I think that when ...Translation in the English classroom - some considerations.

Stefan Rathert

Stefan Rathert, a native German, has been living in Turkey since 2000. He teaches English at Kahramanmaras Sutcu Imam University and speaks Turkish with more pleasure than perfection.
E-mail: srathert@ksu.edu.tr

Menu

1. Translation in the English classroom vs. ELT objectives?
2. Translation
2.1 Word meaning
2.2 Error analysis by translation
2.3 Phrases and sentences
3. Alternatives to translation
3.1 General considerations
3.2 Reconstruction of a sentence
3.3 Structuring a sentence related to its content
4. Conclusion
References

1. Translation in the English classroom vs. ELT objectives?

I think that when you meet the person that you want to spend the rest of your life with, you change.

Do you think that this sentence - taken from a Pre-Intermediate coursebook (Soars & Soars, 2002b: 59) - is comprehensible with ease for a student at that level? I think of the ordinary students, not the over-motivated ones. I think of the students who must learn English because the curriculum of their school prescribes it. And I think of the students whose mother tongue is significantly different from English in terms of syntax. To demonstrate what I mean, I will give you the sentence in a German and a Turkish translation. It is not necessary for you to understand these languages; the different colours indicate where the different clauses of the English version are positioned in the German and Turkish translation, respectively:

I think that when you meet the person that you want to spend the rest of your life with, you change.
Ich glaube, dass man sich verändert, wenn man die Person, mit der man den Rest seines Lebens verbringen will, kennen lernt.
Bence hayatinin arta kalanini geçirmek istedigin kisiyle tanistigin zaman degisirsin.

I can even translate the sentence into German without destroying the structure given in English:

Ich glaube, dass, wenn man die Person kennen lernt, mit der man den Rest seines Lebens verbringen will, man sich verändert.

Although the second translation is grammatically correct, the first one is clearly better for stylistic reasons: the second version is so encapsulated that it might not be comprehensible at first sight. Be that as it may, the structure in the German sentence is similar to the English one while the Turkish translation is certainly not. The separate clauses of the original have different positions in the Turkish translation. Moreover, the subject and the verb of the main clause I think become a disjunct (bence in my opinion), and the subject and verb of the noun clause that you change the subject and verb of the main clause; so the phrase I think that you change becomes in my opinion you change in the Turkish translation. Native Turkish speakers assure me that the word-for-word translation

I think that you change. Degistigini düsünüyorum.

would result in a wrong stress of the sentence: the emphasis would be on the speaker's consideration, not on the change in someone's life. The other main difference is the position of the relative clause that you want to spend the rest of your life with, that is placed in front of its antecedent in Turkish. Moreover, the subject you can be dropped in Turkish because it is a pronoun and therefore, marked in the suffix of the relative participle istedigin:

the person that you want istedigin kisi

Is this hair-splitting grammar analysis with no relevance to the reality of content based ELT? Language analysis with the students' eyes, i.e. analysis under didactic aspects, can obviously relate to the students' language, which is of course only possible in monolingual classes if the teacher shares the students' language. The question is whether ELT teachers should explicitly focus on the students' language. This question affects the dogma that L1, and consequently translation, is unacceptable in the English classroom: Methods applied to teach a foreign language by using the learners' language are considered to be unscientific; nevertheless, the students' language is frequently heard in foreign language classes, and it is reported that its use can be effective to some extent (Tudor, 1987; Butzkamm 2003a, b with further literature). As a result, the taboo on using the mother tongue in the foreign language classroom has been called into question (Deller, 2003).

When answering the question whether to use translation or not we should be clear about both principal objectives of ELT and - as an outcome of these objectives - the language material used in class. Provided that the objectives are meaningful, the use of translation must not be inconsistent with them.

Modern course books present texts (written and recorded) containing longer and more complex sentences from early stages on. They are fairly ungraded or only slightly simplified compared to the learners' level, which means grammatical structures that are dealt with subsequently are not removed from texts, in order to expose learners to authentic language, to give them the experience of real-life English. This preview principle is taken from language learning outside the classroom when a foreign language is learned in a country where the foreign language is spoken. In such a situation learners often face structures they cannot explain or use actively, but process sufficiently that an approximate understanding is acquired. Accordingly, either comprehension of overall meaning, or reading or listening for specific information are the aims of the work with such texts. The student does not have to understand every single word or structure. Understanding and talking about texts is more important than grammatical analysis; in other words language material is taken as a starting point for communication.

