Exploring the cross-linguistic functioning of the Principles of WH-Movements: The case of Pakistani ESL learners

Abstract

This is a cross-linguistic study conducted with Pakistani learners of English in the classroom setting. This study aimed to explore whether the UG Principles of WH-movement supported or restricted Urdu-speaking learners in forming direct WH-questions in English. For data collection, a sample of 260 students was selected using convenience sampling from ten (10) private and four (04) public sector secondary schools from Lahore. To address the research questions of this study, frequency data were collected through Grammaticality judgment Tasks with thirty-six (36) WH-questions in Urdu. Each Urdu sentence is followed by two English translations (T1 & T2). The statistical analysis showed that UG Principles of WH-movement were fully available to Pakistani ESL learners whose L1 was Urdu. The study has important implications for the UG researchers in that more cross-linguistic studies are needed to verify Chomsky’s Hypothesized UG regarding non-English languages. The study also suggests revisiting the Critical Period Hypothesis. Besides, the teachers and materials developers of English grammar in non-native settings like Pakistan should rethink the role of direct instruction of rules and carrier content in teaching WH-questions to learners of English as a second language.

Keywords:

• cross-linguistic approach to WH-movement
• language input and UG
• parameter setting
• UG in the instructed SLA

1. Introduction

Urdu is the national language of Pakistan, which is spoken in all the provinces as a lingua franca. However, English is the official language in Pakistan. It is preferred in documentation, legislation and higher studies. English is also a prerequisite for jobs, which makes it indispensable for Pakistani learners at all levels of education. Sentence patterns in Urdu and English are very different. The word order of an English sentence is Subject-Verb-Object, and that of Urdu is Subject-Object-Verb. Apart from the basic sentence structure, there is another potential difference between English and Urdu: movement operations. Urdu is an in-situ language, whereas the English permits Wh-movement. However, despite the structural differences, Pakistani learners of English as a second language unconsciously internalize overt WH-movement in English and use it well in their language performance. Analyzing WH-movement acquisition in Pakistan’s context is essential to support or reject an internalized grammar constructed by the hypothesized principles of Universal Grammar (UG onwards).

Linguists (Aziz et al., 2020; Maqsood et al., 2019; Taci & Saraci, 2022) working in Chomsky’s tradition acknowledge the significant role of UG in first language (L1 onwards) acquisition. However, UG is just one of the several approaches and perspectives that explain the language acquisition process. This diversity of views not only has attracted the researchers (Perkins & Zhang, 2022) from Linguistics and SLA (Second Language Acquisition), but teachers of English (Saleem et al., 2018), psycholinguists (Diessel et al., 2020), and language policy makers (Han, 2020) also have shown keen interest in this field of study.

One of the dominant theories used to investigate L2 acquisition is UG. While exploring the similarity between L1 and L2 acquisitions, the researchers (Collins, 2022; Cowper & Hall, 2022) started to investigate whether UG works for L2 acquisition. Two main hypotheses are offered: namely, “The No UG access hypothesis”, which states that UG and its acquisition do not constrain L2 acquisition is different from that of L1; “The Full access hypothesis”, which says that L2 acquisition is constrained by UG and principles of UG are available for the L2 learners as well, and are widely tested to analyze if UG is accessible for L2 acquisition or not (Bley-Vroman et al., 1988; Epstein et al., 1996) among many others. This is the focus of the present study regarding Urdu learners of English in Pakistan.

If the Full access hypothesis is accepted, there arises a question about how second language learners access UG. To answer this question, two further hypotheses have been proposed: The Transfer hypothesis, which states that despite L2 learners access to UG, they cannot reset the L1 parameters which they have acquired previously and transfer their L1 parameters to L2, and Parameter Resetting Hypothesis, which says that UG constraints grammars of L2 learners and they reset the parameters of L1 according to the parameters of L2.

Some of the SLA researchers (Atkinson, 2019; Ellis, 2021) believe that the availability of UG for both acquisitions makes it possible for the variable parameters to be reset by the L2 learners who have first acquired the first language (Watanabe et al., 2008). Reglero (n.d.) used WH-parameter to check if parametric resetting occurs in the case of SLA. The study found substantial evidence of parameter resetting in the second language acquisition process. However, Tsimpli and Dimitrakopoulou (2007), in their research, concluded that parametric resetting did not occur while learning a second language.

This study aims to test these propositions regarding Urdu-speaking L2 learners of English in Pakistan. The findings of this study may stimulate teachers and L2 researchers to rethink strategies for teaching English constructions like WH-questions to Pakistani learners. The research may also contribute to the theoretical repertoire of studies regarding the role of Universal Grammar in Second language acquisition. It is hoped that the study may provide new dimensions for future research in understanding the acquisition process in a bilingual mind. It may also bring to light the divergence in UG theory and variations in its application to non-English languages.

2. Statement of the research problem

Urdu and English have vast structural differences in placing WH-question words. Pakistani English speakers tend to choose the WH-in-Situ parameter in Urdu, but they must learn WH-Fronting Parameter for constructing WH-questions in English. This study attempted to determine whether UG Principles of Wh-movements were available to Pakistani L2 learners for parametric resetting in English.

3. Research questions and hypotheses

The current study proposed the following research questions and hypotheses:

1. What is the frequency of UG complied WH- constructions corresponding to each Urdu WH-construction?

H₀: There is no difference in the frequency of the option “grammatical” for the first translation and the frequency of the option “grammatical” for the second translation in WH-constructions.

• (2) Is there any significant association between the two related WH-constructions?

H₀: There is no significant association between the two related WH-constructions.

• (3) Is the acquisition of the WH-movement age dependent?

H₀: Acquisition of the WH-movement is not age dependent.

