Scholarly article on topic 'Wh-Constraints In Interlanguage Grammar Of Persian EFL Learners and its Implication for Teaching English as a Foreign Language'

Wh-Constraints In Interlanguage Grammar Of Persian EFL Learners and its Implication for Teaching English as a Foreign Language Academic research paper on "Languages and literature"

CC BY-NC-ND
0
0
Share paper
OECD Field of science
Keywords
{Wh-constraint / "Universal grammar" / "Sentence-matching task" / "EFL learners" / "Linguistic Knowledge."}

Abstract of research paper on Languages and literature, author of scientific article — Fariba Rahimi Esfahani

Abstract In second language (L2) studies, one direction of research is to investigate whether adult L2 learners still have access to Universal Grammar (UG) . In the related vein, this study is an indirect assessment of availability of Wh-constraints in interlanguage grammar of Persian EFL learners through an on-line sentence-matching task. To do so, 60 university students, both male and female, majoring in EFL at Khorasgan university in Iran were chosen through an Oxford Placement Test (OPT) and divided into three proficiency groups(i.e. 20 Low Intermediate, 20 High Intermediate, and 20 Advanced groups). Moreover, 10 native English speakers were chosen as the control group in this study. Both English native speakers and Persian EFL learners performed an on-line sentence-matching task. The on-line test contained 40 pairs of English sentences, 20 matching grammatical pairs which observed the UG Wh- constraint and 20 matching ungrammatical pairs which violated this Principle. In sentence-matching task students responded to two sentences on a computer screen indicating whether the two sentences were identical or not. The reaction time of EFL learners to both grammatical and ungrammatical pairs was measured and compared to that of English native speakers. The results revealed that native speakers of English can do the task faster than non-natives. Thus, Wh- constraint is not accessible to second language learners to the same extent that it is to first language learners. This study furthers our understanding of the Persian EFL learners’ performance in the area of L2 language acquisition. Moreover, the more proficient groups were faster than the less ones in sentence-matching task. As the proficiency increased, the results got much closer to the results of native speakers. Therefore, proficiency can be considered as a strong factor for UG activation.

Academic research paper on topic "Wh-Constraints In Interlanguage Grammar Of Persian EFL Learners and its Implication for Teaching English as a Foreign Language"

Available online at www.sciencedirect.com

ScienceDirect

Procedia - Social and Behavioral Sciences 192 (2015) 737 - 747

2nd GLOBAL CONFERENCE on LINGUISTICS and FOREIGN LANGUAGE TEACHING, LINELT-2014, Dubai - United Arab Emirates, December 11 - 13, 2014

Wh-Constraints In Interlanguage Grammar Of Persian EFL Learners And Its Implication For Teaching English As A Foreign

Language

Fariba Rahimi Esfahani a*

aIslamic Azad University Shahrekord Branch Iran

Abstract

In second language (L2) studies, one direction of research is to investigate whether adult L2 learners still have access to Universal Grammar (UG) .In the related vein, this study is an indirect assessment of availability of Wh-constraints in interlanguage grammar of Persian EFL learners through an on-line sentence-matching task. To do so, 60 university students, both male and female, majoring in EFL at Khorasgan university in Iran were chosen through an Oxford Placement Test (OPT) and divided into three proficiency groups(i.e. 20 Low Intermediate , 20 High Intermediate, and 20 Advanced groups). Moreover, 10 native English speakers were chosen as the control group in this study. Both English native speakers and Persian EFL learners performed an on-line sentence-matching task .The on-line test contained 40 pairs of English sentences, 20 matching grammatical pairs which observed the UG Wh- constraint and 20 matching ungrammatical pairs which violated this Principle. In sentence-matching task students responded to two sentences on a computer screen indicating whether the two sentences were identical or not. The reaction time of EFL learners to both grammatical and ungrammatical pairs was measured and compared to that of English native speakers. The results revealed that native speakers of English can do the task faster than non-natives. Thus, Wh-constraint is not accessible to second language learners to the same extent that it is to first language learners. This study furthers our understanding of the Persian EFL learners' performance in the area of L2 language acquisition. Moreover, the more proficient groups were faster than the less ones in sentence-matching task. As the proficiency increased, the results got much closer to the results of native speakers . Therefore, proficiency can be considered as a strong factor for UG activation. © 2015TheAuthors.PublishedbyElsevier Ltd.This is an open access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of Academic World Research and Education Center.

Keywords: Wh-constraint, Universal grammar, Sentence-matching task, EFL learners, Linguistic Knowledge.