Although communication is surely the most meaningful aim of ELT, we must not ignore the students who want to know how the English sentence is structured. Teachers who ignore their students' questions about language by refering to the communicative aims of English lessons will forfeit their students' favour and create a negative classroom atmosphere. We have to be prepared - possibly in a lesson phase in which we actually pursue communicative aims - to be forced to place special emphasis on grammar. The teacher's task has to be both that of a facilitator who establishes communication and that of a philologist who helps the students decipher a difficult structure in order to facilitate communication.

When we now reflect on the applicability of L1 in foreign language teaching, conclusions Butzkamm (1998) drew from the analysis of a bilingual history lesson are worth taking into consideration. We can state the issue as follows: foreign language teaching can only be successful if the focus is on both content and language; the complex dynamic of classroom processes can require a quick alternation of focus on content and language; this is often initiated by the students, and most likely verbalised in their first language in order to topicalise language. Since it is impossible to banish the mother tongue from the learners' minds, and since the mother tongue is - observable in nearly every English lesson - used by the students, so to speak, as a natural strategy in foreign language acquisition (Harbord, 1992: 351) to forbid the first language in the classroom means to forgo a "conversational lubricant" (Butzkamm, 1998). This effect of the mother tongue can be exerted if it is ensured that the mother tongue use is controlled in the classroom.

In this article I would like to show by means of some examples that translation in foreign language teaching can establish awareness of divergent modes of expression in different languages. Provided that its functional use is ensured through didactic consideration, translation does not drive the foreign language out, but contributes to communication.

Thus I find it useful to distinguish between translation as a learning tool that functions to work out meaning, correct errors, analyse language, etc., and as the lesson's learning aim itself, which means that students translate in simulated real-life situations, e.g. as a professional interpreter on a conference or in a TV studio, or as someone who happens to translate, for example, for a tourist travelling abroad who does not speak the residents' language (Deller, & Rinvolucri, 2002; Deller, 2004). Generally speaking, when the lesson focus is on translation, learners should be able to realize that a 'correct' translation has to meet the communicative intention of the original as well as to match the grammar of the target language (Macheiner, 1995: 43) and that the same idea can be translated in more than one way. I believe that three points are important when we decide to put translation in the centre of our lesson: firstly, translating of disembodied language (single, incoherent sentences) is virtually impossible because it is also (and above all) the context that builds up the meaning of an utterance. Secondly, the material that is going to be translated should not be too difficult: it does not make sense to make students translate structures they cannot master. Thirdly, a translation phase has to be embedded in a coherent sequence of teaching actions to lead to the intended learning results.

I will focus now on translation as a vehicle to solve language problems, and I should like to give some examples from which I deduce general advice on how translation can be fruitful in the foreign language classroom, as well as examples in which I think translation is not suitable.

2. Translation

2.1 Word meaning

Using translation is surely the most economical way to present or explain vocabulary. It is, however, not the most memorable because it offers only little involvement to the students. There are better ways to work out meaning (Thornbury, 2002, chapter 5).

I will show an example for a single-word translation, initiated by the teacher and in a sentence that has been understood by the students: In a Pre-Intermediate course book (Swan & Walter, 1998: 26) we read a text about a 17-year-old girl who survives a plane crash over the Peruvian jungle. After an adventurous walk through the jungle she finds a hut.

Nobody was there, but the next afternoon, four men arrived. They took her to a doctor in the next village.

These sentences have no difficulties in store. Students who have reached a certain level of proficiency will not have to translate these sentences and in all probability they will not translate them in their minds. You can trigger an 'aha-experience' for your students if you ask them the meaning of the two nexts in the sentences. They are translated differently into Turkish (next [afternoon] sonraki; next [village] en yakin). Interestingly enough, students are often not able to find an adequate translation although they understand the sentence in its overall meaning. There is obviously a difference between overall understanding on the one hand and translation into the mother tongue on the other: if the foreign expression is not foreign to us, we do not have to think about the most suitable translation. If we are forced to find the most suitable expression, we may have serious problems doing so. Translation of single words - and polysems in particular are predestined for that - is justified because an implicit objective of foreign language teaching becomes explicit: improving of native-language competence.

2.2 Error analysis by translation

The linguistic peculiarities of Turkish as an agglutinative and a verb-final language seem to cause specific mistakes such as omitting necessary words. In Turkish, the same idea is normally expressed in less words than in English. The sentence

I am going to school.

is worded in Turkish

Okula gidiyorum.