4. Literature review

Principles and Parameters Theory (PPT onwards) has its origin in the idea of poverty of the stimulus. However, children somehow grasp and internalize the underlying structure of language because they are endowed with innate knowledge of grammar called Universal Grammar. This mental knowledge consists of principles that are the same for all languages and parameters that vary among languages. Haegeman (1994) describes PPT as the view of language acquisition in a parametric setting.

The core idea of PPT is to distinguish the invariant principles across languages and parameters, which collectively act as an innate language acquisition system. PPT also acknowledges that languages not only vary based on parameters of UG, but they also vary in ways UG does not constrain, as is the case of Lexicon. In language learning, principles act the same way irrespective of the input-providing environment of the child. However, parameters take on distinctive value concerning a child’s language learning environment (Snyder & Lillo-Martin, n.d).

One crucial aspect of PPT is that it serves as a bridge between sound and meaning through Phonetic Form (PF). Which is the pronunciation of abstract phonological representation and Logical Form (LF), which is the structurally determined aspect of meaning and becomes the input for the semantic representation (Cook & Newson, 2007a). LF and PF are two external interfaces which connect the syntactic structure with sounds and meaning (Cook & Newson, 2007a). Figure 1 presents this model.

Figure 1. The computational system. Adopted from Cook and Newson (2007a), p.9.

5. The locality principle

The Locality Principle requires that all the grammatical operations tend to be local (Radford, 2009). Syntacticians (Carnie, 2013; Cook & Newson, 2007a) state that “the movement must be within a local part of the sentence from which the moved element originates” (p. 36). Cheng and Corver (2006) also think that locality requires movements to be short and operative within a limited structure. Subjacency Condition captures this shortness of movement. Successive-cyclic Movement is the result of this condition. Thus, Locality Principle requires all the movements or inversions to be local. Local means that the closet constituent is moved to the closest potential landing site. For instance, when a yes/no question is formulated in English through the inversion of auxiliary, the inversion does not occur directly between V→C rather the inversion first takes place from V→T and then from T→C. Same is true for WH-movement. When WH-quantifier moves out of the embedded CP, it first lands on the specifier of the embedded CP and then moves towards the specifier of the main clause CP.

Empty Category Principle (ECP onwards). The movements introduce traces (denoted by t) in the S-Structure. It brings with it the issue of interpreting these traces that ECP resolves. Although traces are not pronounced, their presence is observed in S-Structure. See the following data:

1. Who do you think that John saw?

2. Who do you think [t] John saw?

3*. Who do you think [that t] saw Bill?

4. Who do you think [t] saw Bill?

3 is ungrammatical because it escaped ECP (Saito & An, 2014).

5.1. Parameters of UG

Holmberg (2010) opines that linguistic variations among languages are due to parametric variations in the complex system of interdependent properties. Cook and Newson (2007b) sees principles as invariant across languages because they are “built-in”, and no human language tries to breach these principles. In contrast, parameters confine the variation amongst languages within a specific limit (p. 79). One crucial thing about parameters is that they have clustering properties. Once a parameter is set, it affects all the properties related to it. For instance, a [+prodrop] parameter encompasses the property of omission of subject pronoun, inversion of subject and verb in declarative sentences, and that trace effect. Thus a [+prodrop] language like Italian has all these properties, whereas [-prodrop] language like English does not have any of these properties (Gass & Selinker, 2001). As a variation of parameters leads to variation in language, analysis of parametric resetting can provide substantial evidence of the availability of UG in second language acquisition.

5.2. Wh-movement parameter

In English, WH-words/ question-words move from their original positions to the beginning of the sentence (Radford, 2004). Whereas Urdu and Chinese languages do not indicate any overt Wh-movement, Wh-words remain in the same position in these languages; such languages are called WH-in-Situ languages. According to Subsai (2015), movement is not just recognizing a phrase, taking it up and placing it anywhere, but it largely depends upon the Principle of Structure-Dependency. Apart from this, Wh-movement Parameter also follows the Locality Principle of UG, as all the WH-movements in English tend to be local grammatical operations.

As Wh-movement is the focus of the present study, some more aspects need to be treated, though briefly. Formation of WH-questions is not a random placement of WH-phrases in a sentence. Rather hierarchical relationships amongst the phrases are to be kept in view. There are two types of questions in English: yes/no questions, such as “Are students of syntax happy?” and WH-questions, such as “Who told you about the students’ feelings?”. In order to answer the latter, saying yes/no is not enough; instead, a detailed answer is required. It is important to note that WH-expressions represent phrases when they appear in a sentence and carry theta roles. For example, look at the following sentence with no missing information:

• 5. Fazal is buying shoes.

In English, the verb buy takes two arguments: a VP-external argument to carry the role of agent and a VP-internal one to carry the theme. Fazal is the “agent” and “shoes” the theme in the given sentence. If “Fazal” and “shoes” are replaced with who and what respectively, the sentence becomes a question with two WH-replacements for the missing information:

• 6. Who is buying what?

According to Carnie (2013), in English, WH-phrases appear far from the position where they were theta marked. It is apparent from the following example:

Ali bought what? … >

Chomsky (2005) maintains that movement is a kind of internal merger in which an item contained in an already existing structure is moved to a new position within the same structure. When a WH-word moves, the entire phrase containing it will move. This movement of the entire phrase cannot be a head-to-head movement but a movement to the specifier of CP, which is always a phrase in X-bar theory. Apart from the WH-phrase movement towards the specifier of CP, WH-questions also trigger T→C movement (also known as Subject-AUX inversion) in the main clause to check [+Q] in C (Radford, 2009). This leads to the assumption that the only complementizer compatible for Wh-movement in English is null (Carnie, 2013). “WH-phrase appears to the left of auxiliary in C that means that WH-phrase must raise to a position higher than C. The only position available to us is specifier of CP” (Carnie, 2013, p. 36). According to Sportiche et al. (2014) “the inverted auxiliary in yes/no questions precedes the subject [Spec, TP]. The auxiliary verb in T moves to C and we conclude that WH-phrase moves to [Spec, CP]”.

