The effect of explicit instruction on the acquisition of subject-verb inversion by Macedonian learners of English
Abstract
68
Full text
The effect of explicit instruction on the acquisition of subject-verb inversion by Macedonian learners of English Natasha Stojanovska-Ilievska (Ss. Cyril and Methodius University in Skopje) ABSTRACT This paper reports on the effectiveness of teaching grammar explicitly, with afocus on form, function, and rules. The study tested the hypothesis that explicit instruction of grammar leads to increased accuracy in the grammaticality judgment, using asample of 32 Macedonian learners of English at levels B and C.The structures selected for the task included subject-verb inversion, subject-verb inversion and it-insertion, omission of expletive and anticipatory it and omission of areferential subject, which are typical in the interlanguage of Macedonian learners of English due to cross-linguistic influence from their mother tongue and developmental factors. The study involved apre-test, explicit instruction and controlled practice activities, and apost-test. The pre-test and post-test assessed the accuracy of aseries of sentences, while the instruction involved contrastive analysis of examples in Macedonian and English, eliciting differences between the two languages. Results showed improved accuracy in the assessment of the grammaticality of the target structure at the post-test, but an experimental study with acontrol group is required to verify whether the improvement can be solely attributed to explicit instruction. KEYWORDS accuracy, explicit instruction, grammaticality judgment, subject-verb inversion, word order DOI https://doi.org/10.14712/18059635.2023.1.3 1 INTRODUCTION Whether grammar should be explicitly taught or exposure to ample authentic input is sufficient for learners to master aparticular aspect of foreign language grammar has been asubject of alongstanding debate among practitioners and researchers alike (Terrell 1991; Swan 1985; Norris and Ortega 2001). In the widespread communicative and usage-based teaching approaches an overt focus on form, function and rules is generally given aperipheral role (Krashen and Terrell 1988; Terrell 1991; Teixeira 2018). By and large, recourse to the learners’ native language is similarly frowned upon, whether for contrastive purposes or for translation. Contrary to these views, this study explores if it is possible that learners actually benefit from explicit instruction focusing on grammatical structures that are assumed to pose problems for learners as aresult of transfer from the mother tongue. More specifically, it studies the effectiveness of explicit grammar instruction in the acquisition of subject-verb inversion by Macedonian learners of English. Areas investigated in this study include structures related to subject-verb inversion (SVI), which Macedonian language learners of English find problematic for two OPEN ACCESS
NATASHA STOJANOVSKA-ILIEVSKA 69 main interconnected reasons. Firstly, there are considerable differences in the syntax of the subject in English and in the students’ mother tongue, Macedonian. Macedonian is anull subject language, whereas English requires that the subject position be filled with adummy subject when there is no referential subject (Mitkovska and Bužarovska 2018; Mitkovska et al., forthcoming). Secondly, SVI is governed by different principles in English and in Macedonian. English has agrammatical word order and imposes major constraints on SVI, while the word order in Macedonian is pragmatically motivated and SVI is much more frequently applied (Mitkovska et al., forthcoming). In English SVI is positioned at the syntax-discourse interface, which is interrelated with the syntax-phonology and the lexicon-syntax interface (Lozano and Mendikoetxea 2010). Because of these differences, some negative transfer is expected from the native language (L1) to the target language (L2). The study followed apre-test/intervention/post-test design and aimed to examine whether explicit instruction of grammar focusing on the governing principles and constraints of SVI in English contributes to increased accuracy in the students’ acceptability judgment of SVI sentences. The structures investigated in this study are characteristic of the interlanguage of Macedonian learners of English and stem from the cross-linguistic influence from their native language, as well as certain developmental factors. They comprise subject-verb inversion, subject-verb inversion + it-insertion, omission of expletive and anticipatory it and omission of areferential subject, all of which are briefly outlined below. What is meant by inversion in this study is asubject-verb inversion in declarative sentences, also known as ‘full inversion’ of the subject and the verb, as in: On one long wall hung arow of Van Goghs (Biber et al. 