Keywords

1 Introduction

Relative clauses are sometimes called adjective clauses and follow the noun which they describe. They are dependent clauses which give additional information about a noun in the main clause, and they follow that noun in the complex sentence they create. The most important work on the complexity of relative clauses was the proposal by Keenan and Comrie in 1977 based on language typology (Izumi, 2003), which studies similarities and differences among all languages in the world (Gass & Selinker, 2008, cited in Minako, 2009).

According to Noam Chomsky (1981), The Head-Direction Parameter (HDP) was hypothesized to govern two basic, widely attested word order options: subject-object-verb (Mongolian) and subject-verb-object (English). Many studies have provided evidence that the differences in the principal branching directions between the two languages often cause difficulties in acquiring complex syntax (e.g., Flynn, 1984; Flynn & Espinal, 1985). For example, Schachter (1974) examined the composition data written by Persian, Arabic, Chinese, and Japanese learners of English. She observed that the Chinese and Japanese groups produced significantly fewer relative clauses than did the Persian and Arabic groups. She explained that it is because the native and target languages form relative clauses in strikingly different ways. She also noted that while Chinese and Japanese learners do not use relative clauses with great frequency, they use them with a high degree of accuracy when they do use them (Chang, 2004).

The head-direction parameter plays a significant role in the acquisition of L2 or foreign language relative clauses. When L1 and L2 are the same with regards to the head-direction parameter, the acquisition of relative clauses is facilitated. When L1 and L2 differ with regards to the head-direction parameter, the acquisition of relative clauses is difficult for learners, as they must assign a new value to the parameter in acquisition.

Acquisition of English relative clauses by different language learners has been studied actively for the last few years, but there haven’t been any studies on Mongolian learners’ acquisition of English relative clauses so far. For this reason, the present study examines evidence for head-directionality differences causing difficulties for Mongolian learners in acquiring complex structures such as relative clauses in English.

Moreover, the paper tries to examine whether grammar-translation is good method to help students understand the complex structures of foreign languages such as relative clause word ordering, etc.

2 Differences Between English and Mongolian Relative Clauses

Noam Chomsky’s Principles and Parameters’ Theory in the 1980s introduced the idea that a small number of innate principles are common to every human language (e.g., phrases are oriented around heads), and that these general principles are subject to parametric variation (e.g., the order of heads and other phrasal components may differ). All languages have phrase structure, or a systematic way of organizing the different elements in a sentence. This is a linguistic universal. Not all languages, however, have the same type of phrase structure. Head directionality is connected with the type of branching that predominates in a language: head-initial structures are right-branching, while head-final structures are left-branching.

While English is a head-initial and left branching language, Mongolian is head-final and right branching language. Thus, the relative clause in English follows the head noun phrase (NP) while the relative clause in Mongolian precedes the head noun phrase (NP). Let’s see some examples:

  1. (1a)

    Mongolian: Minii tűűnd őgsőn nom mash sonirholtoi bilee.

    I-Gen 3SG-Dat give-pst book-Nom very interesting Cop

  2. (1b)

    English: The book which I gave him is very interesting.

As the tree shows:

Also, English uses relative pronouns such as who, whose, when, which, where and that.

  1. (2a)

    English: The house [where he lives is very near].

Likewise, there are no explicit markings such as relative pronouns between a relative clause and the head noun in Mongolian as shown in (2b).

  1. (2b)

    Mongolian: [Tűűnii amidardag] baishin mash oirhon.

    3SG-Gen live-Hab house-Nom very near

Diessel (2004) stated that English has a wide variety of relative clauses, which are commonly classified based on two structural features: (i) the syntactic role of the HEAD, that is, the main clause element that is modified by the relative clause; and (ii) the syntactic role of the GAP, that is, the element that is gapped or relativized inside the relative clause. One of the most important components of a noun phrase is the head noun. In English, the head noun becomes the center of attraction of the noun phrase. This part cannot be omitted from the phrase, whereas the pre modifier or post modifier can be omitted.

Most Mongolian relative clauses are pre-nominal. They can be classified into three types, i.e., relative clauses proper, gapless relative clauses, and headless relative clauses. Relative clauses proper are gapped relative clauses with head nouns.

