, that is, narrowing the meaning (English ‘said’ - Hungarian ‘mondta/dörmögte/suttogta/ kiáltotta’ etc.); generalisation, that is, widening the meaning (‘foot/leg’ -  ‘láb’); contraction of meanings  (‘began to speak’ - ‘megszólalt’, ‘said softly’ - ‘suttogta’, etc.); disintegration of meaning (‘refused’ - ‘nem volt hajlandó’, etc.); antonym translation  (‘died’ - ‘nem él’; ‘was closed’ - ‘nem nyitott ki’); total transformation  (‘egg and spoon race’ - ‘zsákbanfutás’; ‘To let the grass grow under one’s feet’ - ‘Várja, hogy a sült galamb a szájába repüljön’).

 The grammatical transformational strategies have similar types: concretisation (Elment aludni.’ - ‘He went to sleep’); generalisation (‘He told her.’ - ‘Mondta neki.’); disintegration (‘He felt a desire to kiss her’ - ‘Vágyat érzett, hogy megcsókolja’; contraction (‘Láttam, ahogyan közeledik’ - ‘I saw him coming’); transpositions: obligatory (‘His parents went for a holiday’ - ‘Szülei vakációra mentek’) or optional (‘But he knew it this time’ - ‘Most azonban tudta’); grammatical change (‘his eyes are beautiful’ - ‘gyönyörû szeme van’ or ‘He is said to be lazy’ - ‘Azt mondják, hogy õ lusta’); omission and insertion: (‘he offered her coffee’ - ‘kávéval kínálta’).

As I have mentioned above, the method of analysis that I followed was parallel comparison. During the application of this method, the different translations of a particular word were written down in two columns. First, the solutions of the trained translators, and then the solutions of the student translators. The same method of parallel comparison was applied to each text, examining the translations for all the important and ambiguous words. An example for the procedure of parallel comparison can be seen in Table 1.

On the word, word string and sentence levels this procedure of parallel comparison worked well, on the discourse level, however, the procedure had to be slightly modified. On this level, the translations were analysed in their full length. An example for the procedure of the analysis on the discourse level can be found in Table 2 on the following page.

At  the discourse level, the different cohesive devices were analysed separately on the basis of Halliday and Hasan’s (1976) taxonomy. This taxonomy illustrates the different devices that make texts cohesive and coherent. Some grammatical cohesive devices from the text are the following: reference (‘..., but the act itself comes to seem more significant and exciting than any purpose it could possible achieve’); substitution (‘The craft of writing has little in common with the craft of winning a video game, but the computer encourages the illusion that the two are essentially identical’); ellipsis  (‘When you write at a computer or play a video game, ...’); conjunctions  (‘As you find the point on screen to which you intend to move a misplaced word, and then drop the word into place ...’). The most important lexical cohesive device from the point of view of my paper is collocations (e.g. ‘The computer flatters you with the illusion that ...’). 

During the procedure of parallel comparison, after each examined unit (a word, construction, sentence or cohesive device), the results were grouped according to the differences and similarities. After completing the parallel comparison, the results were also grouped according to the levels of language,  and within these levels, according to the type of transformational strategies they represent. 

Results and discussion

Numerous differences and similarities were found on all levels of language (word, phrase and clause, sentence and discourse level) in the use of the different grammatical and lexical transformational strategies between professionals and student translators. In the use of obligatory transformational strategies, since they are obligatory, there was no difference between the two groups of translators. In the use of optional transformational strategies, however, there are differences between the two groups. Consequently, I will focus more on the differences than on the similarities, though first, I will give a brief summary of my findings concerning similarities. Then I will discuss the differences found in the use of the grammatical transformational strategies. Finally, the lexical differences will be presented. 

Grammatical and lexical transformational strategies: similarities

Due to their obligatory nature, similarities were found in the following grammatical transformational strategies: change (from noun into verb), generalisation (dropping the personal pronoun and substituting it with a suffix), transposition of complements of the predicate from right to left, compensation for ellipsis by a suitable Hungarian suffix. 