2021: 904). Cases of subject-operator inversion, also known as ‘partial inversion’, as in: And no way could she get Sarah to understand that (Biber et al. 2021: 907) were not within the scope of the study. Keeping in mind that English has arather fixed subject-verb-object (SVO) word order, the use of SVI in English declarative sentences is highly constrained and typically occurs when the following conditions have been met: in sentence-initial position there is an adverbial (usually of place or time) that builds on the preceding discourse; the verb belongs to the category of intransitive or copular verbs and denotes existence or appearance in aparticular setting; and the subject is rather long and brings new information in the discourse (Biber et al. 2021: 903). In Macedonian the word order is more flexible and the linear arrangement of sentence constituents is driven by the communicative needs of the speakers (Minova-Ǵurkova 2000), which is why SVI is more common. This fact is expected to pose difficulties for Macedonian learners of English who are unaware of the constraints on SVI in English. The structure labelled ‘SVI + it-insertion’ applies to the occurrence of non-referential it in sentences with averb-subject (VS) word order, which is occasionally encountered in L2 English (Lozano and Mendikoetxea, 2010; Agathopoulou, 2014). Mitkovska et al. (forthcoming) found that it-insertion is aphenomenon typically occurring in the interlanguage of Macedonian learners of English to agreater or alesser extent depending on the level of proficiency, while English native speakers almost entirely reject sentences with SVI + it-insertion. OPEN ACCESS
70 LINGUISTICA PRAGENSIA 1/2023 As English is anon-null subject language, in the absence of areferential subject in asentence, the dummy pronoun it is used since the subject slot must be filled for the sentence to be grammatical. The English non-referential it has no counterpart in Macedonian and it has no translation equivalent. Unlike English, Macedonian readily allows the omission of the subject since verbal inflections provide information about the person, number, and sometimes even the gender of the subject. Not only can the subject be omitted, but it generally is omitted when it is recoverable from the context. It is specifically mentioned only for the purposes of emphasis or contrast. Thus, in this study, the omission of non-referential it and the omission of areferential subject are considered in view of the cross-linguistic influence from the learners’ L1. 2 THEORETICAL BACKGROUND While the explicit focus on grammar was an indispensable part of language instruction from the sixties to the early eighties of the twentieth century, with the advent of various communicative methods in language teaching, explicit grammar instruction was no longer perceived as central to foreign language teaching (Terrell 1991; Swan 1985). This trend has persisted in the following decades. Yet, findings from multiple previous studies demonstrate that explicit instruction is effective (Rahimpour and Salimi 2010; Bowles and Montrul 2008), or even more effective compared to implicit instruction (Norris and Ortega 2001). Research has revealed that grammatical accuracy cannot be achieved by learners of aforeign language simply by exposure to ample authentic and comprehensible input (Swain 1985; Swain and Lapkin 1982). Similarly, in astudy of French children learning English as their L2, White (1991) found that exposure only to positive evidence was insufficient for learners to realise the impossibility and ungrammaticality of agiven L1 grammatical structure in L2 and that negative evidence (in the form of form-focused instruction and error-correction) was more helpful in this regard. In another study, even though the positive evidence did contribute to the increased use of the targeted L2 word order by the learners, it did little to prevent them from also using aword order, typical of L1 but unacceptable in L2. This might suggest that explicit instruction could be helpful in such cases, but it could also point to the mere coexistence of different forms in the interlanguage for acertain period of time, which is oftentimes attested in second language acquisition (SLA) (Trahey and White 1993). As for the acquisition of SVI in English, there are some indications that it is best acquired at the advanced stages of L2 proficiency. Specifically, for native speakers of European Portuguese (anull subject language like Macedonian), it was found that explicit instruction on the ungrammaticality of “free” inversion can only benefit learners when they are at the advanced level, because that is the period when they are developmentally ready to acquire it (Teixeira 2018: 363–364). Lozano and Mendikoetxea (2010) demonstrate that the VS order in English is restricted by properties operating at the lexicon-syntax, syntax-phonology and syntax-discourse interface. They analysed both the specifics and the interrelatedness of OPEN ACCESS