One of the features of Mongolian relative clauses is that their subjects are mostly marked in the genitive case as shown in (1a), (2b) and (3a) while the subjects of English relative clauses are always marked in the nominative case as shown in (1b), (2a) and (3b).

  1. (3a)

    Mongolian: [Minii tűűnd őgsőn] nom mash sonirholtoi bilee.

    I-Gen 3SG-Dat give-pst book-Nom very interesting Cop

  2. (3b)

    English: The book [which I gave him] is very interesting.

Depending on the time of action, however, verbs in both English and Mongolian relative clauses can be in different tenses and aspects. Let’s see this in some examples:

The verbs in relative clauses are in the past tense:

  1. (4a)

    Mongolian:[Őčigdőr irsen] hűműűs buudald baigaa.

    Yesterday come-Pst people-Nom hotel-Loc exist-Prs

  2. (4b)

    English: The people [who came yesterday] are at hotel.

    The verbs in relative clauses are in the progressive tense:

  3. (5a)

    Mongolian: [Tűűnii hajuud  zogsoj baigaa] ohin manai angiin suragch

    [3SG-Gen next to stand-ImPf to be-Prog] girl-Nom our class-Gen pupil.

  4. (5b)

    English: The girl who is standing next to him is pupil of my class

    The verbs in relative clauses are in the habitual aspect:

  5. (6a)

    Mongolian: [Tűűnii ajilladag] gazar maš tohilog.

    [3SG-Gen work-Hab] place-Nom   very comfortable.

  6. (6b)

    English: The place where he works is very comfortable.

Instead, the Mongolian sentence-final word, or verb of the relative clause, can be in a different tense and/or aspect. Headless relative clauses are gapped relative clauses without overt head nouns. For example:

Dorj zaluudaa hanilsantaigaa nasyg eleej baigaa hűn.

Dorj youth-Loc marry-Pst-Com-Ref age-Acc spend-ImPerf to be-Prs person-Nom.

Dorj is the person who has spent his whole life with the one whom he married to.

As we have seen from the example, there is no overt head noun in Mongolian sentences. Instead hanilsan- of hanilsantaigaa refers the head noun covertly.

3 Mongolian Language Interference in Learning English Relative Clauses

One of the crucial factors that influences and characterizes foreign language learning is language transfer. Sentences in the target language may exhibit interference from the mother tongue of the learner. The basic problems of foreign language learning arise not out of any essential difficulty in the features of the new language itself, but primarily out of the special ‘set’ created by first language habits (Fries, in Lado, 1957; Bley-Vroman, 1989, p. 55).

Ferguson (1965) points out that one of the major problems in the learning of a second language is the interference caused by the structural differences between the learner’s native language and their second language. A natural consequence of this conviction is the belief that a careful contrastive analysis of the two languages offers an excellent basis for the preparation of instructional materials, the planning of courses, and the development of actual classroom techniques (Ferguson ibid, p. 4).

Interference can affect in positive as well as in a negative ways, though with greater differences between the two languages, more negative effects of interference are expected. We should take into account that languages with more similar structures are more likely to be exposed to mutual interference than languages with fewer similar features.

According to Berthold et al. (1997), interference may be viewed as the transference of elements from one language to another at various levels including phonological, grammatical, lexical, and orthographical. Grammatical interference is defined as the first language influencing the second in terms of word order, use of pronouns and determiners, and tense and mood (Berthold et al., 1997, cited in Archvadze, 2015, p. 3)

One of the reasons why Mongolian learners of English are confused in comprehending and producing English relative clauses is because of the differences in the structures of English and Mongolian. In other words, the knowledge of the Mongolian language influences the learner’s English relative clause production. Mongolian learners of English may have problems with English relative clauses due to the fact that Mongolian is a left-branching SOV language while English is a right-branching SVO language. Likewise, the presence of relative pronouns in English and their absence in Mongolian causes problems as well.

4 Research Questions

The present research aims to investigate what difficulties Mongolian students have due to the structural differences between the relative clauses of the two languages. The study attempts to answer the following questions:

  1. 1.

