L2 Texts Research Articles

How does lexical coverage affect the processing of L2 texts?

Abstract Lexical coverage, i.e. the extent to which words in a text are known, is considered an important predictor of reading comprehension, with studies suggesting 98% lexical coverage leads to adequate comprehension. However, no studies to date have examined how the various lexical coverage percentages suggested in the literature are reflected by the cognitive effort involved in processing text and the attention that is devoted to the unknown vocabulary. This study used eye-tracking to examine how lexical coverage affects the processing of text (global measures) and unknown vocabulary (word-level measures), as well as the relationship between processing time on unknown vocabulary and learning. Advanced L2 learners of English read a text in one of four lexical coverage conditions (90%, 95%, 98%, 100%) while their eye movements were recorded. Knowledge of unknown pseudowords in the texts was assessed via an immediate, meaning recall post-test. Results showed that only one of the three global measures examined showed a processing advantage for the 98% condition, reflected by longer saccades and less effortful reading than the 90% and 95% conditions. Crucially, lexical coverage did not have a significant impact on the amount of attention spent on unknown vocabulary. Processing times were found to significantly predict vocabulary gains.

Stance-taking through APPRAISAL in L1 and L2 English argumentative essays: insights from Vietnamese L2 English

Abstract Stance-taking in academic writing is now a well-established topic of interest in English language research, with explorations across a range of theoretical frameworks including metadiscourse and systemic functional linguistics (SFL). This includes studies investigating second language (L2) stance-taking, particularly those comparing stance features deployed by L1 and L2 English writers. However, studies investigating stance-taking using the APPRAISAL framework for evaluative discursive language across L1 and L2 production are relatively rare. Incorporating the APPRAISAL framework into research on stance-taking enhances our comprehension of evaluative language in academic writing, especially when it comes to cross-linguistic contexts. It also provides useful advantages for language assessment and instruction. In this learner corpus-assisted discourse study, APPRAISAL was used to determine how L1 English speakers and L2 English learners from L1 Vietnamese backgrounds expressed attitudes through their written texts. We also investigate the relationship between use of APPRAISAL resources and expert raters’ perceptions of written stance via a stance rubric. Findings show L2 English students are more explicit in argumentative writing than L1 English writers, despite fewer APPRAISAL choices in L2 texts. Besides, while high-rated texts were associated with more judgemental evaluation and invocation, more personal feelings were expressed in low-rated texts. These findings have implications for the instruction of L2 writers in conveying attitudinal meanings in text, as well as for raters tasked with assessing L2 academic essays for stance.

System networks as a resource in L2 writing education

Abstract This paper argues for the value of system networks as a diagnostic tool in L2 writing education. We report on an investigation of Chinese students learning to write in English over a period of a year in Year 9 of high school, characterizing their choices in the system of modality in different writing tasks over that period, and we also report on an extension of this investigation based on two additional sampling points, Year 12 of high school and Year 3 of university. Our findings indicate that even when we sample L2 texts from three different periods in education, the expansion in the use of the meaning potential embodied in the system of modality is quite limited – as long as we focus on texts from one register. We have also seen from Xuan’s (2015. A longitudinal study of Chinese high school students learning English based on systemic functional text analysis. The Hong Kong Polytechnic University PhD thesis) study of Year 9 writing tasks from a few different registers that there was no clear sense of progression throughout the year. The results show that careful selection and ordering of texts from registers that afford learners with complementary opportunities over time to engage with and master different parts of the resources of the system of modality can be recommended in L2 writing education.

Examining student translators’ writing and translation products: quality, errors and self-perceived mental workload

ABSTRACT Although writing and translation are closely related text productions, their interface has rarely been studied in translator training. This study examined student translators’ writing and translation products in terms of their quality, errors and self-perceived mental workload. Data were collected from 11 intermediate-level translation students at a university in mainland China. They were asked to complete four tasks: L1 (Chinese) writing (L1W), L1 (English-Chinese) translation (L1T), L2 (English) writing (L2W), and L2 (Chinese-English) translation (L2T). The results of quantitative analyses revealed no significant linear correlation between self-perceived mental workload and quality/errors, but a significant effect of task type on mental workload. Errors most frequently occurred in translation and L2 text production. The L2W error rate predicted L2T scores and L2T error rates. A discrepancy was noted between the students’ self-evaluation and their actual performance. The findings of this study may help researchers and teachers gain a fine-grained understanding of students’ performance in writing and translation and shed light on the development of pedagogical measures in translator training.