As the lexis of a language is more flexible and freer than its syntactical rules, most of the lexical transformational strategies are not obligatory. Consequently, in the use of lexical transformational strategies no such evident similarities were found as in the use of grammatical ones. There were similarities, but these similarities were mostly due to the similar word choice of the translators, and not the homogenous use of some lexical transformational strategies. 

Grammatical transformational strategies: differences

On the word level, a remarkable difference was detected in the transformational strategy called grammatical change. The general tendency that the English plural forms are expressed in the singular in Hungarian evokes a grammatical change. The title of the source text contains a typical example for this: the plural form of the noun ‘computers’ should be changed into the singular ‘számítógép’ in Hungarian. This grammatical change was carried out by all the six professionals, however, fewer than half (four out of ten) of the student translators made the same change. 

Another difference was found in the change of the parts of speech. This strategy is important, because the languages are not the same in their tendency to use certain parts of speech. In the second half of the first paragraph, the following sentence can be found: ‘The writer’s task has never felt more challenging, ....’. In this sentence, the word ‘challenging’, an adjective, cannot be rendered with an adjective in Hungarian, since the Hungarian word ‘kihívó’ does not have the same meaning as the English word ‘challenging’. Consequently, another part of speech has to be found for this adjective in Hungarian. The professionals used either a noun, or a suitable adjective, but four out of the ten students wrote the incorrect ‘kihívó’. 

On the phrase and clause level, the correct translation of some of the English prepositions caused problems to the student translators. In the first paragraph, there is a clause ‘..., you can congratulate yourself on your marksmanship and control’. In Hungarian, the government of the verb ‘gratulál’ is ‘valamihez’. The professional translators either wrote ‘szakértelmedhez’, ‘uralmadhoz’, etc., or they made a grammatical disintegration and formed a clause: e.g. ‘büszkén gratulálhatunk magunknak, hogy milyen jól céloztunk’. Yet, four student translators used the grammatically incorrect ‘-ért’ government: e.g. ‘gratulálhatsz magadnak a szakértelmedért’.

Remaining on the phrase and clause level, the grammatical transformational strategy in the use of which considerable differences were found between the two groups of translators was the grammatical disintegration of the infinitival, gerund and prepositional constructions. As a general rule, it can be stated that the Hungarian language “prefers” to have more clauses in a sentence than the English. Therefore, grammatical disintegration is a highly recommended transformational strategy when translating from English into Hungarian. The professionals made use of grammatical disintegration, that is, formed two clauses from one more often when translating such constructions than student translators did. For example, in the translation of the sentence ‘Experts can at least take pleasure in winning the game’, almost all the professionals formed two clauses e.g. ‘..., ha megnyernek egy játékot’, but half of the students created one complex sentence e.g. ‘... a videojáték megnyerése’ or ‘a nyerés’.

The first two levels of language have already revealed some differences between the two groups; yet, the most noticeable differences can be found on the sentence level. Grammatical transposition, that is, the change in the order of clauses is a crucial area of this study. English and Hungarian sentences are structured differently, and the order of the clauses cannot always be maintained when translating from one language to another. The reason for this is that the logical flow of a sentence is often expressed differently in the two languages. Student translators, however, tended to maintain the order of the English sentence. For example, the sentence ‘A computer screen flatters you with the illusion (clause a) that you are labouring over style and content when (clause b), with the press of a key, you delete words or transpose sentences (clause c)’ was translated by eight out of the ten students by keeping the order of the English sentence (a, b, c). The professionals, however, changed the order of the clauses to either a, c, b or c, a, b, because the logical flow of the Hungarian text requires it (‘A számítógép képernyõje azt az érzést kelti, hogy amikor egy-egy gombnyomásra szavakat tüntetünk el vagy éppen mondatokat helyezünk át, a stílusunkat és a mondanivalónkat csiszoljuk.’). 