    What kinds of errors do Mongolian learners make in producing English relative clauses?

  2. 2.

    What causes Mongolian learners to make such errors?

  3. 3.

    Is translation an effective way to make students understand the complex structure of relative clauses?

5 Research Methodology

Participants

100 students aged between 15 and 18 (10th–12th grades) who have studied English for 5–7 years were chosen for the study. Subject and object relative clauses had been taught to all participants, and how well each participant learned varied depending on the level of their English knowledge.

Data collection

In order to examine the first and second research questions we had given two tasks to the students in the Pre-test and carried out an error analysis on the students’ tests. In the pre-test, each of the participants was first given 5 item word order test and 5 translating sentences (from Mongolian into English) relating to relative clauses in order to check learners’ background knowledge about English relative clauses. Subjects were given 30 min to finish both tasks (Appendix 1).

In order to examine the third research question we gave the students two Post-test tasks similar to those in the Pre-test. The participants were given 30 min to finish both tasks (Appendix 2).

Data analysis

Pre-test data were analyzed in terms of: (1) the influence of Mongolian relative clause structure, and (2) the type of errors produced by students. Error analysis was used.

Post-test data were analyzed in terms of: (1) the result of translation teaching, and (2) which group of students had done post-test better. Comparative analysis was used.

6 Pre-test

First, in order to check the students’ background knowledge of relative clauses, the students were given an exam with 2 tasks: (1) to translate Mongolian sentences into English and (2) to create sentences by putting given words in the correct order. 72% of the students made errors in translating sentences and the remaining 28% couldn’t translate them at all. In the second task, however, only 57% of the students made mistakes while the remaining 43% were able to do it correctly. As we have seen from the results of the exam, the students have more problems in translating complex sentences than in creating sentences from a given set of words. The errors made by the participants have been analyzed in the next section in detail.

6.1 Error Analysis

Error Analysis is recognized as an essential branch of applied linguistics and constitutes the appropriate start for Second Language Acquisition. The psychological basis of Error Analysis is cognitive theory which is related to Chomsky’s language acquisition and universal grammar. Therefore, it is a methodology for describing second language learners’ language systems.

In fact, the error analysis is basically linguistic analysis on complex phenomena of language learning. The role of error analysis is very important in second and foreign language teaching and learning. Differences between language and culture should be taken into consideration in order to deal with transfer, and as a result, teaching will be more effective.

Errors on part of the learner are seen as an integral part of language learning which is used in teaching grammar, linguistics, psychology, etc. Therefore, an integration of three disciplines is needed to deal with the complexities of second language acquisition and provide empirical evidence for the improvement of teaching methodology, syllabus designs, and teaching techniques in English language classrooms (Akbar, 2012, p. 1031).

Analysis of second language learners’ errors can help identify the learner’s linguistic difficulties and needs at a particular stage of language learning. Moreover, errors made by learners will help teachers to foresee what may be difficult or easy for them, and will provide clues for how to act.

Error analysis should be done in order to understand the influence of the Mongolian language on English relative clause acquisition by Mongolian students. It is noticed that students have made numerous mistakes in word ordering and relative pronoun use. First, let us see the result of the first task: translation of Mongolian sentences into English (Table 1).

Table 1 The representations of total errors in learners’ relative clause translation (task 1)
Full size table

Due to the lack of relative pronouns in the Mongolian language, 68% of the learners omitted relative pronouns entirely. 12% of them selected the wrong relative marker (pronoun) and 79% of the participants made mistakes in word order in the relative clauses. The following sentences (7a, 8a, 9a, 10a, 11a) were given students to translate into English.

  1. (7a)

    Tend suuj baigaa hűn bol minii ah.

    There sit-Conv.ImPF to be-Pr person-Nom to be 1SG-Gen brother.

  2. (7b)

    The man who is sitting there is my brother.

  3. (8a)

    Hődőő amidardag hűűhdűűd bagaasaa mori unaj surdag.

    Countryside-LocШ live-Hab children-Nom childhood-Abl horse-AccШ ride-Imp learn-Hab.