Haegeman (1994) states that WH-phrase lands only in the specifier of CP because of Structure Preserving Principle for Transformation, according to which the structure should not be disturbed by movement operation. That is why [Spec, CP] is the only potential landing site for WH-phrase. WH-movement is obligatory in English for one reason: to attain sentential scope, but this is not a universal principle. In WH-questions V→T movement also takes place, but it is not bound to occur in the main clause as that of T→C movement. Burton-Roberts (2011) states that all the auxiliaries are fronted to C position when forming a question, whereas WH-phrases move to the front of the fronted auxiliary. Thus the fronted WH-phrases occupy the daughter node of CP and sister node of C. When WH-phrases are moved to the specifier position of CP, they leave a trace [t] behind, which is c-commanded by its antecedent (moved WH-phrase). This trace results from composite operation, namely copying and deleting during WH-movement. A critical property of traces [t] is that they are appropriately governed either by their antecedents or the lexical category because of the Empty Category Principle. However, intermediate traces are not correctly governed (Haegeman, 1994).

This phenomenon is explained by Cook and Newson (2007b). Due to ECP, sub-extraction of WH-phrase from the subject position tends to be more difficult than sub-extraction of WH-phrase in the object position because objects are always properly governed by lexical heads. Thus, trace in the position of objects is always governed, but subjects are only sometimes correctly governed by the lexical heads. However, movement from the subject position would be difficult but not impossible because subjects can be properly governed by their antecedents if that antecedent is close, and the relationship will not be interrupted by any closer element like a complementizer. Burton-Roberts (2011) also suggests that WH-phrases can be fronted from the main-clause and the subordinate-clause. In contrast, auxiliary-fronting only occurs in main-clauses (Burton-Roberts, 2011).

The possible motivation for the movement of WH-phrase to the position of specifier of CP is the [+WH] feature which resides in C of CP. WH-phrase moves to specifier position of CP in order to check the [+WH] feature (Carnie, 2013). According to Radford (2009) C contains the EF (edge feature), which enables it to attract WH-quantifier from its original position to the specifier of CP. Once this EF is satisfied, it is deleted or inactive (Chomsky, 2006). WH-phrases also move from embedded clause to the specifier position of CP of the main clause.

5.3. Locality principle and island constraints

The cyclic nature of WH-movement indicates that it also follows Locality Principle. According to this principle, in a long WH-phrase movement, each successive movement of WH-phrase is bound to be local, that is, WH-phrase cannot skip any [Spec, CP] that comes into the way while moving up towards the highest [Spec, CP] position.

5.4. Minimal link condition (MLC)

Apart from the island-constraints mentioned above, MLC is another restriction on the WH-movement. MLC requires that the moving phrase must land to the closest potential site. WH-movement is motivated to get closer to C_[+WH] to check [+WH] feature. But if some other WH-phrase already occupies the closest potential landing site, then movement of the other WH-phrase is restricted (Carnie, 2013).

5.5. Pied piping

Ross (1967) describes pied piping as “any transformation which is stated in such a way as to effect the reordering of some specified NP node or any NP which dominates it” (p. 334). Following bracketed analysis explains this movement:

WH-in-Situ. Languages are dichotomous concerning WH-movement parameters. This variation in languages was accommodated in the Minimalist Program that is grown out of Principles and Parameters or Government and Binding Theory (Edelstein, 2020). In some languages like English, overt WH-movement takes place. But there are other languages which accept covert WH-movement. Urdu is grouped with this latter type of language called WH-in-Situ languages, in which WH-phrase remains at the base position. The non-overt WH-movement in In-Situ languages is caused by the uninterpretable relevant feature, which is weak or absent (Haung, 1982; Adger, 2005). This is a parametric difference between WH-movement and WH-in-Situ languages (Choi, 2009).

In WH-in-Situ languages, WH-phrases undergo movement, but unlike those languages which apply WH-movement before Spell-Out, WH-Movement is covert and occurs at LF level after the Spell-Out (Boeckx, 2006; Yang, 2006). If S-Structure requires movement, the movement will occur before Spell-Out, and it will be visible, whereas if LF requires movement, then it can be either overt or covert and will occur after Spell-Out (Boeckx, 2006; Wu, 1993). Moreover, linguistic Minimalism is an approach to Universal Grammar. It is just a revival of the previous concepts, not their abandonment. UG and Minimalism are not new theories of mental grammar, but rather an idea based on all the pre-Minimalist work done by Chomsky. Edelstein (2020) clarifies that Minimalism does not get rid of syntactic architecture arbitrarily, but it seems to represent our linguistic systems optimally, under the assumption that the system functions optimally. Merge requires feature matching or feature checking that is impossible without movements. So, the derivational levels, not the principles, were reduced to Merge and Move. Principles are there but replaced by the Feature Checking Theory. Feature checking motivates movements; merge will apply if features are checked. If any feature remains unchecked during derivation, the syntactic structure will crash at LF (see, Edelstein, 2020; Radford et al., 2009). That is why the Successive Cycles Movement/Shortest Movement Principle remains very much part of Minimalism and is stated as “Minimize Chain Links Principle” (Boeckx, 2006, p. 170).

