Cross-linguistic influence: the role of L1 and L2 differences in acquiring second language

. This article discusses the role of first language and second language differences in second language acquisition process of L2 learners. This small scale research focuses on finding answers to three main questions: a) what are the main reasons for transferring L1 knowledge into L2; b) what should learners do to deal with this problem; and c) when does transfer start to diminish. The participants of this study are 21 students who are studying in their eleventh grade. According to the analysis of results, (1) main reasons for language transfer are different language structures and not enough knowledge in target language (knowledge gap in the target language) these, in turn, can slow down language learning pace of learners; to deal with the problem (2) learners should devote much time to learning (analyzing speeches, listening to podcasts, reading authentic articles, communicating with native speakers) and expose to L2 as much as possible, finally, (3) the influence of their first language has diminished significantly after time passed, but it didn’t disappear fully.


Introduction
Cross-linguistic influence in second language learners (SLL) is an area of study that has received considerable attention in recent years. Some authors call it transfer while others use it as it is. There is also another older term for this phenomenon -"interference". All these three linguistic terms mean the same, but the former two are dominant in contemporary SLA discourse. Cross-linguistic influence (CLI) refers to the influence that knowledge of one language can influence another within an individual learner. It usually includes two or more languages that can affect one another in a bilingual or monolingual learner.
Obviously, all L2 learners possess L1 knowledge or knowledge of other languages when they start learning a new language. This previous knowledge is a significant source of influence on L2 acquisition which can hinder sometimes L2 development. However, as well as its negative influences, first language knowledge can have a positive impact on L2 learning according to the study carried out by Jarvis [1]. The impact may vary for different learner L1 backgrounds, at different stages of acquisition and in different areas of L2. For example, new learners can face difficulties in grammar because of differences between source language and target languages' grammar structures.
This small scale research focused on cross-linguistic influence theory in SLA tries to answer the following questions: (1) what are the main reasons for transferring L1 knowledge into L2; (2) what should learners do to deal with this problem; and (3) when does transfer start to diminish? So far, a number of studies have been carried out on this topic and most of them have focused on Romance-Germanic language combinations. However, this study is devoted to examine Uzbek and English language pair which has radically different grammar structures by analyzing English language learning process of a monolingual Uzbek learner. It only focuses on differences in question forming of both languages and this is the limitation of this study.
It is obvious that all learners try to find correspondences of what they are learning newly in their first language. They start to build up new knowledge on their previously gained one or on what they already know. It goes with comparing and contrasting the two languages' peculiarities. To understand why learners tend to transfer L1 features into target language we need to go beyond L1-L2 correspondences. According to Ortega [2], a better understanding of the forces that shape cross-linguistic influence rests not with external language comparisons, but with learners' psychological perceptions of those L1-L2 similarities or differences. At first, they do not know what can be transferred or what cannot be.
What can be reasons for transferring knowledge? Sometimes it can happen because of too much reliance on L1 knowledge or it can simple occur when there is a gap in L2 knowledge. Singleton (1987) explored some cases in which a learner committed transfer errors. The participant of the study was Philip -a professional L1 English writer who had picked up French during three short visits to France. In three half-hour interviews in French, 154 transfer errors were identified. When reviewing the interview and commenting on errors, Philip said: "I knew that it probably wasn't right, but it was the nearest I could get to something that might be right" [6]. It confirms that if a learner cannot express himself/herself in target language, the learner tends to rely on source language.
In a series of seminal publications, Eric Kellerman introduced an important notion in the study of L1 influences -transferability. He stated that at the beginning stage of proficiency, young learners who has less instruction and metalinguistic knowledge seem to be unconcerned about the distinctions of morphological features (the examples of transitive and intransitive verbs). Initially, they rely on L1 knowledge and then as they develop a more sophisticated knowledge of the language and they are able to develop 'a sensitivity to a pragmatic distinction (implicitly known)' [7].
There is evidence that L1 transfer can impact the rate of learners' progress in target language development. It was first formulated by Zobl (1982). He proposed that L1-L2 differences account for the pace or rate at which certain morpho-syntactic structures will be learned by different L1 groups [2].
On the other hand, some authors claim that the knowledge of L1 can often have a positive impact on the rate of L2 learning. Hakan Ringbom (1985) has strongly advocated for this idea. He noted that relevant knowledge in L1 can accelerate the rate of L2 learning. His research has focused on the comparison of learning processes of Finnish and Swedish learners who are acquiring English. The results suggest that Swedes have a great advantage to learn English over Finns as both languages (Swedish and English) are from the same language family and they share many typological features. In contrast, Finnish belongs to a different language family, and thus it is unrelated and typologically more different from English. The similarities can account for the overall advantage of Swedes in learning English [8].