* Fariba Rahimi Esfahani Tel.: +672347627843 E-mail address: rahimi_fariba@yahoo.com

1877-0428 © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of Academic World Research and Education Center. doi: 10.1016/j.sbspro.2015.06.089

1. Introduction

The evidence that native language acquisition is possible only because children are born with an innately-determined language faculty - Universal Grammar - is considerable. The evidence that the same innate ability is involved in second language acquisition (SLA) by older learners is superficially less clear. Originally, Chomsky (1980-1981) proposed and developed the theory of UG in order to account for L1 acquisition. He claimed that humans have an innate capacity for acquiring their L1. He postulated UG as an explanation of how it is that learners come to know properties of grammar that go far beyond the input; how they know that certain things are not possible, why grammars are of one sort rather than another. The claim is that such properties do not have to be learned. Proposals for an innate UG are motivated by the observation that, at least in the case of L1 acquisition, there is a mismatch between the primary linguistic data (PLD), namely the utterances a child is exposed to, and the abstract, subtle, and complex knowledge that the child acquires. In other words, the input (the PLD) underdetermines the output (the grammar).This is known as the problem of "the poverty of the stimulus" or the logical problem of language acquisition. (White, 1989)

However, since L2 learners sometimes successfully acquire aspects of the L2 different from their L1 without having received formal instruction in those aspects, some SLA researchers (White, Schachter, Bley -vroman, etc.) believe that an innate mechanism known as UG makes this phenomenon possible.

As an example of a proposed principle of UG which accounts for knowledge too subtle to be learned solely from input, I will consider the Wh-constraint (Chomsky, 1995) as a general and natural constraint which is universally applied. "Wh-constraint" is a component of the Minimalist Program which is an attempt to put flesh on old ideas: the claim that some principle of least effort is characteristic of language faculty ( Smith,1999).

There are several different notions of wh-constaint which are related to phonology, pragmatic, and syntax. The core notion of wh-constaint as it pertains to the syntax can be most simply illustrated by reference to sentences involving wh-movement. Such movement is pervasive in English but it is not unconstrained. Although we can have all of the sentences in the following example, sentence (e) is impossible:

a. I think John saw a buffalo.

b. What do you think John saw?

c. Who do you think saw a buffalo?

d. Who do you think saw what?

e. *What do you think who saw?

In a question with one WH-word (who, what), that item is typically attracted to the front of the sentence as in (b,c) to occupy the Specifier position in the CP. When two constituents are questioned, as in (d), one of the WH-words occurs at the front of the clause (who), and the other remains where it was (what). The question is why can't "what" move to the front of the clause in the above examples, why is (e) ungrammatical? A possible answer is that (e) is less economical than (d) in the sense that while both sentences contain exactly the same words, "who" starts out closer to the Specifier of CP than "what" does. Where either of two elements could move, the "Shortest Movement" condition determines that only the one which has less distance to travel is permitted to move (Neil Smith, 1999). Therefore, according to this principle, shorter moves are preferred to longer ones when a movement is required; otherwise, the outcome will be ungrammatical. Thus, wh-constaint is a derivational condition which eliminates a significant class of derivations.

In the related vein, the very vital question facing generative SLA/FLA at the present time is the availability of UG in L2 acquisition/learning. Assuming that there is indeed a logical problem of L2 acquisition, researchers asked more UG-specific questions. In the 1980s, the UG question seemed relatively straight forward (and relatively global): Is UG available (or accessible) to L2 learners? In other words, do ILGs show evidence of being constrained by principles of UG? A number of principles were investigated, such as Subjacency, the ECP and Binding Principle A. The assumption was that if you can show that a particular UG principle operates/does not operate then this generalizes to other principles, hence to UG availability/non-availability in general. As such, this study is an attempt to answer the question of whether UG is available to Persian L2 learners with specific reference to wh-constreaint in multiple wh-movements. In our study, we explore the use of reaction time in a sentence matching task as one possible supplement to grammaticality judgments in SLA research.

2. Sentence-matching: A supplement to judgment data

Several assumptions which are reasonable in native language research are unlikely to be safe in second language research. And these assumptions are precisely those upon which the use of grammaticality judgments depends. It is therefore prudent to seek additional means of probing learner's grammars. The use of reaction time in a sentence matching task can be a possible supplement to grammaticality judgments in SLA research ( Bley-vroman,1989).The sentence-matching task (SMT) is a relatively simple instrument: participants are asked to decide whether a pair of consecutively presented sentences are identical in form or not. Psycholinguistic interest in the SMT lies in the fact that in most instances grammatical pairs are matched reliably more quickly than ungrammatical ones (Duffield and Matsuo, 2007). For example, if the example the man saw the boy is displayed on a computer screen, followed by the identical the man saw the boy, a subject will be able to say quite quickly that the two sentences are identical. If, on the other hand, the subject sees Man the saw boy the the followed by the identical Man the saw boy the, it will take significantly longer to determine that the sentences match. (Bley-Vroman, 1989)