NATASHA STOJANOVSKA-ILIEVSKA 71 these interfaces and found that the lexicon-syntax interface is reflected in the fact that it is the class of unaccusative verbs that constitutes anecessary (though insufficient) condition for the occurrence of the VS order. Secondly, according to the endweight principle, the order of the subject and the verb is more likely to be inverted when the subject is phonologically heavy, i.e. long and structurally complex, which is indicative of the involvement of the syntax-phonology interface in SVI. Finally, the fact that in English SVI the subject tends to bring new information in the discourse illustrates the role of the syntax-discourse interface in SVI. This is based on the endfocus principle, according to which the element that conveys new information in the discourse is placed towards the end of aclause. The Interface Hypothesis was initially formulated in Sorace and Filiaci’s paper (2006: 340) and postulated that “narrow syntactic properties are completely acquirable in asecond language […] whereas interface properties involving syntax and another cognitive domain may not be fully acquirable”. This hypothesis did not come without acaveat that some narrow syntactic properties, though acquirable, can still cause developmental difficulties. Studies demonstrated that some interfaces are more susceptible to errors than others (Slabakova and Ivanov 2011: 637). Tsimpli and Sorace (2006) found that there are differences between the developmental patterns displayed by the syntax-semantics and the syntax-discourse interface, based on the assumption that properties at the syntax-semantics interface belong to the language system itself and are therefore more easily mastered, unlike the properties at the syntax-discourse interface which pose greater difficulties for foreign language learners as aresult of the greater processing demands placed on the learners trying to integrate language-internal and pragmatic information. The study by Sorace and Serratrice (2009) makes aclear distinction between internal interfaces (between narrow syntax and other linguistic modules) and external interfaces (between syntax and other cognitive systems). It pointed to similar results focusing specifically on the differences in the acquisition of phenomena at the syntax-semantics interface and the syntax-discourse/pragmatics interface in bilingual speakers. It was believed that this external interface is asource of residual optionality for L2 learners even at near-native level of proficiency (Slabakova and Ivanov 2011: 638). This residual optionality is seen as aconsequence of the inefficiency at integrating syntactic and discourse-pragmatic information in actual language use (Sorace 2011). Unlike the narrow syntax domain (which includes the ungrammaticality of null subjects and “free” inversion in English), some optionality is expected to occur at the syntax-discourse interface, regardless of the L1–L2 combinations, as aby-product of bilingualism (Teixeira 2018: 157–158). Having said that, it may be the case that the relationship between particular interfaces and SLA might not be so straightforward. Difficulties in SLA arise across all interfaces, since even in internal interfaces there appear to be phenomena that cannot be completely mastered, whereas some discourse-related properties can be fully acquired (Slabakova and Ivanov 2011: 650). This blurs the supposedly clear-cut correlation between the internal interfaces and the absence of optionality at the near-native level on the one hand, as well as the correlation between the external interfaces and the residual optionality, on the other. OPEN ACCESS
72 LINGUISTICA PRAGENSIA 1/2023 Thus, studies on the syntax-discourse interface have yielded conflicting results, sometimes finding residual optionality and other times detecting no optionality in end-state grammars. Apossible reason for this could be that the methodology employed (generally untimed offline tasks) was not well-suited to the detection of optionality if it was weakly present, since optionality is best detected via timed online tasks that reveal processing inefficiencies (Teixeira 2020: 34). Several generative second language acquisition researchers (Rothman 2010; Valenzuela and McCormack 2013; VanPatten and Rothman 2015; Whong et al. 2014, as cited in Teixeira 2018) have suggested that explicit instruction could assist in the acquisition of properties at the syntax-discourse interface and have even encouraged greater focus on these phenomena in the process of instruction. Along these lines, the present study was intended to raise the students’ awareness of the principles regulating SVI in English and to examine whether students could progress towards native-like proficiency in SVI through explicit instruction. 