The participants of this study are 21 students who are studying in the eleventh grade at public school. Thirteen participants come from Uzbek monolingual families, whereas the other eight are bilingual learners speaking tajik at home for daily communication and chatting with friends. Mode of instruction of the school they all are studying is Uzbek, and all of them are now taking extra classes of English language in education center. Most students' aim for learning English is studying abroad. When the study was started, they had been learning English for three months.
Considering learner interests, their age, and personality is very important while choosing teaching methods that match their personality. Majority of students said that they use different social networking sites daily to communicate with people from around the world. They like communicating with people both lively and online via social networking sites. When I started observing them, I realized that they try to take all chances for speaking English even though there are detectable errors in their speeches. They are not afraid of taking risks and making mistakes which is very good while learning a new language. Moreover, they do not feel shy, embarrassed, or inhibited in front of strangers. As well as their eagerness to learn language, most of them have good long-term memory which can fasten the process of L2 learning. Speculating all about their character and good knowledge of L1, the teacher decided to teach them through interactive methods focusing on communication, using videos, TV shows, English movies, and podcasts. Research design for this small scale study is as following:  pre-interview to collect general information about the participants.  pre-test on the chosen topic.  post-test.
Because of the participants' lack of knowledge in L2, pre-interview was conducted in their first language (Uzbek). It has 5 questions which were aimed to collect data about the participants' current level, their personality, individual differences/variables and, of course, their enthusiasm to acquire a language (Appendix 1). After collecting data about their personality, L1 knowledge, and L2 experience. The author of the study has decided to design a pre-test to know what is problematic for the students in experiment, and how much they are aware of the topic in target language. The test includes only special and general questions. The reason why only these types are included is that forms are totally different from their L1.
As their performance was somehow poor, we put emphasis on this topic to teach and in the third step, after 3 week's preparation and great hard work, they were given a test to check their comprehension. The test consists of 12 questions in which we can see samples of different question types in English (Appendix 3). Moreover, at the end of the process, the participants were interviewed one more time how they feel about their own performance.
When they came first to take extra classes, their English was not very good, and they spoke with mistakes such as wrong word choice, incomplete sentence structures, bad pronunciation and so on. Therefore, pre-interview was decided to be conducted in their first language -Uzbek. During the interview they were asked 5 questions about their personality, learning style preferences, L1 and L2 knowledge as well as their main purpose for learning English language.
According to the results, most of them are sociable in nature and they like asking questions about what they do not know which means they are active learners, not passive recipients. They also answered the question about their L1 grammar knowledge saying that they know it proficiently as nearly all of them went Uzbek schools. Moreover, they are very comfortable with expressing themselves in front of others they do not know well. It is the answer to the next question to know about their personality traits. As they need to learn English faster, the participants said that they put much effort and devote much time every day to learn English since it is their only priority now.
After collecting data about their L1 (proficient) and L2 (somewhere between beginner and elementary) knowledge, the process started. It is obvious that questions are very important for people when they need information, and it is one of the functions (heuristic) of any language according to Halliday. Each language has their own peculiarities on each topic. The same is true with question forming. For example, English is a member of the Indo-European language family and it follows strict SVO structures while making sentences. However, in interrogatives the subject and the verb may be inverted, placing the subject in second position following the verb: What is your favorite color? When it comes to Uzbek, it is a member of Turkic language family, which is of agglutinative nature which means that words are made up of a linear sequence of distinct morphemes and each component of meaning is represented by its own morpheme. Typical word order in Uzbek affirmative sentences is SOV: Men maktabga boraman (I go to school). But sometimes subjects can be omitted as the person and number can be realized from the morpheme of the verb-predicate: maktabga boraman. (I go to school) =>Bor (root morpheme) + -a (suffix for representing future tense or regularity) + -man (first person singular). As for questions, inversion does not occur at all, it is realized by only adding the morpheme -mi at the end of the predicate in general questions (Yes/No questions): Sen maktabga borasanmi? (Do you go to school?) As it is seen, subject and predicate are not inverted to form a question like we normally do in English general question types.
The differences in L1 and L2 discussed above can have a negative influence on students' language learning process. In this study learners' case, they tend to rely on their L1 knowledge and when they were given a test on questions (question forming), they demonstrated very poor results as it can be seen in figure 1 below. The pre-test consists of 8 questions. While designing the test, questions were kept simple because they are not aware of English language (word choice, sentence structures, auxiliary and modal verbs and so on) at the time of testing. Having found 3 correct answers on average out of 8 questions, their results were only 37% which is very low (Figure 1).