  4. (8b)

    The children who live in the countryside learn to ride horses in childhood.

  5. (9a)

    Ene bol tűűnii surch baisan surguuli.

    This is 3SG-Gen study- Conv.ImPF to be-PF school-Nom

  6. (9b)

    This is the school where he used to study.

  7. (10a)

    Tűűnii surdag surguuli endees holgűi.

    3SG-Gen study-Hab school-Nom here-Ablfar-Neg.

  8. (10b)

    The school where he studies is not far from here.

  9. (11a)

    Öčigdör minii hudaldaj avsan üzeg maš goyo.

    Yesterday 1SG-Gen buy-Conv.ImPF take-PF pen-Nom very nice.

  10. (11b)

    The pen which I bought yesterday is very nice.

Let us see some mistakes that students made when they translated the above sentences.

Omission of relative pronoun:

  1. (12)

    My brother is sitting there.

  2. (13)

    *There is my brother sitting.

  3. (14)

    *There sitting man is my brother.

  4. (15)

    *There sitting person is my brother.

  5. (16)

    *Their sitting person is my brother.

  6. (17)

    *This is very nice pen, yesterday I bought.

  7. (18)

    *Yesterday my bought pen is beautiful.

  8. (19)

    *My bought pen is very nice yesterday.

  9. (20)

    *This is his studied school.

  10. (21)

    *It was school he studied.

Most of the students did not use relative pronouns at all when they translated the Mongolian sentences (7a, 8a, 9a, 10a and 11a). The relative pronoun who is omitted in (13), (14), (15), (16); the relative pronoun which is omitted in (18) and (19); the relative pronoun where is omitted in (20) and (21). Instead of using a relative pronoun, they simplified the structure of the sentence in the above examples.

Simplification refers to the situation when learners avoid the use of the complex structures and prefer to use the simpler forms. Sometimes, this also results in errors. Simplification errors are the errors that are caused by simplifying or leaving out some elements. Thus the above sentences and the following are examples of simplification errors.

  1. (22)

    *Yesterday I bought a nice pen.

  2. (23)

    *I bought a good pen yesterday.

  3. (24)

    *His studied school near from here.

  4. (25)

    *He’s school not far from here.

  5. (26)

    *He was studied in this school.

  6. (27)

    *This was his studied school.

Even though some of the sentences are grammatically correct, they were not expected to be translated that way. In other words, the students should have translated the sentences using relative clauses. Let’s see some examples:

  1. (28)

    His school is not far from here.

  2. (29)

    His school is not far away from here.

The expected translation was: The school where he studies is not far from here. The Mongolian sentence is as shown in (10a). In examples (28) and (29) the pronoun his might be caused by interference of the Mongolian language because the subject of the relative clause is mostly marked in the genitive case in Mongolian. Thus the students translated tűűnii -3SG-Gen into his in English in the test item (10a).

  1. (30)

    He used to study in this school.

  2. (31)

    He studied in this school.

  3. (32)

    It was his school.

Expected translation was: This is the school where he used to study. The Mongolian sentence is as shown in (9a). It is noticeable that the students avoided relative clauses due to the difficulties arising from the difference between the relative clause structures of the two languages.

Word ordering is considered as the syntactic arrangement of words in a sentence, clause, or phrase. In other words, it is the order in which words occur in sentences. Furthermore, it refers to the different ways in which languages arrange the constituents of their sentences relative to one another. A noun phrase with a relative clause is a phrase that consists of a pronoun or noun with an adjective clause or relative clause which can consist of a subject and a verb and (optionally) an object. Word order in a noun phrase with a relative clause is relative clause + head noun in Mongolian. On the contrary, word order in a noun phrase with a relative clause is head noun + relative clause in English. It is observed that most of the participants had problems with ordering in a noun phrase with a relative clause when they translated Mongolian sentences with a relative clause into English. As mentioned above, 79% of the participants made ordering errors. In the translation task they made errors by putting the relative pronouns in different positions instead of putting them after the head noun as shown in (33) and (34)

  1. (33)

    *That pen is very nice which I bought yesterday.