3 STUDY DESIGN AND METHODOLOGY The study was carried out among students from Ss. Cyril and Methodius University and FON University in Skopje. Atotal of thirty-two native Macedonian learners of English (mean age: 20.88 years, standard deviation: 0.98; male=6, female=26) at levels B and C according to the Common European Framework of Reference (CEFR) participated in the study. Of these, ten were at level B (intermediate to upper-intermediate) and twenty-two were at level C (advanced to proficiency). The students’ level of English proficiency was determined with aplacement test, which was administered at the two universities and comprised 46 multiple choice questions that tested knowledge of grammatical structures and vocabulary at various levels of complexity. The data collection process, as well as the explicit grammar instruction and controlled practice activities, took place offline, in aclassroom setting. For the purposes of this study ashort experiment was designed involving three stages: apre-test, explicit instruction and controlled practice activities, and aposttest. The pre-test and post-test were conceived as timed tasks in the form of agrammaticality judgment and correction task (GJCT) and centred on the assessment of the accuracy of aseries of sentences, samples of which are available in the Appendix. The pre-test comprised 29 sentences, which included sentences that tested the students’ understanding of SVI in English, as well as distractor sentences. Students were asked to label these sentences as either correct or wrong. They were also asked to correct the sentences they thought were wrong. The instruction involved contrastive analysis of examples in Macedonian and English, eliciting differences between the two languages. This was followed by aset of controlled practice activities which were done in pairs and involved identifying errors related to SVI, as well as subject realization and subject omission in individual sentences and in short texts. The post-test consisted of 21 sentences structured according to the same principles as the ones in the pre-test. In the Results section of this paper, the pre-test and post-test are also referred to as Test 1 and Test 2, respectively. OPEN ACCESS
NATASHA STOJANOVSKA-ILIEVSKA 73 The study period spanned over fifteen days, with the pre-test and the post-test being administered at the beginning and at the end of this time-frame, respectively. The instruction and controlled practice activities were carried out on Day 8 of the study period, midway between the pre-test and the post-test. The structures tested in this study included subject-verb inversion, subject-verb inversion + it-insertion, omission of expletive and anticipatory it and omission of areferential subject, all of which are typical for the interlanguage of Macedonian learners of English due to cross-linguistic influence from their L1 as well as some developmental factors. Seeing that the conditions for parametric statistics (including normal distribution) were not entirely met at either level B or level C, aWilcoxon signed-rank test was performed for both levels under the null hypothesis that the median of the underlying population of differences is equal to 0. This test made possible the comparison between the pre-test and the post-test results per category within the same proficiency level. In addition, test results were compared between the two proficiency levels by applying the Wilcoxon rank-sum test. Acomparison between the pre-test and the post-test results of individual students on both levels is offered as well. As is widely accepted, ap-value of less than 0.05 was considered statistically significant. Since the study was performed without acontrol group, diverse confounding variables are assumed to pose threats to the causal validity of the study (Marsden and Torgerson 2012), some of which are addressed herein. As far as maturation is concerned, the interval between Test 1 and Test 2 was rather short for adiscernible maturation effect to take place. With regard to history, no known past event was identified that could have potentially influenced the students’ scores. As for the test effects, several distractor sentences were inserted in each GJCT so as to mask the true purpose of the tests, though it cannot be dismissed that some awareness-raising might have occurred as aresult of the questions on the pre-test. 4 RESULTS The study findings are summarized in sections 4.1–4.5. Sections 4.1 and 4.3 offer acomparison between the pre-test and post-test results for levels B and C per category, while sections 4.2 and 4.4 present pre-test/post-test results of individual participants at these two levels respectively. Section 4.5 offers acomparison between the results on level B and level C. 4.1. LEVEL B RESULTS PER CATEGORY As shown in Table 1, the data indicate improvement across all 4 categories. In Test 1 in the category of SVI 47% of the sentences were wrongly classified as correct, and that number significantly dropped to alittle over 20% in Test 2. Similarly, there has been amarked decline in wrong answers in the next two categories: SVI with it insertion and Omission of expletive and anticipatory it. As far as the Omission of areferential subject is concerned, in Test 2 no sentences were wrongly labeled as correct. The Wilcoxon signed-rank test demonstrated that there was statistically significant difference between Test 1 and Test 2 in the category of SVI and Omission of areferential OPEN ACCESS