6. Parametric setting in the first language acquisition

As principles are universal and common to all the languages, language acquisition is seen as setting the value of the parameters according to the parametric value of the language to be learned (Ayoun, 2005). Scholars (Cook, 1989; Radford, 2009) state that principles of language are an in-built part of the mind; they do not need to be learned; language learners automatically apply them whenever they start with any language. But parameters vary according to the language. Clahsen and Hong (1995) believe that language learning revolves around a parametric setting. Parameters Setting Theory states that children tend to set the values of parameters with subset/superset relations to the default value. If that default value proves wrong, the child resets its parametric value (Crain et al, 1994). Hyams (2013) also suggests that the initial value for the parameter is the default, and then based on receiving data, children reset their parameters for their L1.

7. Parameter setting in the second language acquisition

To identify whether UG is available in L2 acquisition or not, various hypotheses have been proposed such as No UG Access Hypothesis, Partial Access and Full UG Access Hypothesis. Proponents of “No Access Hypothesis” state that UG is not available in L1 acquisition but L1 learners “reconstruct much of it by observing the native language (Bley-Vroman, 1990). In contrast to No Access Hypothesis, Transfer Hypothesis assumes that Universal Grammar is available to L2 learners but in a limited way. It is argued that only those parameters come into action for L2, which were already available in L1 (Schachter, 1989; Strozer, 1992). The hypothesis assumes that L2 learners have full access to UG in L2 acquisition, but second language learners will be quicker in acquiring the same parameters for both L1 and L2. White (2003) proposed that five basic combinations can be possible under the canopy of the level of access and transfer. According to this approach L2 learners have access to UG via L1 grammar. If a parameter or principle is not present in L1 grammar, it will not be available for L2. Figures 2-8 visually sum up the range of hypotheses regarding the availability of UG in L2 acquisition:

Figure 2. Adopted from Sadeghi (2006), p. 54.

Figure 3. Adopted from Sadeghi (2006), p.55.

Figure 4. Adopted from Sadeghi (2006), p. 57.

Figure 5. Adopted from Sadeghi (2006), p. 57.

Figure 6. Adopted from Sadeghi (2006), p.57.

Figure 7. Meta-analysis Random effect model T1*T2 all constructions.

Figure 8. Meta-Analysis 2 of Age and all T1 (Random effect model).

8. Paradigm and research design

The primary purpose of this research was to collect empirical evidence to address whether UG was available to Pakistani L2 learners or not for WH-parametric resetting. Therefore, focusing on the fundamental purpose of the research, a descriptive research design based on a positivist paradigm was chosen for this study.

Population and Sample. Sample of this research consisted of 10 private and four government schools in Lahore city, which were selected using snowball sampling. Both male and female volunteers of grade 10, who studied question formation in Urdu and English, aged from 14.5 to 17, were selected as participants. The convenience sampling selected 260 students (109 males; 151 females).

Method of data collection. A Grammaticality judgment Task was devised as an instrument for data collection. The Task consisted of 36 Urdu questions and various Urdu and English question-words. Each question-word was used twice in a different context (see, Appendix A for T1 & T2).

The procedure of Statistical Analysis. SPSS was used for the statistical analysis of the data, whereas for Meta-analysis, the software “Comprehensive Meta-Analysis (3.0) was used. Frequencies were counted to check the number of respondents who selected correct grammatical construction and the number of respondents who selected ungrammatical construction (Research Question 1). Fisher’s Exact Test was used to test if there was any significant association between T1 and T2 constructions, and to measure the strength of association, effect size was calculated though Hedges” G Test. In addition, meta-analysis, with both Fixed and Random model was done for all T1 and T2 constructions to reach the conclusion (Research questions 2 & 3 and the three null hypotheses set for this study).

9. Results

In this section, the following tables (1-4) display the results of the analyses and their interpretations to provide evidence to address the research questions. Whereas, Table 1 presents the information regarding the age of particpants in the sample.

Table 1. Age of participants in the sample

Age	Frequency	Percent	Cumulative Percent	M	SD
Between 14.5 to 15.5 year	87	33.5	33.5	1.66	0.47
Between 15.6 to 16.5 year	173	66.5	100.0
Total	260	100.0

The total number of participants was (n = 260). Respondents aged between 14.5 to 15.5 years were 87 (33.5% of the total accessible population). Respondents aged 15.6 to 16.5 years were 173 (66.5% of the total population). Mean of the age of sample population was M = 1.66, and the Standard Deviation of the age of the sample population was SD = 0.47.