There have been two proposed explanations for this phenomenon. One plausible theory is that when the sentences are grammatical, the language processing system immediately and automatically produces a unified high level representation of the examples; and identity can be determined on the basis of comparing unitary representations at this level. When the examples are not grammatical, however, no high-level representations can be computed, and matching must be done by other less efficient strategies, for example, by a word-by-word comparison. This idea has its roots in word matching tasks in reading research. Real words are faster to match than non-word letter sequences. This explanation is that advocated by Freeman and Forster (1985). A second possible explanation is that when an example is ungrammatical, the mind constructs two representations: one of the example as is, another of a corrected version. In this second explanation it is mental correction rather than ungrammaticality per se which slows down matching time for ungrammatical examples. According to Duffield and Matsuo (2007), Whether or not this explanation is correct is perhaps less important here than the fact that - in most cases - it works, and that it provides an implicit measure of grammaticality. Clearly, the debate over these two interpretations has important implications for the theory of language processing. For our purposes, the particular explanation will not be crucial. What will be crucial is the reality of the phenomenon. Obviously, this phenomenon might be exploited to investigate grammaticality status independently from obtaining overt judgments, and might be used as a supplement to judgments. This study also uses a reaction time method in sentence-matching task to present evidence that wh-constaint in Multiple wh-questions which operates in L1(English)is also available to Persian learners of English while working with multiple wh-questions.

3.Background

A goal that L2 researchers have set for themselves is the elicitation of learner behavior from which they can infer certain knowledge of the target language. In other words, they investigate the knowledge of a UG principle in L2. According to Cook and Newson (1996),the question of whether L2 learners have access to UG has been perhaps the main topic of research among those interested in applying principles and parameters theory to second language acquisition. What could be the role of UG in L2 learning? Cook (1985) put it as a choice between three possibilities. One is that L2 learners start from scratch; they have direct access to UG and are uninfluenced by the L1.The second one is that, they start from their knowledge of the first language; they have indirect access to UG via the L1.The third one is that, they do not treat the L2 as a language at all; they have no access to UG and learn the L2 without its help. Flynn (2002) also presents three hypotheses to explain role of UG in SLA: "No Access" Hypothesis: UG is totally inaccessible to the adult L2 learner; learning takes place in terms of non-linguistic learning strategies, "Partial Access" Hypothesis: UG is partially available to the learner; only those parametric values characterizing the L1 grammar are available, the rest must be learnt in terms of non-linguistic learning strategies. Finally, in "Full Access" Hypothesis, UG is fully available; differences in patterns of acquisition between L1 and L2 learners and the lack of completeness can be accounted for in other ways.

In sum, Chomsky proposed the theory of Universal Grammar to account for the fact that children know properties of grammar that goes far beyond the input and they know far more than PLD (utterances a child is exposed to) has provided them. Since L2 learners seem to know more about the L2 than they were taught, some researchers suspect UG's role in SLA. The field is relatively new and does not have a set answer. Many researchers in the field feel that more studies in different settings and different languages with various UG principles should be done. The claim that UG is available in SLA is a very attractive one and there is a way to test L2 learner's intuitions

by which one can measure their knowledge of UG principle, such as grammaticality judgment task that many researchers have adopted. The present work is one such study aimed to contribute to the development of the field. Before discussing in detail, it makes perfect sense to turn our attention to wh-constraint in multiple wh-questions as a principle of UG, which is the focus of this study.

4. Theoretical Background

4.1 Movement

An important syntactic point within WH-questions is movement. Movement is a process in generative grammar that was formulated from the early 1970s. In fact, movement is a syntactic operation by which constituents move and leave behind silent trace copies. In 2000s, traces, which came into existence as the result of movement, were reformulated in terms of copying according to which units in one position were duplicated in another (Matthews, 2007).Furthermore, movement is characterized in terms of chains and is constrained by the principle of economy (i.e. "move as few constituents as possible the shortest distance possible") (Radford et al. 2009, p.301).

Movement takes place both overtly (i.e. before spell-out) and covertly (i.e. after spell-out at LF). Different movement operations are triggered by different causes. For example, feature checking causes wh-movement to occur (Radford,2004).Noting that a variety of movement types exist, Uriagerika (1998) argued that it is the strength of movement-requiring morphemes that serves as a distinguishing factor among languages and not the movement types.