The expected translation was: The pen which I bought yesterday is very nice. The Mongolian sentence is as shown in (11a). In the example, the relative clause which I bought yesterday is placed right after the adjective nice.

  1. (34)

    *The school was not far from here where he studied.

The expected translation was: The school where he studies is not far from here. The Mongolian sentence is as shown in (10a). In the example, the relative clause where he studied is placed right after the word here. The errors violate the structure dependency principle of phrase structure.

Some of the participants translated Mongolian sentence (7a) as follows:

  1. (35)

    *There is my brother who is sitting. In the example,

Some of the participants translated Mongolian sentence (11a) into English as shown in:

  1. (35)

    *Yesterday I bought a pen which is beautiful.

In the example, even though the relative clause which is beautiful grammatically follows the head noun pen, the expected relative clause was which I bought yesterday.

Selection refers to the problem of improper selection of certain forms. According to the study, 13% of the students had problems with selecting relative pronouns. In other words they chose the wrong relative pronoun for the translation. Let’s see some examples:

  1. (36)

    *His school which he studies isn’t far here.

In this example, the relative pronoun which is used instead of the relative pronoun where But where is a location-relative pronoun that joins a modifying clause. In the example, the clause modifies a place. If which is used, it should have preposition in denoting a place in that case. Thus, the correct form should be as shown in (9b) This is the school where he used to study. The Mongolian sentence is as shown in (9). Or it can be as following: This is the school in which he used to study. The following are also improper selection examples:

  1. (37)

    *This is school which he was study.

  2. (38)

    *This is a school which he studied.

  3. (39)

    *This school is that he studied school.

In this example, students had two kinds of errors: improper selection of relative pronouns and disordering. The other types of errors are not considered at this time.

The second task was to put given words in the correct order to create sentences. This task aimed to check the students’ understanding of the word order of relative clauses.

In general, word order errors can significantly complicate comprehension. Let’s see some detailed examples:

  1. 1.

    freezer, ice cream, the, I, ate, the, that, was, in. The intended sentence is “I ate the ice-cream that was in the freezer”

    1. (40)

      *I ate the ice-cream was in that the freezer.

    2. (41)

      * I ate that ice-cream was in the freezer.

    3. (42)

      * I ate that the ice-cream in the freezer.

    4. (43)

      *I was in the freezer that ice-cream ate.

    5. (44)

      *I ate in the freezer that was the ice-cream.

In the above (40), (41), (42), (43), (44) examples, the students might be confused between demonstrative pronoun that and relative pronoun that.

  1. 2.

    a lot, ask, who, learn, students, questions. The intended sentence is “Students who ask questions learn a lot”

    1. (45)

      *Students learn a lot questions who ask.

    2. (46)

      *Students learn who ask a lot questions/*Students learn who ask questions a lot.

    3. (47)

      *Students learn a lot who ask questions.

    4. (48)

      *Students ask questions who learn a lot.

    5. (49)

      *Students who learn a lot questions ask.

    6. (50)

      *Students ask who a lot questions.

    7. (51)

      * Who ask students a lot questions learn?

    8. (52)

      * Students learn “Who ask a lot questions?”

    9. (53)

      * Who ask questions, students learn a lot?

    10. (54)

      * Who learn students ask a lot of questions?

According to the study, it is obvious that the second scrambled sentence was the most difficult one for the students to reorganize into the correct sequence. As shown in Table 2, ten variations of the incorrect sentences (45)–(54) were made by 84% of the students. If we examine the above sentences, the students might confuse who as the interrogative pronoun as shown in (45), (46), (47), (48), (50), (51), (52), (53) and (54). In example (49) even though who might be considered as relative pronoun, the sentence is ungrammatical due to the improper ordering of the other words in the sentence.

Table 2 The representations of word ordering in sentences with relative clauses (task 2)
Full size table
  1. 3.

    children, I, a, woman, two, know, has, who. The intended sentence is “I know a woman who has two children”

    1. (55)

      *I know who a woman has two children.

    2. (56)

      *I know who has a woman two children.

    3. (57)

      *I has a woman who know two children.

    4. (58)

      *I has know who a woman two children.