Table 2. Descriptive analysis of the data of Urdu Language

Sentences	N	M	SD	Frequency (translation 1) Grammatical (Movement observed)	Frequency (translation 1) Ungrammatical (Movement not observed)	Frequency (translation 2) Grammatical (movement observed)	Frequency (translation 2) Ungrammatical (Movement not observed)
Ap iss sab ka ilzaam kis par dy rahay ho? (Whom are you blaming for all this?)	260	1.05	0.23	94.2%	5.7%	6.9%	93.1%
Tumharay pas kitnay jootay hain? (How many shoes do you have?)	260	1.06	0.25	93.1%	6.9%	7.7%	92.3%
Ayesha nay computer kyun tora? (Why did Ayesha dash the computer?)	260	1.09	0.29	90.8%	9.2%	11.9%	88.1%
Tum Rizwan sey kab milay? (When did you meet Rizwan?)	260	1.08	0.27	91.9%	8.1%	11.2%	88.8%
Tum kahan jaa rahay thay jab mein aya? (Where were you going when I reached?)	260	1.06	0.25	93.1%	6.9%	13.5%	86.5%
apki tabiyat kaisi hai itni taweel allalat k baad? (How are you now after such a prolonged ailment?)	260	1.09	0.29	90.4%	9.6%	12.3%	87.7%
Kis ki pencil shelf par parri hui hai? (Whose pencil is this on the shelf?)	260	1.07	0.26	92.3%	7.7%	10%	90%
Aap haftay kay din ghair-hazir kyun thay? (Why were you absent on Saturday?)	260	1.16	0.37	83.1%	16.9%	22.7%	77.3%
Apkay pas kitni copies hain? (How many copies do you have?)	260	1.07	0.26	92.7%	7.3%	8.5%	91.5%
Wahan kon baitha hai safaid kameez mai? (Who is he in the white shirt?)	260	1.07	0.26	92.7%	7.3%	14.6%	85.4%
Train kab station par ai? (When did the train arrive at the station?)	260	1.36	0.48	63.8%	36.2%	43.1%	56.9%
Admi kis ka peecha kar raha hai? (Whom was that man chasing?)	260	1.29	0.45	70.4%	29.6%	38.5%	61.5%
Ap iss qatal k baray mai kiya jantay hain? (What do you know about this murder?)	260	1.08	0.28	91.2%	8.8%	13.5%	86.5%
Sumaira kab sy mera intezaar kar rahi hai? (For how long had Sumaira been waiting for me?)	260	1.09	0.29	90.4%	9.6%	13.5%	86.5%
Yeh bacha kis kay pas bhaag raha hai? (Whom is this kid running to?)	260	1.33	0.47	66.2%	33.8%	36.2%	63.8%
Mulaqaat kay liye konsa din muntakhib kiya gaya? (What day was a day for the meeting?)	260	1.12	0.33	87.3%	12.7%	16.2%	83.8%
Rohail kitni bar daant saaf karta hai? (How many times does Rohail brush his teeth?)	260	1.11	0.32	88.1%	11.9%	15.4%	84.6%
University yahan sy kitni door hai? (How far is the university from here?)	260	1.39	0.49	60.4%	39.6%	41.9%	58.1%
Aap kay pas kitnay paisay hain? (How much money do you have?)	260	1.15	0.36	84.6%	15.4%	16.9%	83.1%
Usnay kiya dawa kiya k akhbaar mai kiya parha? (What did he claim that he read in the newspaper?)	260	1.11	0.32	88.2%	11.9%	22.3%	77.7%
Wo kiya maanta hai kis nay kya kiya? (What does he believe? Who did what?)	260	1.17	0.38	82.3%	17.7%	22.3%	77.7%
Apkay kitnay bhai hain? (How many brothers do you have?)	260	1.15	0.36	84.6%	15.4%	18.1%	81.9%
Fakhir nay Rehan ko kyun mara? (Why did Fakhir beat Rehan?)	260	1.23	0.42	76.2%	24.8%	25.2	74.8%
Aap cake kab kaatain gy? (When will you cut the cake?)	260	1.16	0.36	83.8%	16.2%	16.2%	83.8%
Kamran kay mutabiq wo kahan apna khoya bhai dhoond ly ga? (Where, according to Kamran, will he find his lost brother?)	260	1.09	0.29	90.8%	9.2%	13.5%	86.5%
Aap ka safar kaisa raha? (How was your journey?)	260	1.08	0.29	91.2%	8.8%	8.8%	91.2%
Wo kis ki notebook hai wahan? (Whose notebook is that over there?)	260	1.08	0.27	91.5%	8.5%	8.5%	91.5%
Mujhe samjh nahi ai teacher aaj itni naraaz kyun thi? (I did not understand why the teacher was so angry today?)	260	1.18	0.39	81.2%	18.8%	19.6%	80.4%
Wo kon hai? (Who is he?)	260	1.07	0.26	92.3%	7.7%	7.7%	92.3%
Atir kis ky baray mai baat kar raha tha? (Whom was Atir talking about?)	260	1.24	0.43	75.4%	24.6%	22.2%	77.8%
Larka kis road ki tarf jaa raha tha? (Towards which road was the boy going?)	260	1.10	0.30	90%	10%	18.1%	81.9%
Aap kitni bar apnay doston ko miltay ho? (How often do you meet your friends?)	260	1.09	0.29	90.8%	9.2%	10%	90%
Apka ghar yahan sy kitni door hai? (How far is your house from here?)	260	1.08	0.27	91.5%	8.5%	12.3%	87.7%
Wahan kitni kitaabain hain? (How many books are there?)	260	1.08	0.29	91.9%	8.1%	8.1%	91.9%
Mujhe iss kay liye kitni qeemat ada karni paregi? (How much would it cost me?)	260	1.10	0.31	89.2%	10.8%	13.5%	86.5%
Ap kitnay arsay sy yeh kahani likh rahi hain? (Since how long have you been writing this story?)	260	1.11	0.31	88.8%	11.2%	11%	89%

Table 3. Meta-Analysis Random effect model T1*T2 all constructions

Hedges’ g	S.E	VAR	LL	UL	z-value	p-value
2.004	0.12	0.01	1.75	2.25	15.93	0.00**

Note: SE = Standard Error, VAR = Variance, LL = Lower Limit, UL = Upper Limit

Table 4. Meta-Analysis Random effect model T1 (all constructions) *Age

Hedges’ g	S.E	VAR	LL	UL	z-value	p-value
0.20	0.02	0.00	0.16	0.25	9.06	0.00

Note: SE = Standard Error, VAR = Variance, LL = Lower Limit, UL = Upper Limit

Data for the meta-analysis below was extracted from inferential analysis in which Fisher’s Exact Test and Hedge’s G Test were used for the analysis of the association between English translation 1 (movement observing translation) and English translation (movement non-observing translation) 2 of each Urdu question separately (see, Table A1).

Results in Table 3 indicate (S.E = 0.12, VAR = 0.01, LL = 1.75, UL = 2.25 and z-value = 15.93). The overall p-value reflects a highly significant association as p = 0.00, far less than the cut-off point p = 0.01. The overall effect size for this analysis (g = 2.00) is found to exceed Hedges’ g convention for a large effect size (g = 0.80).

Data for the meta-analysis below was extracted from inferential analysis in which the Chi-Square Test for independence and Phi Test for analysis of the association between Wh-Movement acquisition and age was applied to English translation 1 of each Urdu question separately (see, Table A2.).