Learners’ reception of translanguaging as a model for reading English L2 texts

Language is the mediator of two critical aspects of a child's education: identity and knowledge. Through language, learners can make sense of who they are and make sense of the world around them. However, the teaching of English second language in several schools in South Africa is such that a learner's identity, best described in their own indigenous languages, is not mirrored in the classroom and, in its place, a monolingual system (English-only) becomes central. Their acquired linguistic repertoire as multilingual learners is considered a hindrance to learning, not a resource. This paper assesses learners’ reception of translanguaging pedagogy as a model for reading English texts. It further seeks to take the debate about the paradox of translanguaging forward to determine if translanguaging is a unitary, differentiated system, or both. Exploring data in the form of class observations and interviews, the study reveals that translanguaging as a model for reading English L2 texts is well-received by learners as a practical and valuable pedagogical approach that can potentially improve English L2 performance. The results further establish the existence of both the unitary and the differentiated systems of translanguaging. In the end, the study's recommendations for transforming multilingual classrooms are discussed.

The development of sentence and clause lengths in Czech L2 texts

The development of sentence and clause lengths in Czech L2 texts

Theoretical Considerations on AI-based Business Models for Lexicography

Abstract AI-generated text production is on the rise Zandan (2020), and AI writers seem to be playing an increasingly important role in marketing, L2 text production, lexicography and language teaching, cf. Simonsen (2020b; 2021a; 2022a; 2022b; Sharples/Pérez Y Pérez 2022; ChatGPT 2023; RYTR 2023; Writewithlaika 2023). In addition to that large national language datasets are being developed in many countries, cf. for example Kirchmeier et al. (2020), and these national word registers are expected to become an important backbone in AI-based lexicographic services. On this background, there seems to be a need for AI-based business models for lexicography. This article draws on a literature review focussing on business models and business-related considerations of relevance for lexicography. The insights from the literature review led to the development of a number of theoretical considerations on AI-based business models for lexicography. The article suggests three AI-based business models for different types of lexicography and demonstrates how these business models could be implemented in three concrete projects.

Inferential evaluation and revision in L1 and L2 text comprehension: An eye movement study

Abstract Text comprehension frequently demands the resolution of no longer plausible interpretations to build an accurate situation model, an ability that might be especially challenging during second language comprehension. Twenty-two native English speakers (L1) and twenty-two highly proficient non-native English speakers (L2) were presented with short narratives in English. Each text required the evaluation and revision of an initial prediction. Eye movements in the text and a comprehension sentence indicated less efficient performance in the L2 than in L1 comprehension, in both inferential evaluation and revision. Interestingly, these effects were determined by individual differences in inhibitory control and linguistic proficiency. Higher inhibitory control reduced the time rereading previous parts of the text (better evaluation) as well as revisiting the text before answering the sentence (better revision) in L2 comprehenders, whereas higher proficiency reduced the time in the sentence when the story was coherent, suggesting better general comprehension in both languages.

Exploring collocational complexity in L2 Russian: A corpus-driven contrastive analysis

Objective: The objective of this article is to discuss the pedagogical and practical need for automated assessment tools that enable teachers, researchers, and other language practitioners to relatively quickly and automatically assess the general proficiency of second language (L2) speakers according to a number of different linguistic parameters, specifically the use of collocations. Introduction: The Introduction discusses existing approaches to the concept of collocation and its role in L2 corpus studies. Methods: In the main part, Toward the automatic detection of collocational strength, we describe an approach to assessing collocations in Russian L2 written production, which is based on the corpus-driven contrastive analysis. In brief, this approach proposes ranking L2 collocations in accordance with the “collocational scores” of bigrams extracted from a reference (L1) corpus, with the assumption that the greater number of native-like bigrams in an L2 text and the higher collocational scores associated with them, the more proficient is the learner who created the text. To test this hypothesis, we conducted a series of experiments based on an L1 reference corpus and split the L2 datasets into four proficiency groups (Beginner, Intermediate, Advanced, and Superior). Results and evaluation: The findings, presented in the section Results and evaluation of the methods, established an empirical minimum and maximum of mean reciprocal rank (MRR) values, and the variation of the MRR across the four proficiency levels. In the last sections, we discuss shortcomings of the methods and propose further steps for overcoming them. We believe that the proposed method for assessing the collocational skill of Russian L2 writers can be successfully integrated into a bigger system capable of multifaceted automatic assessment of Russian learner texts.

Chunking in simultaneous interpreting: the impact of task complexity and translation directionality on lexical bundles.