74 LINGUISTICA PRAGENSIA 1/2023 subject. The results showed improvement on Test 2 even in the other two categories, only this improvement could not be labelled as statistically significant. More precisely, the p-values measuring statistically significant difference between the scores on Test 1 and Test 2 per category were as follows: p=0.0106 for SVI, p=0.1238 for SVI with it insertion, p=0.0648 for Omission of expletive and anticipatory it, and p=0.0455 for the Omission of areferential subject. There was also astatistically significant difference (p=0.0107) between the mean for Test 1 and the mean for Test 2. The z-values and the p-values for all categories are summarized in Table 2. SVI SVI with itinsertion Omission of expletive and anticipatory it Omission of areferential subject T1 T2 T1 T2 T1 T2 T1 T2 Mean 0.47 0.22 0.48 0.33 0.55 0.30 0.10 0.00 Median 0.58 0.20 0.50 0.33 0.58 0.25 0.00 0.00 Standard Deviation 0.31 0.24 0.30 0.35 0.25 0.31 0.13 0.00 Minimum 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Maximum 0.83 0.60 0.80 1.00 0.83 0.75 0.25 0.00 Table 1.Level B: Proportion of sentences wrongly classified as correct (Test 1 vs Test 2) Level B z-value p-value SVI (Test 1 vs Test 2) 2.555 0.0106 SVI with it insertion (Test 1 vs Test 2) 1.539 0.1238 Omission of expletive and anticipatory it (Test 1 vs Test 2) 1.847 0.0648 Omission of areferential subject (Test 1 vs Test 2) 2.000 0.0455 Mean test results (Test 1 vs Test 2) 2.552 0.0107 Table 2.Level B: Comparison between Test 1 and Test 2 4.2. LEVEL B RESULTS PER PARTICIPANT It is obvious from Figure 1 that most level B students performed better on Test 2, and some students even demonstrated considerable improvement, as was the case with student 8B. On the other hand, there were students, such as students 4B and 10B, who did not make any progress at the post-test. Section 5.5 outlines some of the reasons that could potentially explain the individual variation encountered among level B students. 4.3. LEVEL C— RESULTS PER CATEGORY Table 3 shows acomparison between the scores on Test 1 and Test 2 among the students at level C.It is evident that the data indicate considerable improvement across OPEN ACCESS
NATASHA STOJANOVSKA-ILIEVSKA 75 all four categories. The scores indicate less than 20% of wrong answers to start with on Test 1 in each of the 4 categories, and this has been reduced more than twofold on Test 2 in each of the four categories. Considering p-values of less than 0.05 as statistically significant, the Wilcoxon signed-rank test demonstrates that there was significant improvement in all categories except the first one. Specifically, the p-values measuring statistical significance between the scores on Test 1 and Test 2 per category are as follows: p=0.0940 for SVI, p=0.0049 for SVI with it insertion, p=0.0041 for Omission of expletive and anticipatory it, and p=0.0249 for the Omission of areferential subject. There was also astatistically significant difference (p=0.0047) between the mean for Test 1 and the mean for Test 2. The z-values and the p-values for all categories are summarized in Table 4. SVI SVI with it insertion Omission of expletive and anticipatory it Omission of areferential subject T1 T2 T1 T2 T1 T2 T1 T2 Mean 0.14 0.06 0.16 0.02 0.20 0.06 0.14 0.02 Median 0.17 0.00 0.10 0.00 0.17 0.00 0.00 0.00 Standard Deviation 0.16 0.13 0.20 0.07 0.19 0.13 0.24 0.07 Minimum 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Maximum 0.50 0.40 0.60 0.33 0.67 0.50 1.00 0.33 Table 3.Level C: Proportion of sentences wrongly classified as correct (Test 1 vs Test 2) 0% 10% 20% 30% 40% 50% 60% 70% 1B 1B 2B 3B 4B 5B 6B 7B 8B 9B 10B 11B 12B 13B 14B 15B 16B 17B 18B 19B 20B 21B 22B 2B 3B 4B 5B 6B 7B 8B 9B 10B Average Test 1 Average Test 2 0% 10% 20% 30% 40% 50% 60% 70% Average Test 1 Average Test 2 Figure 1.Level B results per participant (Percentage of sentences wrongly classified as correct in Test 1 vs Test 2) OPEN ACCESS