The subject is 'hidden' in the verb ending -um, and does not have to be expressed specifically. The preposition to is the ending -a in okula, expressing here a direction. The same ending is also used as dative case marker signifying the indirect object, for example:

Mektubu Murat'a verdim.

I gave Murat the letter.

While in the first example school is an adverbial phrase through its preposition to and in the second example the word order identifies Murat as the indirect object, in Turkish this difference plays no role in morphological terms. The common mistake *I am going school can be visualised easily and effectively by translation. The teacher writes the defective English sentence and its Turkish translation on the board, and asks where the ending -a in okula is expressed in English. Additionally, *I am going school can be translated word for word into Turkish to make the mistake perfectly clear to the students.

The mistake of omitting words does not always depend on different linguistic structures. When students produce sentences such as

*He works a teacher.

the teacher can ask the students to say it in L1. Being native speakers my Turkish students produce a correct sentence, of course:

Ögretmen olarak çalisiyor.

Now, the students are asked where or how olarak (˜ as) is expressed in English to bring the sentence into its correct form.

Translation is a tool that helps teachers understand their learners' problems and develop ways to overcome them. Also, students can use it to check their sentences themselves. Translation is one method of error correction which can be counted as one of the most effective for developing grammar awareness (Thornbury, 2004).

2.3 Phrases and sentences

In an elementary course book, we read a text about a woman's extraordinary home (Soars & Soars, 2002a: 41). In the heading

The lady who lives on a plane

there is a difficulty that is possibly hidden for speakers whose native languages have relative clauses. The lady lives on a plane is easy to understand, but "What is who doing in this sentence?", I hear my students ask. They know who as a question word, but not as a relative pronoun. In this case I write the heading on the board and explain that who cannot be a question word because the phrase is not a question. The next step (if necessary, taken by the teacher) is to translate who with ki o, a temporary measure, that is often used to introduce relative clauses, but not regarded as good Turkish. So we get as basis to work out the literal translation: Bayan, ki o uçakta yasiyor. Now the students themselves can bring the phrase into a proper form, which I write on the board then:

The lady who lives on a plane


Uçakta yasayan bayan

To illustrate where the different parts of the English version take part in the Turkish translation I mark them optically:

The lady who lives on a plane

Uçakta yasayan bayan

There is more to work out with the students: There is no equivalent for who in Turkish. Since Turkish does not have relative clauses, there is no relative pronoun. Who is the ending in the participle yasayan (yasamak to live). While in English the relative clause follows its antecedent, the participle phrase precedes in Turkish; uçakta yasayan is an attribute to bayan: 'the on a plane living lady' would be the analogous translation.

In a description of the plane the book says:

Then there's the living room with four emergency exit doors, which she opens on summer evenings.

you have to be prepared to be asked what which means. Especially the false beginners among the students will know which as a question word. The procedure now is similar to the one applied to explain the expression The lady who lives on the plane. Again, I point out that the sentence is not a question. Since it is only important to understand how the clauses are connected, I write the sentence paraphrased on the board:

There are four emergency exit doors, which she opens on summer evenings.

There are two obstacles to be negotiated: it is difficult to discern that which she opens on summer evenings is a subordinate clause, there are four emergency exit doors the main clause. Students tend to understand the sentence to mean She opens the four emergency exit doors on summer evenings. The second difficulty is to realize the function of which as a direct object replacing doors. Here I suggest giving which as ki onlari (onlari is accusative plural meaning them). With the knowledge of the lady who lives on a plane the students are now able to give a literal translation: Dört tane acil çikis kapisi var, ki onlari yaz aksamlari açar. Then they can transform it into natural Turkish, which is correspondingly documented on the board in this form:

There are four emergency exit doors, which she opens on summer evenings.


Yaz aksamlari açtigi dört tane acil çikis kapisi var.

In the same text, there is a third example of a difficult syntactic structure. The lady enthuses about her plane:

'The plane is 27 years old and it's the best home in the world,' says Joanne. 'It has all the things you want in a home: a telephone, air conditioning, a cooker, a washing machine, even a dishwasher.(...)'