    5. (59)

      *Who I know a woman has two children?

    6. (60)

      *I woman who has know two children.

In examples (55), (56), (58), (59), the students might be confused about the interrogative function of who versus the relative function of who. But in examples (57) and (60) even though who might be considered a relative pronoun, it is disordered. As we have seen from the examples, the students have been unfamiliar with the structure of the English relative clause.

  1. 4.

    countryside, who, have, I, the, some, live, in, friends. The intended sentence is “I have some friends who live in the countryside”

    1. (61)

      *I live in the countryside who have some friends.

    2. (62)

      *I have who the some friends live in the countryside.

    3. (63)

      *I have the some who friends live in the countryside.

    4. (64)

      *Who have some friends I live in the countryside?

According to the study, this scrambled sentence was the easiest one for students to arrange. It can be proven that 27% of the participants had problems as shown in Table 2. In the example, the relative pronoun was placed in many different places instead of being placed after the head noun friends. In the example (64), the pronoun who was considered an interrogative pronoun who.

  1. 5.

    my, bought, gave, friend, he, the, pen, which, yesterday, me. The intended sentence is My friend gave me the pen which he bought yesterday.”

    1. (65)

      *My bought pen which he gave me friends yesterday.

    2. (66)

      *My friend bought the pen yesterday which he gave me.

    3. (67)

      *Yesterday my bought the pen which friend he gave me.

    4. (68)

      *My friend gave me yesterday which he bought the pen.

    5. (69)

      *Yesterday he bought the pen which my friend gave me.

    6. (70)

      *My friend gave me which he bought the pen yesterday.

    7. (71)

      *Yesterday my friend bought the pen which he gave me.

According to the study, 76% of the participants had problems properly ordering the constituents of this scrambled sentence. The relative pronoun which is placed separately from the head noun in examples (66), (68) and (70). In examples (65), (67), (69) and (71) the positions of the subject and object of the relative clause and main clause are exchanged.

In general, if we look at the participants’ errors we find one type of mistake that was particularly frequent in the second task: in the sentences, the participants changed the word order such that they produced a relative clause different from the one in the test item.

6.2 Findings

When Mongolian learners produce English relative clauses, the influence of native language interference occurs in most cases. According to the result of the test, it can be said that more learning difficulties might be expected due to substantial differences between English and Mongolian relative clauses. One of these differences is the position of the relative clause in a sentence. The Mongolian learners face difficulties in learning English relative clauses due to the head direction or word order difference between English and Mongolian. In other words, the interference from the word order of the Mongolian adjective clause (relative clause) creates much more difficulty in acquiring English relative clauses. When students translated the sentences, they commonly made errors by putting the relative clause and head noun separately. The relative clause pronouns were omitted in English due to the absence of the relative pronoun in the Mongolian language.

On the other hand, it can be said that some of the errors made by the students have been caused by the lack of knowledge of the structure dependence principle of a language. Thus teachers should explain the structure dependence principle to their students.

7 Treatment

In order to check whether translation is a good way to solve the problems of Mongolian students with English relative clauses and its importance for teaching and learning the complex structure of relative clauses, we carried out another experiment on the same students. In other words, we tried to find out the answer to the 3rd research question as mentioned above in Sect. 4. Before we explain the study in detail, let us briefly introduce the basis of the translation method.

While some scholars consider translation an unsuitable activity for language learning and teaching, many theorists, linguists, and teachers agree on the importance of using translation in foreign language classes. For instance, Schaffner (1998) claims that translation and related exercises could be beneficial to foreign language learning for the following reasons: (1) To improve verbal agility. (2) To expand students’ L2 vocabulary. (3) To develop their style. (4) To improve their understanding of how languages work. (5) To consolidate L2 structures for active use. (6) To monitor and improve the comprehension of L2.

The importance of translation in foreign language classes lies in the comparison of grammar, vocabulary, word order and other language features in the target language and the student’s mother tongue. Students are directly exposed to the contrasting language systems of the target and native languages. Therefore, the learners should be required to discuss and correct common mistakes. Correcting mistakes in translations is a challenging activity for students. Some incorrect word-for-word translations are handed out, and the students are encouraged to discuss and correct the mistakes. Translation exercises are considered to be a positive learning resource for them to comprehend and memorize and to acquire English skills.