76 LINGUISTICA PRAGENSIA 1/2023 Level C z-value p-value SVI (Test 1 vs Test 2) 1.675 0.0940 SVI with it insertion (Test 1 vs Test 2) 2.817 0.0049 Omission of expletive and anticipatory it (Test 1 vs Test 2) 2.873 0.0041 Omission of areferential subject (Test 1 vs Test 2) 2.243 0.0249 Mean test results (Test 1 vs Test 2) 2.827 0.0047 Table 4.Level C: Comparison between Test 1 and Test 2 4.4. LEVEL C— RESULTS PER PARTICIPANT Focusing on individual students’ results, at level C the improvement seems more remarkable, as 14 of the 22 participants (ca. 64 %) made no mistakes at all at the posttest, which is evident in Figure 2. This statement comes with acaveat that there are also several students who made no progress at all or only aslight progress on the second test (students 6C, 8C, 21C), as well as afew students who actually performed worse on the post-test (students 4C, 13C, 16C, 17C). These students had one or two wrong answers more on Test 2 than on Test 1. Possible reasons for this intra-level variation are expounded in section 5.5. Figure 2.Level C results per participant (Percentage of sentences wrongly classified as correct in Test1 vs Test 2) 4.5 COMPARISON BETWEEN LEVELS B AND C Table 5 displays the results from the Wilcoxon rank-sum test whereby level B and level C results have been compared per category, both on the pre-test and on the posttest. The data demonstrate statistically significant difference between the two levels on both tests and across all categories, except for the Omission of areferential sub0% 10% 20% 30% 40% 50% 60% 70% 1B 1B 2B 3B 4B 5B 6B 7B 8B 9B 10B 11B 12B 13B 14B 15B 16B 17B 18B 19B 20B 21B 22B 2B 3B 4B 5B 6B 7B 8B 9B 10B Average Test 1 Average Test 2 0% 10% 20% 30% 40% 50% 60% 70% Average Test 1 Average Test 2 OPEN ACCESS
NATASHA STOJANOVSKA-ILIEVSKA 83 the controlled practise activities as negative evidence, greater grammatical accuracy may be within closer reach. The statistically significant improvement in the overall acquisition of English SVI-related structures both on an intermediate (p=0.0107) and on an advanced level (p=0.0047) following instruction centred on grammar and contrastive analysis between the two languages, contrary to predominant views in the late twentieth century, lends some support to the claim that “L1 can and should have arelevant role at least in L2 grammar teaching and learning” (Teixeira 2018: 348). 5.7 LIMITATIONS The results of this study should be interpreted with caution, keeping in mind that it does not allow the inference of definitive or generalized conclusions regarding the effect of explicit instruction on the acquisition of SVI by Macedonian learners of English, due to the lack of acontrol group in the study design. The observed improvement may have resulted, at least partially, from acombination of confounding variables, the effect of which cannot be determined with certainty. Other limitations of this study include the fact that there was no long-term follow-up to verify if the knowledge related to SVI was permanent and the rather small sample size, knowing that the increase in the sample size contributes to greater statistical power (Warne 2018: 251) and greater precision of the study inferences (Miah 2016: 246). This is why random allocation of participants to an experimental and acontrol group, adelayed post-test and recruiting alarger number of participants might well be considered, should this study be replicated at some point in the future. Another idea that could be taken into consideration for some future study is assessing the students’ productive language skills in relation to SVI in English. The results from such astudy could complement the findings from this study which relied on the students’ assessment of the (non-) acceptability of given sentences. 6 CONCLUSIONS In this exploratory study of the effect of explicit instruction on the acquisition of SVI in English, Macedonian learners of English at levels B and C generally performed significantly better at the post-test than at the pre-test. This indicates that in this study explicit instruction may have led to improved assessment of the acceptability or nonacceptability of the target structure— SVI in English. Understandably, this needs to be verified in an experimental study with acontrol group. This study revealed that the effectiveness of explicit instruction depends on several factors: the cross-linguistic influence from L1; the developmental readiness of learners to acquire the target structure, which is closely related to the level of proficiency in English; and the domains involved in the target structure. As for the first factor, the different principles governing SVI in English and Macedonian inevitably contribute to the difficulties of the Macedonian learners in the acquisition of the SVI in English. Typologically, English and Macedonian are different in that Macedonian is anull subject language, while English is anon-null subject OPEN ACCESS