The focus of our interest is: It has all the things you want in a home. If you check comprehension by having sudents translate it, you will often see that - confronted with an unknown structure - students tend to give the sentence a meaning that halfway fits the context: You want all these things in a home may be the first understanding of the sentence, which the students possibly change then into You want to have all these things in your home or, the lady speaking: I want to have all these things in my home. If students are unable to analyse the structure, they bring the sentence into line with what they expect using the context but often disregarding the grammatical structure. To use (and more important: to arouse) students' expectations is without doubt one of the best tools in the teacher's hand - but it can lead to misinterpretation as seen in this example.

It has all the things you want in a home is a sentence that is more complex than it seems at first sight. It contains a contact clause (so the connector is invisible because dropped), uses the impersonal you denoting people in general, and consequently formulates in a home to mean in any home in contrast to at home.

I am now trying to go through a translation process of this sentence, but I don't advocate applying this process identically in the classroom. It is, however, a good preparation for teachers to go through such a process because it increases awareness of the distances students have to cover in order to grasp a foreign language structure.

Here is a literal translation, in which the Turkish, while not very odd, does not sound entirely natural; the different constituents are marked in different colours:

It has all the things you want in a home.
O bütün seylere, ki onlari sen bir evdeistersin, sahiptir.

Looking in more detail, the Turkish equivalent of in a home has a different word order ('a home in') and ki onlari has no equivalent in the English (because of the omitted that). The definite article the cannot be translated because there is no definite article in Turkish. The colouring shows that the order of constituents is largely parallel; only the position of the verbs vary: in English they are placed after the subjects because English is a SVO language, and in Turkish at the end of the clauses because it is a SOV language; the verb of the main clause (sahiptir) is at the end of the whole sentence.

In order to produce a sentence that sounds natural, the expression ki onlari first has to be replaced. The relative pronoun, the subject, e.g as in the phrase The lady who lives on a plane, the EN-participle (here: yasayan) is used to translate the verb of the relative clause; we imitated this in the literal translation the on a plane living lady. In the sentence It has all the things you want in a home the dropped relative pronoun is not the subject, so the verb of the relative clause is translated with another participle: the DIK-participle (von Heusinger, 1996: 20). Its subject is a genitive, here: senin ( you) that governs the relative participle istedigin (from istemek want). The possessive suffix -in in istedigin refers to the genitive-subject. While the suffix -dik, adjoined to the verbal stem iste- refers to the antecedent bütün seylere. So the DIK-participle in the "relative clause construction points in two directions" (von Heusinger, 1996: 26): it is predicate modifying the genitive-subject and attribute to its antecedent. In a way, it takes on the double function the relative pronoun has in English being both a connector representing a noun of another clause and constituent in the relative clause. By the way, the genitive-subject can be dropped as well as o ( it) because they are pronouns:

Bir evde istedigin bütün seylere sahiptir.

This is quite a good translation that can alternatively be varied by replacing the verb istemek with ihtiyaç duymak (˜ need) and bütün seylere with her seye (˜ everything):

O, bir evde ihtiyaç duydugun her seye sahiptir.

Another variant strongly prefered by native Turkish speakers is:

Bir evde istedigin her sey onda var.

Here var (˜ there is/are) is used, which is much more common than sahip olmak (˜ to have). Accordingly, the sentence changes (literally translated) to There is everything you want in a house on it. We can omit onda (˜ on it) as it is clear from the context that the location of the plane is meant. Of course, there are many more conceivable variations. It is of interest here that in a very common variant

Bir evde istedigin her sey var.

with all its subvariants, the English structure is unrecognizable. This observation has to be taken into consideration when such a sentence is dealt with in the classroom.

In English lessons, to end our short digression here, it is possible to obfuscate the sentence by contortions done in English. The most elegant and easiest way is to give the crucial part of the sentence in a translation that is parallel to the 'good' translations in the examples before (the lady who lives on a plane/emergency exit doors, which she open) and covers the intended meaning of the English sentence:

all the things you want


istedigin her sey

When these three examples are presented on the board, the students gain the impression that syntactic structures (here: adjective clauses) are expressed in reverse order in English and Turkish, a common phenomen that students have come across, for example, in prepositional expressions as in Istanbul Istanbul'da or uçakta on the/a plane. Words such as who, which, when are linking words in affirmatives or negative sentences, and they are often not expressed in Turkish specifically. This is not a minor point for Turkish students who have just started learning English and a lot of practice and experience is needed to grasp such sentences. Furthermore, the process of translation becomes apparent to the students. The English structure is first decoded, i.e. it is understood and transformed into a literal Turkish translation that is capable of improvement, and then encoded, i.e. it is put in the adequate Turkish form:

The lady who lives on a plane

is decoded to

Bayan, ki o uçakta yasiyor

is encoded to

Uçakta yasayan bayan

Autonomous problem solving can only be acquired by learning the appropriate methods. Therefore decoding and encoding should be topicalised in the lesson without introducing these terms; instead, the two steps can be named 'literal translation' and 'free translation'/'translation into good Turkish'. Hopefully the students will get the idea that a translation is mostly bad if it is too near to the original because it follows the linguistic conventions of the source language, and that it is very often good if it is far from it (Macheiner, 1995: 38; 46f., 229), but experience shows that it takes students a long time to put this idea into practice.