Another grammatical transformational strategy on the sentence level showing remarkable differences was grammatical change, and within that, the changing of the passive constructions  in English into active  ones in Hungarian. English is full of passive constructions, which sound alien in Hungarian. The results of the analysis revealed that in contrast to the professional translators, the student translators did not tend to change passive constructions into active ones. For example, the passive construction ‘... the actions you can perform are rigidly limited by the rules ...’ was changed into an active one by most of the professionals (‘szigorúan behatárolják’, ‘merev szabályok határozzák meg’), but seven of the ten students kept the original passive construction, creating phrases like ‘szigorúan be vannak határolva’.

Finally, notable differences were found on the discourse level as well. Ellipsis  is more common in English than it is in Hungarian. In English, personal pronouns, auxiliaries, prepositions, etc. can be omitted. In most of the cases, the translation of these ellipses into Hungarian does not cause a problem, because the Hungarian language automatically compensates for these ‘losses’ due to its morphological structure. There are some cases, however, when the elliptical word in English cannot be omitted from the Hungarian translation. For example, in the first paragraph of the source text, there is a sentence: ‘The writer’s task has never felt more challenging, the instruments for accomplishing the task never more responsive and luxurious’. This sentence can only be translated into Hungarian properly by inserting the omitted ‘has felt’ in the second phrase, as all the professional translators did e.g. ‘Az írók számára munkájuk még sosem jelentett ekkora kihívást, mint ma, s a feladat végrehajtásához sem álltak rendelkezésükre ilyen gyors és pazar eszközök’. Although the ellipsis should not be transferred unchanged into the Hungarian translation, almost half of the student translators (four out of ten) did so, creating sentences such as: ‘Az író feladata még sohasem érzõdött ennyire kihívónak, a megvalósításhoz szükséges céleszközök ennyire érzékenynek vagy pazarnak’.

The use of conjunctions  is probably the area where the discourse level grammatical differences between the two groups of translators are most vividly present. The professionals inserted several additional conjunctions into the text which enhance cohesion and the understanding of the text in general. The average number of inserted conjunctions in the professional translations is 7, but this number is only 3 in the student translations. As a result of the high number of inserted conjunctions, the professional translations became easy-to-follow, logical, cohesive and coherent. 

Lexical transformational strategies: differences

The differences in the use of the lexical transformational strategies by the professionals and the student translators are not as numerous as the ones in the use of the grammatical transformational strategies. Yet, they are also important and worth analysing. 

On the word level, a difference between the professionals and the student translators was in the use of literal translation. Literal translation can obviously be used, but the translator should be aware that in many cases the literal translation of a source word means something different in the target language. An example of this phenomenon is the word ‘prose’. This word was translated into Hungarian by a student translator as ‘próza’, which cannot be accepted in this context, as in Hungarian ‘próza’ means only a literary genre, and not ‘stílus’ or ‘írás(készség)’ as the context of the text implies. 

Remaining on the word level, another difference was that the students used foreign words, even if the word in question had a correct, usable and used Hungarian equivalent. The English word ‘computer’ was translated by a student as ‘komputer’, and by another student as ‘kompjúter’, instead of the Hungarian word ‘számítógép’. The expression ‘frustrating task’ (third paragraph) was translated by three students as ‘frusztráló feladat’, and the verb ‘hesitate’ (third paragraph) was translated by a student as ‘hezitál’. The use of foreign words is not recommended, they should be limited as much as possible, a technique that the professionals pursued, giving words like ‘számítógép’ for ‘computer’ and ‘haboz’, ‘töpreng’, ‘tétovázik’ for ‘hesitate’.

On the phrase and clause level, a difference was detected between the professionals and the students in the lexical transformational strategy called antonym translation. This strategy is completely optional, yet, it is highly recommended to use, especially in cases when the target language prefers to ‘see’ a certain segment of ‘common reality’ from the opposite angle than the source language. The analysis of the translations revealed that the professionals use antonym translation more often than student translators do. For example, the sentence ‘The craft of writing has little in common with the craft of winning a video game’ was translated by almost all the professionals by using antonym translation: ‘nem sok köze van’. Half of the students, however, used literal translation, writing ‘kevés köze van’. 