84 LINGUISTICA PRAGENSIA 1/2023 language and this fact is also asource of confusion for some Macedonian learners. With regard to the second factor, we saw greater progress in students at level C than in students at level B, which means that intermediate students might not be as developmentally ready to master these structures as advanced learners. And we also saw that, despite the general trend towards improvement at both levels, mastering SVI was still beyond the grasp of individual students at both levels within the twoweek time frame of the study. When it comes to the last factor, studies demonstrate that grammatical issues in the strict syntactic domain (like Omission of areferential subject) can be acquired even by learners who do not achieve native-like proficiency. However, learners find it hard to come to grips with complex issues such as SVI, which are at the interface between several domains (such as the syntax-discourse domain) and it is in these areas that learners could benefit from explicit instruction (Teixeira 2018). Although the findings from this study unambiguously point to asignificant improvement, it might be sensible to replicate this study in the future, with the addition of acontrol group, adelayed post-test and with alarger respondent sample size, so as to examine whether the findings from this study would be confirmed. Acknowledgments Iam highly indebted to Liljana Mitkovska and Eleni Bužarovska for their conceptualization of the entire study, the creation of the study-related materials and their generous support throughout the study, as well as to Iskra Stamenkoska for her practical involvement in the study as an instructor alongside myself and for her assistance with the data analysis. REFERENCES Agathopoulou, E. (2014) “Automatically Arises the Question Whether…”: ACorpus Study of Postverbal Subjects in L2 English. In: Lavidas, N., Th. Alexiou and A.M.Sougari (eds) Major Trends in Theoretical and Applied Linguistics 2, 169–186. London: Versita. Biber, D., S.Johansson, G.N.Leech, S.Conrad and E.Finegan (2021) Grammar of Spoken and Written English. Amsterdam: John Benjamins. Bowles, M. and S.Montrul (2008) The Role of Explicit Instruction in the L2 Acquisition of the a-personal. In: de Garavito, J.B. and E.Valenzuela (eds) Selected Proceedings of the 10th Hispanic Linguistics Symposium, 25–35. Somerville, MA: Cascadilla Proceedings Project. Bužarovska, E., L.Mitkovska and A.KirkovaNaskova (2019) Zbororedot vo prostata rečenica: inverzija na subjektot ipredikatot. Journal of Contemporary Philology 2, 7–25. https://doi.org/10.37834/JCP192007B Judy, T. and J.Rothman (2010) From asuperset to asubset grammar and the semantic compensation hypothesis: Subject pronoun and anaphora resolution evidence in L2 English. In: Franich, K., K.M.Iserman and L.L.Keil (eds) BUCLD 34: Proceedings of 376 the 34th annual Boston University Conference on Language Development, 197–208. Somerville, MA: Cascadilla Press. Krashen, S. and T.Terrell (1988) The Natural Approach: Language acquisition in the classroom. New York: Prentice Hall. Levin, B. and M.Rappaport Hovav (1995) Unaccusativity at the Syntax–Lexical Semantics Interface. Cambridge, MA: The MIT Press. Lozano, C. and A.Mendikoetxea (2010) Postverbal subjects in L2 English: OPEN ACCESS
NATASHA STOJANOVSKA-ILIEVSKA 85 Acorpus-based study. Bilingualism: Language and Cognition 13/4, 475–497. Marsden, E. and C.J.Torgerson (2012) Single group, pre-and post-test research designs: Some methodological concerns. Oxford Review of Education 38/5, 583–616. Miah, A.Q. (2016) Applied Statistics for Social and Management Sciences. Singapore: Springer. https://doi.org/10.1007/978-981-10-0401-8 Minova-Ǵurkova, L. (2000) Sintaksa na makedonskiot standarden jazik. Skopje: Magor. Mitkovska, L. (2011) Makedonskite se-konstrukcii inivnite ekvivalenti vo angliskiot jazik. Skopje: Makedonska reč. Mitkovska, L. and E.Bužarovska (2018) Subject pronoun (non)realization in the English learner language of Macedonian speakers. Second Language Research 34/4, 463–485. Mitkovska, L., E.Bužarovska and N. StojanovskaIlievska (forthcoming) Acquisition of Structures at Syntax-Discourse Interface: Post-Verbal Subjects in L2 English. Theory and Practice of Second Language Acquisition. Norris, J.M. and L.Ortega (2001) Does Type of Instruction Make aDifference? Substantive Findings from aMeta-Analytic Review. Language Learning 51/1, 157–213. Prado-Alonso, C. (2011) Full-verb inversion inwritten and spoken English. Bern: Peter LangAG. Prentza, A. and I.Tsimpli (2013) The Interpretability of Features in Second Language Acquisition: Evidence from Null and Postverbal Subjects in L2 English. Journal of Greek Linguistics 13, 323–365. Rahimpour, M. and A.Salimi (2010) The impact of explicit instruction on foreign language learners’ performance. Procedia Social and Behavioral Sciences 2, 1740–1746. https://doi. org/10.1016/j.sbspro.2011.11.378. Rutherford, W.E. (1989) Interlanguage and pragmatic word order. In: Gass, S.M. and