4.2 Multiple Wh-questions

In English multiples, Wh-in-situ is merely allowed in those structures where one wh-phrase has been moved (Ambar, 2001). In fact it is believed that in English only one wh-phrase undergoes fronting. According to Attract Closest Principle (Radford, 2004), a head which attracts a given kind of constituent attracts the closest constituent of the relevant kind. Following this principle, the movement of wh-words in multiple wh-questions in English is restricted to the wh-word which is closer to the beginning of the sentence. Therefore, according to the Principle of Economy, short moves are more economical than longer ones. Long moves over shorter ones make the sentence less economical and hence ungrammatical.

As for Persian, overt optional wh-fronting exists, and wh-in-situ is also allowed; in addition, scrambling is an integral part of Persian. In an experimental study, (Lotfi, 2003b) argued that pair-listing interpretation is always seen in Persian multiples due to the fact that all wh-features by no means move to C. Moreover, he asserted that Persian overt wh-fronting shows both superiority effects and superiority violations on an ad hoc basis, and that the existence of /ra/ is of great assistance to Persian due to the fact that its presence can eliminate superiority violations and help scrambling.

4.3 Island constraints

(Chomsky, 1964) observed that there was a restriction on the family of displacement operations collectively known as A-bar movement or extraction . Ross(1967) suggested that this phenomenon is part of a larger pattern of constraints on movement that limit movement from certain kinds of constituents. In Ross's terminology, a syntactic constituent that disallows movement from within itself is known as ' island constraints.' In addition to Ross's island, a number of other island types have been proposed, including negative islands (Schafer, 1995) and Wh-islands( Chomsky, 1973). The latter are perhaps the most important kind of island.

In the earliest version of the minimalist program (Chomsky, 1994), island effects were captured by making reference to a version of relativized minimality, whereby there are economy constraints that compare derivations of sentences and look for the most economical. One such constraint is shortest Move( SM).This constraint gives preference to derivations where each movement moves to the closest potential landing site.(Carnie, 2006)

By the "shortest move", apart from the movement of the shortest distance, one could mean the fewest possible steps as well (Chomsky, 1988). Moreover, a movement should not happen unless there is really a "necessity". NP-movement, for instance, is triggered by the need to check the morphological features of the NP. If those features are

not checked and erased, the output representation would not be licensed or grammatical. Only when there is a need to check features, rising occurs. Once licensed, there should not be any movement. (Economy in Generative Grammar: 125).

Therefore, according to the derivational economy proposed in MP, derivations must be as economical as possible. The more economical derivations block the less economical ones. There are several instantiation of the principles of Economy: Ross's island constraint( 1967), Wh-islands ( Chomsky, 1973), Minimal Link Condition (Chomsky, 1995), Rizzi's Relativized Minimality(1990), Subjacency in wh-island constructions, Superiority effect, Scrambling, Attract closest Principle(Radford 2004). According to these principles, movement is a constrained operation and needs to be kept local. Short moves are preferred to longer ones and are more grammatical.

One constraint on wh-movement which can be subsumed under economy principle is Wh- constraint. Radford (2004) talks about Wh- constraint which specifies that wh-clauses (i.e. clauses beginning with a wh-expression) are islands, so that no constituents can be moved out of a wh-clause. Consider the ungrammaticality of the following sentence:

*He is someone [who nobody knows [what the FBA did to]]

Radford (2004) explains the ungrammaticality of the above sentence due to the fact that moving 'who' out of the 'did clause' will lead to the violation of Chomsky's wh-islands. He states that, the 'did clause' is a wh-clause (by virtue of containing the preposed wh-word ' what') and since wh-clauses are islands, movement of the relative pronoun out of the 'did clause' will lead to the violation of the Wh-island .

5. Empirical background

Various types of research are conducted to show whether UG access in L2 is done directly or via L1.To demonstrate whether the resemblance between the ways in which UG operates in both L1 and L2 can be of any positive effect to L2 learners, Johnson (1988) conducted a study on adult native speakers of Spanish; a language where UG principles operate similarly as they do in English. The results show that Spanish speakers did not show native-like success in observing Subjacency in English despite the fact that the two languages are similar in this regard.

In a research study on Chinese learners of English, (Hawkins, 1997) found out that speakers of Chinese (a language without wh-operator movement in overt syntax) learning second language English (a language with wh-operator movement in overt syntax) establish mental representations for English which involve pronominal binding rather than operator movement. It will be suggested that this divergence from native-speaker representations is an effect of the inaccessibility of features of functional categories in second language acquisition, what he will refer to as the 'failed functional features hypotheses.