Results in Table 4 indicate S.E = 0.02, VAR = 0.00, LL = 0.16, UL = 0.25, z-value = 9.06 and p-value = 0.00. The overall significance value i.e., p-value is reflecting a highly significant association between gender and wh-movement acquisition as p = 0.000 which is far smaller than cut off point p = 0.01. The overall effect size for this analysis (g = 0.20) is found to be far less than Hedges’ g convention for a small effect size (g = 0.3).

10. Discussion

The primary purpose of this research was to investigate the relationship between the acquisition of English WH-questions by Pakistani learners and the involvement of principles of Universal Grammar. The study also tested which hypothesis of UG availability in SLA was operative in Pakistani context. Frequency analysis was done to find the answer to the first research question:

What is the frequency of UG complied WH- constructions corresponding to each Urdu WH-construction?

The following null hypothesis was constructed for the above research question:

H₀: There will be no difference in the frequency of the chosen option “grammatical” for T1 and the frequency of chosen option “grammatical” for T2 WH-constructions for each Urdu WH-construction.

In light of the frequency analysis, it was found that a vast majority of the respondents chose T1 as grammatical for each Urdu construction. In contrast, most respondents chose T2 as ungrammatical for each Urdu construction. Based on the frequency analysis, the null hypothesis “There will be no difference in frequency of the chosen option ‘grammatical’ for T1 and the frequency of chosen option ‘grammatical’ for T2 WH-constructions for each Urdu WH-construction” was rejected, and the alternative hypothesis was retained. The result of frequency analysis indicates that most of the respondents could identify not only WH-movement but also all the relevant uninterpretable features [+WH], locality conditions and locality constraints. This finding reflects that the respondents were not guessing the correct WH-construction by simply seeing the WH-phrase at the front of the sentence. Rather they have internalized the mechanism involved in the construction of WH-questions. The internalization was only possible with parameter resetting by the Urdu-speaking learners of English.

Respondents were able to access the grammatically correct WH-construction in which WH-movement occurred even though neither the WH-movement was not primarily present in Urdu nor was any explicit instruction given to the students of grade 10 about the locality conditions. If it was directly taught, WH-movement could be taken as indirect negative evidence in learning WH-questions in English. This study result is consistent with that found in Snape (2006) but contradicts those of Meisel (1991). The second research question was:

Is there any significant association between the two related WH-constructions?

To find answers to this research question, Fisher’s Exact Test and Hedges’ G Test were employed on the two related WH-constructions; namely, T1 and T2 for each Urdu WH-construction and meta-analysis was done for the related WH-constructions. Therefore, the null hypothesis for this question was:

H₀: There is no association between the two related WH-constructions.

The statistical analysis indicated a significant association between two related WH-constructions for each pair of WH-construction. The meta-analysis, with both fixed effect and random effect models, indicated a strong association between the two related WH-constructions. Based on statistical evidence, the null hypothesis “There is no association between the two related WH-constructions” was retained and the alternative hypothesis was rejected. The association between the two related WH-constructions is indicative that the respondents have evaluated the two related WH-constructions with the support of UG principles, which helped them to identify the correct English WH-construction. The results of this study provide evidence for Parameter Resetting and Full Access Hypothesis. The results of this study are supported by Tayyabi (2012), Enjavinezhad and Paramasivam (2014), Aldwayan (2008), (and Epstein et al. (1996). However, the findings contradict those found in Bley-Vroman (1990), Muftah and Eng (2011), and Prentza (2012). The third research question and related hypothesis was:

Is WH-movement acquisition age dependent?

H₀: WH-movement is not age dependent

The individual statistical analysis of each T1 concerning age indicated no significant association between age and correct WH-construction. The finding was the same for most of the WH-constructions and the correct WH-constructions, significant association with age was observed. Although there was relationship between some WH-constructions and age, the strength of the association was negligible. However, the meta-analysis, with both Fixed effect and Random effect models, indicated a significant association between age and WH-question movement. The analysis indicated that WH-movement was age dependent. Overall, though WH-movement acquisition was age-dependent, the strength of this association was not considerable. However, this result provides statistical evidence that age can play a role in resetting specific parameters. Thus, the null hypothesis “WH-movement is not age dependent” cannot be entirely rejected.

11. Conclusion

In the present research, WH-movement parameter of English has been used to analyze the accessibility of Universal Grammar in SLA by Urdu speaking Pakistani learners of English. Statistical analysis of the data provided empirical evidence that Pakistan English learners utilize internal support of UG in SLA. The results of this study do not support No Access Hypothesis proposed by (Bley-Vroman 1990; Clahsen & Peter, 1986; Schacter, 1988). The findings do not support Hasebe’s (2014) argument in favor of Transfer Hypothesis, as the current research showed that Pakistani English learners were able to reset their covert WH-movement parameter. Based on empirical evidence, the present research supports the Parameter Resetting Hypothesis, that is also supported by other studies (Finer & Broselow, 1986; McLaughlin, 1998).

The availability of UG in SLA plays a vital role in second language teaching. UG helps the second language acquirer create a mental representation of the target language via positive, direct, negative, and indirect negative evidence. Full access to UG would help the L2 learner in resetting parametric values of L1 in L2 through minimal input of the target language from their surroundings. This is also important to note that the quantity of the input data is less helpful than the quality. Even a few sentences could help the L2 learners reset their parameters of L1 if the data provided reflects the property of the parameter to be reset.