Fig. 1. Pre-test results
Analyzing their performance, the results show that they did not much pay attention to inversion (questions 1 and 3) in question forming of English. Besides this, they relied on their L1 knowledge and grammar structure while doing the test. Most participants said that "I translated the questions word by word and then I chose the appropriate one", which are

RIGHT ANSWERS WRONG ANSWERS
the examples of questions 6 and 7. They did not pay attention to modal verb to invert (question 2). It seems that those right answers are likely to come from their communication experience with native speakers via social networking sites. After results were taken, they started to be instructed about what is inversion, when it is used and how we use it while forming questions in English. For example, auxiliary verbs and modals should be put before subject in each type of question regardless of special, general, or alternative questions: General questions: -Does Jane know about you new job? (Yes, she does/No she does not) -Can I call my sister? (Yes, you can/No, you cannot) Special or Wh-questions: -When did you come here? (I came here two weeks ago) -How can I go to the hospital? (You can go by train) Alternative questions: -Does she like ice cream or sweets? -She likes ice-cream. Apart from question tags which is a bit different in structure: -She sent him an invitation, didn't she? -Yes, she did.
After being taught, they came to realize that Uzbek and English have different structures and rules for making questions. They did some grammar exercises first and then as they love watching movies, videos and listening to songs and podcasts, we turned to analyze the speeches of native speakers in videos and podcasts. At home, they devoted 4 to 5 hours for English and did listening tasks on the topic. In classes which were conducted daily, dialogues were included as well. The students practiced role plays, dialogues about the topics of interest using different verb tenses to know how it is like to make questions with auxiliary verbs in tenses. (Main purpose is forming question types correctly and rather fast). Practice really helped the learners, and they started making questions fast and correctly ( Figure 2).

Fig. 2. Post -test results
Having had much practice on the topic both at home and in classes, they felt that they already acquired the required knowledge and able to use it correctly in both speaking and writing. After 3 weeks, they were tested again by giving two option questions to check their comprehension and gained knowledge. It is a post-test part of this small-scale study, and the test has 12 questions in total and time is limited (5 minutes only) now as the participants are aware of the topic. Moreover, it has a mixture of questions. When I checked the test, the results showed higher scores with at least 8 and 9 correct answers, which is better, still not perfect. The figure 2 shows the results in percentage. Since it is not very high, I came to realize that they still confuse about L1 and target language differences. Again, they did not pay enough attention to modals and auxiliary verbs and their person (singularity or plurality). However, the best part is that they did not do any inversion mistakes in forming questions this time, which is really satisfying.

Conclusion
Each learner passes their own path of learning language. They may face some difficulties. Of course, it is normal, and it means the learner is acquiring something new. The participants of this study are very enthusiastic about learning English. It is somehow challenging for them as most of them started learning after puberty. However, the motivation they have never ends as I know them now for a while. It can play a major role in SLA process.
According to previous studies on Cross-linguistic influence, the differences between L1 and L2 can account for slowing down the pace in which a learner is acquiring a new language. The case is true with these learners as well. Even though they are motivated and devote much time to learning the pace seems quite slower. However, after time passes it may disappear with hard work and much practice.
The study put three main questions at the beginning of the research and through the investigation of learning processes of 21 students, I have revealed that (1) main reasons for language transfer are different language structures and not enough knowledge in target language (knowledge gap in the target language), these, in turn, can slow down language learning pace of learners; to deal with the problem (2) learners should devote much time to learning (analyzing speeches, listening to podcasts, reading authentic articles, communicating with native speakers) and expose to L2 as much as possible, finally, (3) according to participants' results, the influence of their first language has diminished significantly after time passed, but it didn't disappear fully.
The result is in concord with the ones of previous studies mentioned in the literature review part and for further research, identifying and analyzing positive impact of CLI in SLA process can be a good topic to go deeper.