In one study, (Chen and Haung 2006) examined the acquisition of subjacency and Empty Category Principle by English and Japanese speakers of Chinese. The participants were forty intermediate students of Mandarin Training Centre National Taiwan Normal University: half were English native speakers and half Japanese. In addition there were 20 native controls. Two tasks (i.e. a preference and an ordering task) were designed on the basis of the following properties concerning subjacency and Empty Category Principle: Wh-island constraint, complex NP constraint, sentential subject constraint, and superiority effect. The results show that, except for the superiority effect, neither group of L2 learners carried their L1 knowledge to acquire subjacency and ECP, suggesting that L1 influence is not significant. Furthermore, it was found that Japanese speakers did not perform significantly better than English speakers. This shows that Universal grammar is still available, since our participants have reset their L1 parameters to proper L2 values. In addition, among these features our participants did less well on non-superiority, and the native controls rejected island violations more strongly than either of L2 participants. Finally Subjacency and the ECP were found equally easy for our participants to acquire.

In an experimental study upon the acquisition of English restrictive relative clauses (RRCs) by native speakers of Cantonese , (Hawkins & Chan's, 1997) tested L1 speakers of Cantonese at three proficiency levels in English and compared them with native speaker controls. Using a grammaticality judgment task (59 relevant items out of a total of 101 sentences) involving 20 grammatical and 39 ungrammatical RRCs ,Hawkins and Chan tested L1 speakers of Cantonese at 3 proficiency levels in English, and compared them with native speaker controls. On the basis of the results, Cantonese speakers of English appear, as proficiency increases, to approximate to the target language. The advanced speakers accept grammatical RRCs, and reject doubly-filled C and resumptive pronouns. This might lead

one to the conclusion that they have established a movement operation for RRCs in their ILGs. However, some results are problematic for this view. Cantonese speakers appear to become less target-like with proficiency in detecting the grammaticality of violations of 'shortest move'. The elementary speakers apparently know about constraints on wh-movement while advanced proficiency speakers apparently allow a high proportion of such violations.

In another study, White, (Travis and Maclachlan, 1992) conducted research upon the constraints on English wh-movements by Malagasy learners of English. The learners observed the constraints on English wh-movements despite the fact that the constraints were quite different in the two languages. Using a GJT and a Written Elicited Production Task to tap the learners' knowledge of grammatical wh-question formation in English, the researchers found that almost all the high-intermediate group and half of the low-intermediate group perform like the native English speaking control group; hence, the L1 is not the only source of the learners' UG-like knowledge. Rather, principles of UG remain accessible in adult L2 acquisition, and can be triggered properly by increasing the level of general proficiency (White, 1996).

Finally, (Rahimi Esfahani, 2009) investigated the availability of UG in Persian learners of English in different contexts using a grammaticality judgment task. To achieve this end, two groups of Iranian learners, one group studying English in Iran and one in the United States at the time of study, were tested on the Minimal Link Condition in English. There was also a control group, composed of native speakers of English. The Iran group consisted of 30 students, both male and female, all of whom were majoring in EFL in Khorasgan University. The subjects in this group were chosen based on their TOEFL scores. The US group consisted of 25 subjects, both male and female, majoring in six different fields in the United States. They were asked to report their TOEFL scores (they were more proficient than the Iran group based on their TOEFL scores). Then a Universal Grammar (UG) MLC test containing 40 items was administered to the students in Iran group. Besides, the MLC test was sent to the Iranian students in US and also to 15 English native speakers via E-mail. The results revealed that the performance of the US group was significantly different from that of the Iran group. Moreover, both groups differed from native speakers of English significantly in terms of their grammaticality judgment. These results suggest that UG does play a role in SLA but not to the same extent as it does in first language acquisition, and the poorer performance of the students in Iran (Iran group) is due to the need for a process of parameter resetting.

All of the above studies were based on grammaticality judgments. As (Bley-vroman, 1989) says, several assumptions which are reasonable in native language research are unlikely to be safe in second language research and these assumptions are precisely those upon which the use of grammaticality judgments depends. It is therefore prudent to seek additional means of probing learner's grammars. In this study, we investigated the availability of UG in L2 learners via reaction time method in sentence-matching task instead of grammaticality judgments. Moreover, the role of L2 proficiency was examined in performing sentence-matching task and the results were compared with native speakers of English to see if UG is available in the same way in both L1 and L2.

6. Research Questions

To investigate whether the wh- constraint in multiple wh-questions is available to Persian learners of English, the following questions were raised:

Q1. How do native speakers of English differ from L2 learners in Reaction Time Method in sentence-matching task?