The role of UG in learning English in the Pakistani context also challenges the traditional rules-based methodology of English teaching. Explicit teaching and learning of grammatical rules play a minor role in creating mental representations of the target language in the mind of L2 learners. What is required in this regard is just the quality input of the relevant parameter. The availability of UG in SLA also demands feedback to the L2 learners on their language constructions because feedback would help them test their hypothesis about a particular parameter and then set it to the required value. Concerning the syllabus, it is suggested that it should be designed keeping in view the principles and parameters of UG. As principles of UG are available in all the languages, so they need not be included in the syllabus. However, only those aspects of grammar should be included which involve parametric differences between L1 and L2. It is also implied that while making a syllabus for L2 learners, instead of teaching them concrete “rules”, structural input should be provided in L2. Teachers need to enrich their knowledge of UG’s principles and parameters, which can help them understand the differences between L1 and L2 and thus lead to knowledge-based teaching strategies.

In addition, the age variable should be considered while devising syllabus and methodology for L2 learners to make the process of parametric resetting natural and attain the optimal level of L2 acquisition by Pakistani learners, whose L2 acquisition often remains behind the optimum level of attainment. The researchers concede that very few aspects of the issue have been explored in this study, therefore, they suggest that more cross-linguistic data be explored to test the explanatory power of Chomsky’s principles and parameters or UG theory. This study was a mild attempt, especially to invite the attention of Pakistani researchers to this so far less attended domain.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The authors received no direct funding for this research.

Appedix A

Translation 1 (T1) and Translation 2 (T2) of Each Urdu Question

1. Aap iss sab ka ilzaam kis ko dy rahay hain?

T1. Who are you blaming for all this?

T2. Who you are blaming for all this?

2. Tumharay pas kitnay jootay hain?

T1. How many shoes do you have?

T2. How many do you have shoes?

3. Ayesha nay computer kyun tora?

T1: Why did Ayesha break the computer?

T2. Ayesha did why break the computer?

4. Tum Rizwan sey kab milay?

T1: When did you meet Rizwan?

T2. Did you meet when Rizwan?

5. Tum kahan jaa rahay thay jab mein aya?

T1. Where were you going when I came?

T2. Where you were going when I came?

6. Apki tabiyat kaisi hai itni taweel allalat k baad?

T1. How are you after this long span of illness?

T2. How you are after this long span of illness?

7. Kis ki pencil shelf par parri hui hai?

T1. Whose pencil is lying on the shelf?

T2. Whose is lying on the shelf pencil?

8. Aap haftay kay din ghair-hazir kyun thay?

T1. Why were you absent on Saturday?

T2. Why you were absent on Saturday?

9. Apkay pas kitni copies hain?

T1. How many copies do you have?

T2. How many do you have copies?

10. Wahan kon baitha hai safaid kameez mai?

T1. Who is sitting there in the white shirt?

T2. Who there is sitting in the white shirt?

11. Train kab station par ai?

T1. When did the train arrive at the station?

T2. When the train did arrive at the station?

12. Admi kis ka peecha kar raha hai?

T1. Whom is the man chasing?

T2. Whom the man is chasing?

13. Ap iss qatal k baray mai kiya jantay hain?

T1. What do you know about this murder?

T2. You what know about this murder?

14. Sumaira kab sy mera intezaar kar rahi hai?

T1. How long has Sumaira been waiting for me?

T2. Has Sumaira been how long waiting for me?

15. Yeh bacha kis kay pas bhaag raha hai?

T1. To whom is the boy running?

T2. Whom is the boy running to?

16. Mulaqaat kay liye konsa din muntakhib kiya gaya?

T1. Which day was selected for the meeting?

T2. Which was selected day for the meeting?

17. Rohail kitni bar daant saaf karta hai?

T1. How often does Rohail brush his teeth?

T2. Does Rohail brush his teeth how often?

18. Mein sochti hun university yahan sy kitni door hai?

T1. I wonder how far the university is from here?

T2. I wonder how far is the university from here?

19. Aap kay pas kitnay paisay hain usnay pucha?

T1. How much money do you have?

T2. How much do you have money?

20. Usnay kiya dawa kiya k akhbaar mai kiya parha?

T1. What did she claim that she read in the newspaper?

T2. What did she make the claim that she read in the newspaper?

21. Wo kiya maanta hai kis nay kya kiya?

T1. Who does he believe has done what?

T2. Who does he believe what has done?

22. Apkay bhai kitnay hain?

T1. How many brothers do you have?

T2. How do you have many brothers?

23. Fakhir nay Rehan ko kyun mara?

T1. Why did Fakhir beat Rehan?

T2. Fakhir beat Rehan why?

24. Aap cake kab kaatain gy?

T1. When will you cut the cake?

T2. You will cut when the cake?

25. Kamran kay mutabiq wo kahan apna khoya bhai dhoond ly ga?

T1. Where does Kamran believe that he would find his lost brother?

T2. Where does Kamran have the belief that he would find his lost brother?

26. Aap ka safar kaisa raha?

T1. How was your journey?

T2. How your journey was?

27. Wo kis ki notebook hai wahan?

T1. Whose notebook is that over there?

T2. Whose is that over there notebook?

28. Mujhe samjh nahi ai teacher aaj itni naraaz kyun thi?

T1. I did not understand why the teacher was angry today?

T2. Why I did not understand the teacher was angry today?

29. Wo kon hai?

T1. Who is he?

T2. Who he is?

30. Atir kis ky baray mai baat kar raha tha?

T1. About whom was Atir talking?

T2. Whom was Atir talking about?

31. Larka kis road ki tarf jaa raha tha?

T1. Towards which road was the boy going?

T2. Which road was the boy going towards?

32. Aap kitni bar apnay doston ko miltay ho?

T1. How often do you visit your friends?

T2. Do you visit your friends how often?

33. Apka ghar yahan sy kitni door hai?

T1. How far is your home from here?

T2. Is your home how far from here?

34. Wahan kitni kitaabain hain?

T1. How many books are there?