On closer examination of the examples we can also go into detail about the teaching methods. The sentences are gradually paraphrased. In the last example only the phrase ,the things you want is translated. In order to avoid an overuse of the learners' language in the English classroom, only the crucial parts of the sentence should be translated. It is important to make the structural differences clear through visualization. The 'X' made by the crossed lines between the English and Turkish sentences becomes a symbol for the syntactic relation between the two languages. Although syntax is the central topic during the lesson phases in which the structures are analyzed, a grammar-centered instruction style should be avoided; the structures are explained so far as to ensure the learners' understanding of the structure; grammatical explanation, e.g. on adjective clauses, is not done in the lesson (and would be absolutely meaningless at that level). The timing of the translation phases in the lesson is flexible, according to the specific demands of the class. If it is not necessary to deal with them earlier, the teacher should take up the difficult structures at the end of the lesson, but students' initiative may suggest a different procedure. Provided that language work (here: translation) is done collaboratively, the disadvantage of distracting from the actual content is compensated by better teaching results (Ellis, R., Basturkmen, H., & Loewen, S., 2002: 427). In any case, the presented translation sections should be done as briefly as possible in order not to distract the students from the topic: the extraordinary home of an eccentric lady. And finally: to analyse language by means of translation is an option, not a must - for all the examples here.

3. Alternatives to translation

3.1 General considerations

One method to clear up a difficult sentence or paragraph is to translate it. The teacher makes a good student translate it, or translates it himself or herself and all problems are solved - but only in the language the sentence is translated into. Especially when long or difficult structures are translated which part of the translation corresponds to which part of the original can remain unrecognisable, all the more when a ready translation is presented to the students instead of going through the translating process with them. Students (and teachers) deceive themselves when they think that a sentence must have become clear by translation.

Translation means to turn the spotlight on the learners' language. Learning a foreign language, however, means habitualising it by using it regularly; if students make frequent detours via their mother tongue, they will not internalise the foreign language structure; a large majority of the Turkish students I have come in contact with cannot communicate actively in English, even if they have been studying it for many years, because it seems English lessons in Turkey are being held in Turkish in large measure. Students cannot develop a "feeling" for English when the comprehension and discussion of texts is, if done at all, mainly in their first language. In such a teaching situation, texts are often used as quarries where grammar is dug up out of the sentences. As a consequence of this method (the infamous 'grammar translation method'), learning English is mainly understood as learning English grammar using the mother tongue as lingua operandi at quite an abstract level. The effect is that classes are boring and frustrating for both students and teachers. Teachers applying those 'teaching methods' have never asked themselves why their students should learn English, and they are actually not aware of their teaching objectives. Missing a didactic basis, the mother tongue use is uncontrolled, and English is crowded out.

For these reasons it can be advantageous in specific cases to apply a monolingual grammar explanation (in order to facilitate communication, not as an end in itself!). I should like to illustrate this with two examples, and give reasons for restriction to English.

3.2 Reconstruction of a sentence

A text (elemantary level) about the invention of the drug Aspirin (Soars & Soars 2002a: 62) reads:

Felix Hofman, a 29-year-old chemist who worked for the German company Bayer, invented the drug Aspirin in April 1879.

In some ways this sentence is easy to understand, in some ways it is not: it has some pieces of information that the students can separate easily from each other: name, age, occupation, place of work, invention of Aspirin, year of the invention. On the other hand it is difficult to see how the pieces are bound together. Attentive students may also notice the structure 29-year-old and clever students will ask why it is not *29-years-old. If you have a class that is not absolutely without humour, you can try this way: Ask a student for his name and age and write on the board: Mustafa is 20 years old. Ask him now for his job and change the sentence into Mustafa is a 20-year-old student. Explain them that 20-year-old is an adjective which could be changed with any other adjective (e.g. Mustafa is an intelligent student.) and that we do not say *20-years-old. Ask him (and the other students) now what he invented - this is why your students should have a sense of humour! For instance, one of my students recently answered that he had invented ice cream (the place where I live is famous for its ice cream). So I write the original sentence on the board, erasing is, putting commas and leaving a space:

Mustafa, a 20-year-old student, invented ice cream.