Q2. What is the effect of L2 proficiency on the reaction time of students in sentence-matching task?

Q3. What is the effect of grammaticality type on the reaction time of students in sentence-matching task? ( Will the

ungrammatical examples take both L2 learners and native English speakers longer to match than the grammatical

examples?)

7. Hypotheses

On the basis of the above questions, our hypothesis would be as follows: H1. There is a significant effect for native language, that is native speakers of English should be able to do the task faster than non-natives (via sentence matching task).

H2. The more proficient learners are faster than the less ones in sentence-matching tasks.

H3. There is a significant effect for grammaticality type, which is the ungrammatical examples will take both L2

learners and native English speakers longer to match than the grammatical examples.

8. Method

8.1 Participants

The subjects for this study were three groups of university students, both male and female, majoring in EFL at Khorasgan university, Esfahan, Iran..60 students were chosen from 100 applicants through an Oxford Placement Test (OPT) as the criterion for their selection and division into three proficiency groups(i.e. 20 Low Intermediate , 20 High Intermediate, and 20 Advanced groups) based on their scores which fell into the ranges of 80-135 for Low Intermediate Level , 135-150 for High Intermediate Level, and 150-170 for Advanced levels. Moreover, 10 native English speakers were chosen from University of Sydney as the control group.

8.2 Material

The materials consisted of a task which was administered to three groups of university students and a group of English native speaker to tap their knowledge of forming multiple wh-questions in English. The task was in the form of sentence-matching task to examine the L2 learners' interpretation of English multiple wh-movements. It was concerned with examining the effects of L2 proficiency and grammaticality type. We constructed a test instrument of 40 items, half of which were matching pairs and half of which did not match. In fact, we were only interested in the times for the matching pairs: the others were included to make the task realistic. Moreover, we used The Authorware Software (2005), SAT-based software to measure the response latencies (in milliseconds) of the subjects for both grammatical and ungrammatical matching pairs.

8.3 Design and Procedures

Upon being administered the grammatical component of the OPT, the EFL participants under study were assigned to three proficiency-based groups in accordance with their proficiency scores, namely, Low Intermediate , High Intermediate, and Advanced groups. The required data were gathered by administering the afore-mentioned sentence-matching task to both Persian EFL participants and native speakers of English and recording their response latencies via the authorware software.

The experiment was administered on personal computers. First, a set of instructions and example sentences was displayed, and then a trial set of 3 items, then the actual test instrument. A given test item is presented as follows:

1. The first of the pair of sentences is displayed in the upper center of the screen, with ordinary capitalization (not all caps).

2. After a short delay, the second of the pair appears about one inch beneath it, and offset about one inch to the right. Now both sentences are visible on the screen.

3. If the sentences match, the subject is to press the "m" key on the keyboard; if they do not match, the "n" key is to be pressed.

4. After the key is pressed, the subject's response is echoed at the bottom of the screen.

5. At the bottom of the screen, the words "Press SPACE-BAR for next item" now appear. The subject presses the space bar when ready.

9. Results and Discussion

After administering the sentence-matching task to the three L2 groups and English native speakers, the required response times were obtained. For each subject we computed a mean response time for the 20 matching pairs in the task. Table 1 shows the mean response latencies of subjects in all four groups. As expected, the native speakers respond much more rapidly than the L2 groups. Moreover, as proficiency increases, the reaction time of subjects decreases.

Fariba Rahimi Esfahani / Procedia - Social and Behavioral Sciences 192 (2015) 737 - 747 Table 1. Descriptive Statistics

N Std. Deviation Mean Proficiency

10 20 20 20 125.99848 111.75904 156.02990 87.56297 1351.0000 3100.0000 3444.2500 4823.2500 Native Speaker Group Advanced Group High Intermediate Group Low Intermediate Group

70 1014.91457 3589.7143 Total

Table2. ANOVA

Sig. F Mean Square df Sum of Squares

.000 1321.956 23811532.262 3 71434596.78 6 1329597.500 72764194.28 6 Between Groups

20145.417 66 69 Within Groups Total

Dependent Variable: 4.00 Scheffe

Table3. Multiple Comparisons

(I) Proficiency (J) Proficiency

Mean Difference (I-J) Std. Error

95% Confidence Interval

Native Speaker Group

Advanced Group

High Intermediate Group

Low Intermediate Group

Advanced Group High Intermediate Group

Low Intermediate Group Native Speaker Group High Intermediate Group

Low Intermediate Group Native Speaker Group

Advanced Group Low Intermediate Group

Native Speaker Group

Advanced Group High Intermediate Group

-1445.00000(*) -1803.25000(*)