J.Schachter (eds) Linguistic perspectives in second language acquisition, 163–182. Cambridge: Cambridge University Press. Slabakova, R. (2015) The effect of construction frequency and native transfer on second language knowledge of the syntax–discourse interface. Applied Psycholinguistics 36/3, 671–699. Slabakova, R. and I.Ivanov (2011) Amore careful look at the syntax–discourse interface. Lingua 121/4, 637–651. Sorace, A. (2003) Near-nativeness. In: Doughty, C.J. and M.Long (eds) The handbook of second language acquisition, 130–151. Oxford: Blackwell Publishing. Sorace, A. (2011) Pinning down the concept of ‘interface’ in bilingualism. Linguistic Approaches to Bilingualism 1, 1–33. Sorace, A. and F.Filiaci (2006) Anaphora resolution in near-native speakers of Italian. Second Language Research 22, 339–368. Sorace, A. and L.Serratrice (2009) Internal and external interfaces in bilingual language development: Beyond structural overlap. International Journal of Bilingualism 13/2, 195–210. Swain, M. (1985) Communicative competence: Some rules of comprehensible input and comprehensible output in its development. In: Gass, S. and C.Madden (eds) Input in second language acquisition, 235–253. Rowley: Newbury House. Swain, M. and S.Lapkin (1982) Evaluating bilingual education. ACanadian case study. Clevedon: Multilingual Matters. Swan, M. (1985) Acritical look at the Communicative Approach (1). ELT Journal 39/1, 2–12. Teixeira, J.A. (2018) L2 acquisition at the interfaces: Subject-verb inversion in L2 English and its pedagogical implications. Doctoral dissertation. Universidade Nova de Lisboa. Teixeira, J. (2020) Gradient optionality in L2 acquisition at the syntax-discourse interface: Evidence from inversion in advanced and near-native English. Lingua 245/102947. https://doi.org/10.1016/j.lingua.2020.102947. Terrell, T. (1991) The role of grammar instruction in acommunicative approach. Modern Language Journal 75, 52–63. Trahey, M. and L.White (1993) Positive evidence and preemption in the second language classroom. Studies in Second Language Acquisition 15, 181–204. OPEN ACCESS
86 LINGUISTICA PRAGENSIA 1/2023 Tsimpli, I. and A.Sorace (2006) Differentiating interfaces: L2 performance in syntax– semantics and syntax–discourse phenomena. Proceedings of the 30th Annual Boston University Conference on Language Development, 653–664. Somerville, MA: Cascadilla Press. Valenzuela, E. and B.McCormack (2013). The syntax-discourse interface and the interface between generative theory and pedagogical approaches to SLA. In Whong, M., K.-H. Gil, and H.Marsden (eds), Universal Grammar and the second language classroom 101–114. Dordrecht: Springer Netherlands. VanPatten, B. and J.Rothman (2015) What does current generative theory have to say about the explicit-implicit debate? In: Rebuschat, P. (ed) Implicit and explicit learning of languages, 89–116. Amsterdam: John Benjamins Publishing. Warne, R.T. (2018) Statistics for the Social Sciences— AGeneral Linear Model Approach. Cambridge: Cambridge University Press. White, L. (1991) Adverb placement in second language acquisition: Some positive and negative evidence in the classroom. Second Language Research 7/2, 133–161. Whong, M., K.-H. Gil and H.Marsden (2014). Beyond paradigm: The ‘what’ and the ‘how’ of classroom research. Second Language Research 30/4, 551–568. Natascha Stojanovska-Ilievska Ss. Cyril and Methodius University in Skopje Faculty of Philology, Department of English Language and Literature ul. Goce Delcev 9a, 1000 Skopje, North Macedonia ORCID ID: 0000-0003-1221-0999 [email protected] APPENDIX Sample questions (some of which are distractors) extracted from the grammaticality judgment and correction tasks (GJCT)2: 1. In example (5) instead of Present Continuous is used Present Simple Tense. 2. Here is very hot and the sea is blue and lovely. 3. On the position of post-modifiers can stand either phrases or clauses. 4. We can hear “h” when is in initial position, for example: house. 2 The GJCTs were compiled by Liljana Mitkovska and Eleni Bužarovska. OPEN ACCESS
NATASHA STOJANOVSKA-ILIEVSKA 87 5. All uses of the Present Simple Tense are examined in this study. 6. Sometimes is hard to read and understand English poetry. 7. So it should be used the adjective comfortable instead of comfortably. 8. Here in Bitola is very nice, and the people are kind, too. 9. Ihave studied the present perfect tense in English and how is translated in our language. 10. In some schools, beside English and German it is also studied French. 11. It is quite obvious the omission of the definite article ‘the’. 12. In this paragraph through examples will be illustrated which of the verbs of thinking have finite complement clauses. 13. The grass was soft like sand and it was hard to walk on it. 14. Tomorrow are coming my grandmother and grandfather, because it’s my birthday. OPEN ACCESS