T2. How many are there books?

35. Mujhe iss kay liye kitni qeemat ada karni paregi?

T1. How much do I need to pay for this?

T2. I need to how much pay for this?

36. Ap kitnay arsay sy yeh kahani likh rahi hain?

T1. For how long have you been writing this story?

T2. How long you have been writing this story for?

Appendix

Table A1. Fisher’s Exact & Hedges’ G Test of association between T1 and T2 of each Urdu question

Fisher’s exact test p-value	Hedges’ g	S.E	VAR	LL	UL	z-value
0.000**	2.968	0.199	0.040	2.579	3.358	14.929
0.000**	2.799	0.186	0.035	2.435	3.163	15.066
0.000**	2.359	0.158	0.025	2.049	2.669	14.918
0.000**	2.481	0.166	0.027	2.156	2.806	14.972
0.000**	2.455	0.168	0.028	2.126	2.784	14.646
0.000**	2.315	0.156	0.024	2.010	2.620	14.875
0.000**	2.578	0.171	0.029	2.242	2.914	15.041
0.000**	1.551	0.122	0.015	1.311	1.790	12.689
0.000**	2.709	0.180	0.032	2.357	3.062	15.058
0.000**	2.370	0.163	0.027	2.051	2.690	14.547
0.000**	0.467	0.099	0.010	0.272	0.661	4.712
0.000**	0.735	0.103	0.011	0.534	0.936	7.170
0.000**	2.309	0.156	0.024	2.002	2.615	14.760
0.000**	2.258	0.153	0.023	1.958	2.558	14.755
0.000**	0.682	0.101	0.010	0.484	0.881	6.734
0.000**	1.968	0.138	0.019	1.697	2.239	14.232
0.000**	2.039	0.142	0.020	1.762	2.317	14.401
0.000**	0.411	0.098	0.010	0.219	0.604	4.187
0.000**	1.814	0.131	0.017	1.557	2.072	13.817
0.000**	1.788	0.134	0.018	1.526	2.050	13.391
0.000**	1.533	0.121	0.015	1.295	1.771	12.633
0.000**	1.770	0.130	0.017	1.516	2.025	13.649
0.000**	1.222	0.112	0.013	1.002	1.441	10.921
0.000**	1.813	0.131	0.017	1.556	2.070	13.821
0.000**	2.283	0.155	0.024	1.980	2.586	14.760
0.000**	2.568	0.170	0.029	2.235	2.902	15.104
0.000**	2.622	0.173	0.030	2.282	2.962	15.112
0.000**	1.580	0.123	0.015	1.340	1.820	12.897
0.000**	2.736	0.181	0.033	2.381	3.091	15.099
0.000**	1.210	0.112	0.012	0.991	1.428	10.848
0.000**	2.042	0.144	0.021	1.759	2.324	14.148
0.000**	2.468	0.164	0.027	2.147	2.789	15.058
0.000**	2.392	0.161	0.026	2.077	2.707	14.877
0.000**	2.678	0.177	0.031	2.330	3.025	15.111
0.000**	2.189	0.149	0.022	1.897	2.480	14.709
0.000**	2.285	0.153	0.024	1.984	2.585	14.896

Appendix

Table A2. Chi-Square Test for independence and Phi Test for analysis of association between Wh-Movement acquisition and age

Sentence	X^2	DF	P	Φ
Sentence 1 (T1)	15.485^a	1	.000	−.244
Sentence 2 (T1)	1.048^a	1	.306	−.063
Sentence 3 (T1)	3.248^a	1	.072	−.112
Sentence 4 (T1)	11.313^a	1	.001**	−.209
Sentence 5 (T1)	1.048^a	1	.306	−0.063
Sentence 6 (T1)	1.380^a	1	.240	−.073
Sentence 7 (T1)	6.854^a	1	.009**	−.162
Sentence 8 (T1)	.637^a	1	.425	−.049
Sentence 9 (T1)	14.895^a	1	0.000**	−.239
Sentence 10 (T1)	3.383^a	1	.066	−.114
Sentence 11 (T1)	.022^a	1	.881	−.009
Sentence 12 (T1)	.049^a	1	0.826	.014
Sentence 13 (T1)	8.513^a	1	0.004**	−.181
Sentence 14 (T1)	11.586^a	1	0.001**	−.211
Sentence 15 (T1)	.186^a	1	.666	−.027
Sentence 16 (T1)	5.533^a	1	.019	.019
Sentence 17 (T1)	3.522^a	1	0.061	−.116
Sentence 18 (T1)	.867^a	1	.352	.058
Sentence 19 (T1)	2.827^a	1	0.093*	−.104
Sentence 20 (T1)	.065^a	1	.799	−.016
Sentence 21 (T1)	.807^a	1	0.369	−.056
Sentence 22 (T1)	5.807^a	1	0.016*	−.149
Sentence 23 (T1)	10.002^a	1	.002**	−.196
Sentence 24 (T1)	.483^a	1	0.487	−.043
Sentence 25 (T1)	3.248^a	1	0.072	−.112
Sentence 26 (T1)	1.137^a	1	.286	−.066
Sentence 27 (T1)	7.090^a	1	.008**	−.165
Sentence 28 (T1)	3.547^a	1	0.060	−.117
Sentence 29 (T1)	9.680^a	1	0.002**	−.193
Sentence 30 (T1)	.016^a	1	0.899	.008
Sentence 31 (T1)	2.090^a	1	0.148	−0.090
Sentence 32 (T1)	1.818^a	1	.178	−.084
Sentence 33 (T1)	1.553^a	1	0.213	−.077
Sentence 34 (T1)	.000^a	1	0.990	.001
Sentence 35 (T1)	.478^a	1	0.489	−.043
Sentence 36 (T1)	.293^a	1	.588	−.034