Now I ask him when he invented it and write:

Mustafa, a 20-year-old student, invented ice cream in 1995.

Normally no one in the classroom takes offence at the fact that Mustafa was not 20 years old in 1995, so do not go into this problem. Until this point the sentence is easy to understand, not least because it was developed step by step together with the students. Now we come to the most difficult point, especially for students whose mother tongue uses a different syntactic structure to express what is formulated in a relative clause in English. Ask Mustafa where he worked in 1995. Mustafa worked for Café Yasar (the most famous ice cream parlour here):

Mustafa, a 20-year-old student, invented ice cream in 1995.
Mustafa worked for Café Yasar.

And now you can connect the sentences moving the comma and replacing Mustafa with the relative pronoun:

Mustafa, a 20-year-old student who worked for Café Yasar, invented ice cream in 1995. .

It is not a disadvantage if your students have heard about who-sentences before. The students will not have learnt relative clauses systematically at this stage, and probably they will not know the term "relative clause". Do not teach it here; the knowledge that there is such a grammatical phenomenon has no function, but the understanding of the structure in the Felix-Hofman sentence is a good preparation for an easier understanding of relative clauses when they are systematically dealt with later. Because of the easily separable information clusters of the sentence there is neither need to go deeper into grammar nor to translate it; as an alternative a new sentence is created using the structure given in the sentence. It is recomposed, so to speak.

3.3 Structuring a sentence related to its content

Finally I will discuss how the sentence quoted at the beginning of this article can be illustrated to students.

I think that when you meet the person that you want to spend the rest of your life with, you change.

One difficulty of the sentence is the fact that it does not have difficult vocabulary! The sentence about Felix Hofman had some eye-catchers which this sentence misses totally. It is nearly impossible for the untrained eye to pick up pieces of information out of the sentence.

The first step is to write the sentence on the board and to ask the students to find syntactic breaks in the sentence. They normally name the breaks before that (2x) and after with (because there is a comma), but they might have problems finding the break after the first that probably because they do not expect it at this place as there is a break before that. If they do not find it, remind them that linking words are a signal for a new clause. So, they will find the break between that and when:

I think | that | when you meet the person | that you want to spend the rest of your life with, | you change.

Experience shows that it is usually difficult to see that you change is the continuation of I think. Therefore, it is better to leave the area of syntax and to sort out the problem keeping a close eye on meaning. Starting from the beginning of the sentence (I think) it is natural to ask: 'What do you think?' This question is documented on the board in the following way:

I think | that | when you meet the person | that you want to spend the rest of your life with, | you change.
I think what?

The students will search the sentence to find an adequate answer. Since it is a frequent phrase, they will immediately see that I think and that form an integrated whole. It is now manageable for them to check which part of the sentence is its continuation - when you meet (...) and that you want (...) have to be ruled out because the clauses are introduced by linking words and two linking words in succession are not possible within one clause. So the next step can be developed as follows:

I think | that | when you meet the person | that you want to spend the rest of your life with, | you change.
I think what?
that you change

The next question is determined in advance by the sentence itself:

I think | that | when you meet the person | that you want to spend the rest of your life with, | you change.
I think what?
that you change when?
when you meet the person

Since the person was not introduced in the previous sentence, it (the noun 'person') needs a modification which can only be given in the remaining clause that you want to spend the rest of your life with. So the board gets its final appearance:

I think | that | when you meet the person | that you want to spend the rest of your life with, | you change.
I think what?
that you change when?
when you meet the person what person?
that you want to spend the rest of your life with.

Breaking up a sentence into parts - the etymological sense of syntactic analysis - has always been done in English lessons and its importance will remain. This method has an enormous advantage: the students can see that there is a relation between formal structure and meaning. When analysing a sentence in this way, it is therefore better to go over its content instead of explaining it grammatically, e.g. by identifying noun clauses, relative clauses, etc.