3152.25000(*) 1445.00000(*) 358.25000(*)

1707.25000(*) 1803.25000(*)

358.25000(*) 1349.00000(*)

3152.25000(*)

1707.25000(*) 1349.00000(*)

54.97102 54.97102

.000 .000

Upper Bound

-1602.7116 -1960.9616

Lower Bound

-1287.2884 -1645.5384

54.97102 .000 -3309.9616 -2994.5384 54.97102 .000 1287.2884 1602.7116 44.88365 .000 -487.0210 -229.4790

44.88365 .000 -1836.0210 -1578.4790 54.97102 .000 1645.5384 1960.9616

44.88365 .000 229.4790 487.0210

44.88365 .000 -1477.7710 -1220.2290

54.97102 .000 2994.5384 3309.9616

44.88365 .000 1578.4790 1836.0210

44.88365 .000 1220.2290 1477.7710

* The mean difference is significant at the .05 level.

As table2 shows, the significance is .000. This is smaller than .05 and .01, so the differences between the groups are significant. The results of the Schefe test (table 3) show that the differences between native speakers and all L2 groups are significantly different. Moreover, the differences between L2 learners with different proficiency levels are also significant. The more proficient learners did the task significantly faster than the less proficient ones. Therefore, Hypotheses land 2 are confirmed. That is, native speakers of English can do the task faster than non-natives and the more proficient learners are faster than the less proficient ones in sentence-matching task.

Moreover, we computed the mean response time of the grammatical and ungrammatical matching pairs

separately for each subject to see the effect of grammatically type on the reaction time of the students. To compare the response latency means of the 4 groups of participants separately on grammatical and ungrammatical sentences in sentence-matching task, we used 4 matched t-tests. As the results of T-tests show, the difference between the reaction times of the students in grammatical and ungrammatical sentences in all four groups was significant. In all four groups, the matching task took significantly longer when the matching pairs were ungrammatical sentences violating the Wh-constraint principle of UG.

Table4.Paired Samples Test (advanced group)

Paired Samples Test

Paired Differences t df Sig. (2-tailed)

Mean Std. Deviation Std. Error Mean 95% C onfidence Interv al of the Difference

Lower Upper

Pair GRAMATICAL REACTION 1 TIME - UNGRAMATICAL REACTION TIME -909.000 167.83137 37.52823 -987.547 -830.453 -24.222 19 .000

Table5. Paired Samples Test (high intermediate group)

Paired Samples Test

Paired Differences t df Sig. (2-tailed)

Mean Std. Deviation Std. Error Mean 95% C onfidence Interv al of the Difference

Lower U pper

Pair GRAMATICAL REACTION 1 TIME - UNGRAMATICAL REACTION TIME -972.500 309.53063 69.21315 -1117.36 -827.635 -14.051 19 .000

Table6. Paired Samples Test (low intermediate group)

Paired Samples Test

Paired Differences t df Sig. (2-tailed)

Mean Std. Deviation Std. Error Mean 95% C onfidence Interv al of the Difference

Lower U pper

Pair GRAMATICAL REACTION 1 TIME - UNGRAMATICAL REACTION TIME -1746.50 237.02709 53.00087 -1857.43 -1635.57 -32.952 19 .000

Table7. Paired Samples Test (native group)

Paired Samples Test

Paired Differences t df Sig. (2-tailed)

Mean Std. Deviation S td. E rror Mean 95% C onfidence Interval of the Difference

Lower Upper

Pair GRAMATICAL REACTION 1 TIME - UNGRAMATICAL REAC TION TIME -134.000 80.99383 25.61250 -191.939 -76.06051 -5.232 9 .001

As the results of t-tests show, the grammatical sentences were matched faster than the ungrammatical ones in all L2 groups and in native speakers. This provides a way of determining the availability of structural representation that is UG wh-constraint principle in L2 learners. The idea behind this is that the presence of structure in the stimuli facilitates the same/different decision. In general, a subject's reaction time to a particular sentence pair can be taken to be a function of its grammaticality. Grammatical sentences can be matched faster than the ungrammatical ones. Therefore, Wh-constraint Principle is available to Persian learners of English, since subjects' reaction times to grammatical sentences were faster than ungrammatical ones. Moreover, the result of ANOVA shows that all L2 learners were significantly different from native speakers in their response latencies. Also, they were different from each other: The more proficient learners were faster. This indicates that Wh-constraint principle is not available to L2 learners to the same extent that it is to native speakers.