Differences were detected in the use of the lexical transformational strategies on the discourse level, too, especially in the use of collocations. The correct translation of the collocations of the source language into suitable collocations in the target language is a difficult task. The student translators tended to give a literal translation of the source language collocations, while the professionals tried to find the appropriate collocation which is used in Hungarian in the particular context. For example, the collocation ‘flatters you with the illusion’ was translated by all the professionals by an acceptable Hungarian collocation e.g. ‘abba az illúzióba ringat’ or ‘azzal áltat’, but half of the students created an unacceptable Hungarian collocation such as ‘azzal az illúzióval kecsegtet’.

Although several differences were found on all the four levels of language, still, the importance of these differences was not equal on each. On the word level, the differences were fewer in number and smaller in importance. On the word string, sentence and discourse levels the differences were more obvious and remarkable. This result reveals that the main difference between the professional translators and those who simply know two languages can be detected on the higher levels of language. 

Implications for teaching translation

The results of the present study revealed that differences in the use of grammatical and lexical transformational strategies between professionals and student translators may only be detected in the use of optional transformational strategies. Consequently, if the translation competence of a translator needs to be improved, then the appropriate use of these optional transformational strategies has to be taught to the translator. 

For the teaching of these language-pair specific transformational strategies several methods may be used. Source text analysis  (Klaudy, 1997b) may come in two types: source language-oriented source text analysis investigates the source text itself by asking a number of questions in connection with the text which help reveal the communicative situation, the message and the intentions of the author. Target language-oriented source text analysis, however, looks at the source text as a text for translation, focusing on possible translation problems.

Translation criticism and translation analysis  (Newmark, 1988) is the critical comparison of a translation (or several translations) with the original text in order to identify the correct and incorrect transformational strategies used in translation.

Guided translation (Klaudy, 1997b) is a translation exercise in which the translators have to use certain expressions or grammatical constructions where indicated. These prescribed solutions serve as indicators of the correct transformational strategies that the teacher wants the students to master.

Translation with variants (Klaudy, 1997b) means that the translators have to translate the source text by using several options. By using one or several of these teaching methods in the English language classroom, the language teacher can ensure that the translations are done more efficiently and proficiently. 

Conclusion

The use of a single original text and only a small number of professional and student translators limited the scope of my research. In many cases, only one or two examples were found for a particular transformational strategy. In order to carry out further research, a larger number of samples and a wider variety of genres would be desirable.

This fact, however, does not prevent us from seeing tendencies or stop us from drawing conclusions even from these limited number of examples. Revealing the differences between the strategies used by professional translators and language learning students can be highly useful for the language teacher. Being aware of the reason for the differences, we can ‘train’ our students to translate more efficiently.

Notes

1. This paper is an abridged version of the author’s unpublished M.A. thesis (Klatsmányi, E., 1999. Efficient methods for the teaching of written translation - developed on the basis of professionals’ versus students’ strategies. Eötvös Loránd University, Budapest) (back)

References

Bart, I., Klaudy, K., & Szöllõsy, J.  (1996).  Angol fordítóiskola.  Budapest: Corvina.
Halliday, M. A. K., & Hasan, R.  (1976).  Cohesion in English.  London: Longman.
Király, D. C.  (1995).  Pathways to translation. Pedagogy and process.  Kent, Ohio: Kent State University Press.
Klaudy, K.  (1994).  A fordítás elmélete és gyakorlata. Budapest: Scholastica.
Klaudy, K.  (1997a).  Fordítás I. Bevezetés a fordítás elméletébe.  Budapest: Scholastica.
Klaudy, K.  (1997b).  Fordítás II. A fordítás gyakorlata.  Budapest: Scholastica.
Krings, H. (1986). The use of introspective data in translation.  In: C. Faerch  & G. Kasper (Eds.),  Introspection in second language research (pp. 159-176).  Clevedon: Multilingual Matters
Newmark, P.  (1988).  A textbook of translation.  New York: Prentice Hall International.