As stated before, a translation of a complex sentence very probably does not elucidate how the clauses are linked because where each part of the English sentence is positioned in the translation is not clearly visible. Only the most difficult link (for Turkish students) the person that you want... is worth translating and the most suitable translation for I think can be discussed (see first paragraph). Translation then has a different function: learning control. Nevertheless, it may not be necessary to check the students' understanding after clarifying the sentence in the way described above. In summary, decomposition is a good alternative to translation, especially for longer and more encapsulated sentences.

4. Conclusion

Translation in ELT can raise awareness of different methods of grammatical encoding in different languages. It especially makes sense to illustrate differences in syntax between languages. Such differences normally appear at link joints in sentences, so it is usually enough, and in fact better, only to translate the link joints and not the whole sentence. To explain the grammatical structure of longer, nested sentences, it is better to use analysis techniques. If translation is used in the English classroom, it is important that the process of translating is done by the students as independently as possible.

Translation and the learners' language can be used in many more ways than presented here (Deller, & Rinvolucri, 2002). Using the mother tongue in ELT means applying 'natural' methods children use growing up in bilingualism (Butzkamm 2003b). In foreign language acquisition, the learners derive benefits from their own language with respect to both languages: Native-language competence is developed, and the foreign language acquisition is improved because it is connected to the mother tongue. It is a truism that we process and remember learning topics more successfully if they are anchored in existing knowledge. Translation in the English classroom ensures that learners feel in control of what they are learning; they learn in a meaningful way because their mother tongue is the coordinate system in which they move and approach the foreign language: "native language skills in the phonological / orthographic, syntactic, and semantic codes form the basic foundation for FL [foreign language] learning." (Ganshowe & Sparks, 2001: 87). So why should the mother tongue be excluded from the English classroom?

In foreign language teaching, translation is just one method among others. Its use enhances the variety of methods and can meet the special needs of students whose mother tongue has a grammar which is fundamentally different from the English grammar.

Acknowledgements

I am grateful to my colleagues at the Foreign Languages Department of Kahramanmaras Sutcu Imam University who contributed to the translation examples in this article.

References

Butzkamm, W. (1998) Code-switching in a Bilingual History Lesson: The Mother Tongue as a Conversational Lubricant. International Journal of Bilingual Education and Bilingualism, 1:2, 81-99
Butzkamm, W. (2003a) Die Muttersprache als Sprach-Mutter: ein Gegenentwurf zur herrschenden Theorie. französisch heute, 34, 174-192
Butzkamm, W. (2003b) We only learn language once. The role of the mother tongue in FL classrooms: death of a dogma. Language Learning Journal, 28, 4-14
Deller, S. (2003) The Language of the Learner. English Teaching Professional, 26, 5-7.
Deller, S. & Rinvolucri, M. (2002) Using the Mother Tongue: Making the most of the learner´s language. London: Baskerville Press Ltd.
Ellis, R., Basturkmen, H., & Loewen, S. (2002) Doing focus-on-form. System, 30, 419-432
Ganshowe, L. & Sparks, R. (2001) Learning difficulties and foreign language learning: A review of research and instruction. Language Teaching, 34, 79-98.
Harbord, J. (1992) The Use of the Mother Tongue in the Classroom. English Language Teaching Journal, 46: 4, 350-355
von Heusinger, K. (1996) Turkish Relative Participles. A Reanalysis in Categorial Grammar. Arbeitspapier 78. Fachgruppe Sprachwissenschaft, Universität Konstanz. Retrieved July 17, 2006 from: http://ling.uni-konstanz.de/pages/publ/PDF/ap078.pdf
Macheiner, J. (1995) Übersetzen. Ein Vademecum Frankfurt am Main: Eichborn
Soars, J. & Soars, L. (2002a) New Headway English Course. Elementary. Students' Book English. 7th impression. Oxford: Oxford University Press
Soars, J. & Soars, L. (2002b) New Headway English Course. Pre-Intermediate. Students' Book English. 8th Impression. Oxford: Oxford University Press
Swan, M. & Walter, Catherine (1998) The New Cambridge English Course. Student 2. Thirteenth Printing. Cambridge: Cambridge University Press
Thornbury, S. (2002) How to Teach Vocabulary. Essex: Pearson Education Limited
Thornbury, S. (2004) English Teaching Essentials: Grammar. English Teaching Professional, 32, 40-41
Tudor, I. (1987) Using Translation in ESP. English Language Teaching Journal 41, 268 - 273.

Please check the English for Teachers course at Pilgrims website.
Please check the Teaching Advanced Students course at Pilgrims website.