These results can doubt on the theory of direct access to UG. The L2 groups of this study did not have the same access to Wh-constraint principle and UG, in general, as native speakers of English since the mean response latencies appeared to be significantly different. In other words, Persian learners of English do not possess the same linguistic competence as L1 adults. Consequently, the results to some extent tend to support the indirect access to UG in FLA. We can claim that learning is not so directly tight with UG since the reaction time of L2 learners differed from native speakers. Therefore, the results of this study do not reject the whole idea that Persian learners of English do not access Wh-constraint principle at all. What it claims is that its extent is rather a variant.

Moreover, 'Proficiency' should not be overlooked easily in l2 since the performance of L2 groups with different proficiency levels were different in sentence-matching task .As the proficiency increases, the results get much closer to the results of native speakers. The findings of this study reject the claim that Persian learners of English should unconsciously realize wh-constraint principle as soon as they discover multiple wh-movements in English.

The results of this study support previous findings that suggest that UG does play a role in SLA but not to the same extent as it does in first language acquisition. Rahimi Esfahani (2009), White, Travis & Maclachlan (1992), and Hawkins & Chan's (1997)). Moreover, the findings support those claims that as the proficiency increases, the learners' interlanguage becomes closer to the native speakers' competence.

This research also indicates that Wh- constraint Principle could be confounded by proficiency. This implies the feasibility of UG theory, by itself, should not be overemphasized; teachers or instructors should not leave students in the hope that they figure out principles or conditions by themselves. Rather, they should focus on 'how to teach' to set the right parameters in L2 teaching.

Reference

Ambar,M.(2001). Wh-Asymmetries. (pp.209-249) &Universidade de Lisboa.

Bley-Vroman, R. W., Felix, S., & Ioup, G. L. (1988). The accessibility of universal grammar in adult language learning. Second Language Research, 4, 1-32.

Bley-Vroman, R ., & Masterson,D.(1989).Reaction time as a supplement to grammaticality judgments in the investigation of second language

learners' competence. University of Hawai'i Working Papers in ESL, 8, 207-237. Carnie, A. (2006). Island constraints" in Encyclopedia of Language and Linguistics,edited by Keith Brown.Second Edition. 6, 48-51. Oxford:Elsevir.

Chen,C.-Y. Doris and H.-Y. Candy Haung.( 2006).Is it difficult to acquire Subjacency and the ECP? Taiwan Journal of Linguistics, vol.4.1, 89130.

Chomsky, N. (1972). Language and mind. NY: Harcourt, Brace and World. Chomsky, N.(1995). The Minimalist Program. MIT Press.

Cook, V. J. (1985). Chomsky's universal grammar and second language learning. Applied Linguistics, 6, 1-18. Cook, V. J. (1991). Second language learning and language teaching. London: Edward Arnold.

Duffield,N., Matsuo,A. ,& Roberts,L.(2007). Acceptable ungrammaticality in sentence matching. Second Language Research 23,2); pp. 155-177 Flynn, S.& Lust, B. (2002). "A minimalist Approach to L2 solves a Dilemma of UG" in Portraits oof the L2 User, Clevedon, Multilingual Matters, ch.4, p.93-120.

Lotfi,A.R.(2003b).Persian wh-riddle.In Boecks,C. and K. Grohmann(eds.)Multiple wh-fronting,John Benjamins Matthews, P.H. (2007). Oxford concise dictionary of linguistics (2nd ed.). Oxford: Oxford University Press.

Muller, Gereon, (1997) Order preservation, Parallel movement, and the Emergence of the Unmarked. www.uni-del~muellerg/mu19.pdf Radford, A. (2004). English Syntax: An Introduction. Cambridge: Cambridge University Press.

Radford, A., Atkinson, M., Britain, D., Clahsen, H., & Spencer, A. (2009). Linguistics: An introduction (2nd ed.).Cambridge: Cambridge University Press.

Rahimi Esfahani, F.(2009). Minimal Link Condition and UG Operation in SecondLanguage Acquisition. The International Journal of Learning. 16(11),47-62.

Schachter, J. (1989). Second language acquisition and its relationship to UG. . Applied Linguistics, 9, 219-235.

Smith, Neil, (1999). Chomsky Ideas and Ideals. Cambridge: Cambridge University Press.

Uriagereka, J. (1998). Rhyme and Reason: An Introduction to Minimalist Syntax.

White, L. (1989). Universal grammar and second language acquisition. Amsterdam: Benjamins.

White, L. (1996). Universal Grammar and Second Language Acquisition: Current trends and new directions. In: n.d. (eds.), Handbook of Second Language Acquisition (pp: 85-120). Academic Press, Inc.