According to Ross (2000), if students are aware of the differences, language interference (transfer) from their own language is likely to be reduced. Moreover, translation in the L2 classroom offers a way to highlight similarities and differences between L1 and L2 forms. Word-for-word translation, that is, literal translation, turned out to be a big problem. By translating sentences in parallel, translators may transfer sentence lengths and characteristics of L1 into L2, making the translation obscure or unnatural. L1 influence appears to be strongest in complex word order phrases and in word-for-word translations of phrases.

Benjamin (1968) states that a faithful word-for-word translation will not transmit the original sense. Thus, according to Sharwood Smith (1974) in order to produce acceptable English sentences, the learner needs to know the restrictions on word ordering. In particular, they must know the certain areas within the sentence where no extra words may be placed or at least where an extra word would create a feeling of clumsiness for a native speaker.

As mentioned above, in order to examine whether the translation method, especially the grammar translation method, is an effective way to teach and learn the complex structure of English, we divided the participants into two groups. In this experiment, the first group was given exercises to translate Mongolian relative clauses into English while the second group was given exercises to fill out or put in the correct order. In other words, we would like to check which one is effective for students to understand the complex structure of relative clauses.

The first group was taught how to translate Mongolian relative clauses into English and vice versa. During the translation practice session they had particular problems with the word order of Mongolian and English relative clauses, since with no prior practice these translation exercises were difficult for them. When they translated Mongolian relative clauses into English they made common mistakes such as omitting relative pronouns and ordering the words improperly in the sentences. During the practice session, the students were taught about the presence and absence of relative pronouns and the differences in word order in the two languages and were given tasks to translate sentences with relative clauses from Mongolian to English and vice versa. After the practice session they understood the structure dependence principle and were subsequently able to translate subject relative clauses into both languages.

The second group was also taught about the structural differences and the structure dependence principle for relative clauses. During the practice session, the students were given the task of putting given words in the correct order to make sentences. It seemed that the students could carry out the tasks easily as they understood the structure of English relative clauses. Compared to the translation practice, ordering the words properly was an easy task for them. The students had three hour practice totally for the experiment.

8 Post-test

In order to check the results of the practice sessions for both groups, the students were tested on relative clauses. Like the first experiment, the relative clause test has two tasks: translation and proper ordering of phrasal constituents. After three hour of practice, 86% of the first group could do the above two tasks successfully, while 65% of the second group could.

The first group of students could do both tasks (translation and putting the words in the correct order) more successfully than the second group of students with fewer errors as shown in (Fig. 1).

Fig. 1
figure 1

The presentation of errors in the second experiment

Full size image

Even though the grammar translation method is considered an ineffective, boring, and ‘old-fashioned’ method in communicative classrooms, it is quite useful to make learners understand the structural difference of complex sentences between source and target languages. According to the study, I hypothesize that translation is the most difficult task for language learners, but it is also the best task to improve their foreign language knowledge. It enhances better understanding of the structures of the two languages in foreign language classes as well. In other words, translation is the most effective way to practice producing L2 or foreign language sentences such as complex sentences with relative clauses.

9 Conclusions

In conclusion, our study has shown that the main differences found in relative clause structures between English and Mongolian cause one of the many difficulties that Mongolian learners face. When students recognize the differences in the internal structure of the Mongolian and English noun phrases and the use versus non-use of relative pronouns in the two languages, they easily acquire English relative clauses. It is important for both teachers and students to understand that knowledge of the structure dependency principle is required to teach and learn relative clauses that contain a subordinate clause in any languages.

If teachers analyze errors and problems of first language interference faced by the learners of foreign languages, they can find appropriate methodologies, strategies, and materials for teaching second and foreign languages. Although some scholars consider the grammar-translation method as a passive, ineffective, and boring method, I have suggested that grammar-translation is the effective way to teach and learn complex structures such as relative clauses based on my second experiment result. In other words, translation plays an important role in foreign language education in general.