This study explored the use of phraseological frames (p-frames), a type of lexical bundle, by simultaneous interpreters as a strategy for managing cognitive loads. Specifically, using a comparable corpus of United Nations Security Council conferences, the study employed kfNgram to automatically identify the p-frames, and investigated their variations, regarding frequency, fixedness, structures, and functions among L1, L1-L2, and L2-L1 texts, which differ in cognitive loads due to task complexity and directionality of interpreting. The findings indicated that interpreters used more p-frames as cognitive loads increased; No significant difference was identified in fixedness as all texts tended to employ heavily formulaic and relatively fixed p-frames. Directionality correlated with grammatical preferences, with retour interpreting relying more on content-word-based p-frames. Additionally, task complexity correlated with functional preferences, with self-expression characterized by more stance expressions in the simple task. This study innovatively addressed the interaction of two factors that generate cognitive loads in interpreting and filled a research gap by providing empirical evidence on how directionality affects the use of formulaic language.

På, i, for, or til: A comparative analysis of prepositions in the writing of L1 and L2 Danish users

Abstract This corpus study compares the use of Danish prepositions in the writing of L1 users of Danish to that of L2 users of Danish (with English as their L1). It examines to what extent non-standard usage is a specific L2 phenomenon and to what degree the challenges with prepositions are shared between the L1 and L2 group. We analyze the distribution and characteristics of non-standard usage involving four frequent Danish prepositions: til ‘to’, i ‘in’, på ‘on’, and for ‘for’. The results show that L2 texts are characterized by a larger share of non-standard tokens and by confusion of prepositions and non-prepositions. Yet, both L1 texts and L2 texts have non-standard usage and both have possible English calques. Based on the similarities between the two groups, we recommend that future studies include naturally occurring L1 production as a baseline for describing characteristics of L2 production.

Exploring the Impact of Interlinear vs. L1 Marginal Glosses on Iraqi EFL Learner’s L2 Vocabulary Learning

This study investigates the impact of interlinear vs. L1 marginal glosses on the Iraqi Learning of vocabulary in EFL instructional setting. Along with all the efforts made during past 3 decades to improve the teaching of vocabulary via various strategies and techniques countywide, the classical grammar-translation method and its offspring are still widely used in most schools in Iraq. By analyzing the elicited data out of the management, treatment, and assessment of 92 pre-university students studying in two high schools in Iraq, the inquirer observed a reliable improvement of vocabulary learning in participants. To examine the effectiveness of the two glossing strategies, i.e., the implantation of L1 marginal and interlinear glosses, the results of pre and post-test of three groups were compared using One-Way ANOVA. We found reliable difference between the two groups in post-test vocabulary learning. Consequently, L1 marginal and Interlinear glosses both have come to make the learners understand L2 texts well, in comparison with the other group. Ultimately, Marginal glossing is more effective than the interlinear (text) glossing in the improvement of vocabulary development of Iraqi EFL learners.

Students’ Need to Increase Ethically Acceptable Paraphrasing of Academic Texts

The lack of Indonesian students performing their abilities in L1 paraphrasing becomes the primary reason they are labeled as persistent plagiarizers. Students, in Indonesian context and circumstances, need to broaden their perspectives regarding how to document the sources, using them in their original papers to support and defend the arguments. Literacy should be developed sustainably in order for students might critically think and state things with evidence. However, to say that students still developing English writing text as intentional plagiarism will appear to be overly harsh. Teachers, at the university level, should acknowledge that students’ commitment to plagiarism is the process of knowing its definition, characters, impacts, and habit of copying others’ work without any attempt to document it properly. Moreover, it also signs the teachers that students’ ability to use academic English needs improving. Observing these facts urges us to facilitate students to make them ready to conduct the paraphrase of the L2 texts while their EAP is still developing. This study seeks the fundamental preparation that should be made to generate acceptable paraphrasing texts.

The Influence of L1(Chinese) on Writing in L2(English): A Case Study

This paper aims to explore the influences of L1 in the aspect of writing in L2, and it serves as a qualitative case study to disclose how Chinese as L1 influences writing in English (L2) in the context of IELTS writing exercises as part of mock exams. The influences of concern include: First, the study objects habitual or frequent misuse and mistakes of the L2 resulting from L1s existing language system; these errors can be grammatical, lexical, syntactic, structural, word choice, tense, etc.; and second, how the study objects previous experience of writing similar texts in L1 is extending its influence in L2 writing. The data were collected from students who used to prepare for IELTS academic tests to achieve the ends mentioned above. Writing tasks prepared for IELTS mock exam were collected for error collection and in-depth word-by-word analysis. The data for the study were collected after the IELTS exam was taken; these essays were part of the preparation for the IELTS exam. Therefore, the analysis can be used as a good indicator of how students perform and write formally in an academic context, not to mention the pressure experienced during the exam. The data analysis shows that there is a wide range of mistakes or misuses in writing in L2, all of which show a link to the interference or transfer of L1 to writing in L2 in aspects of grammar, semantics, and syntax. The interference results in ungrammatical or broken English. The analysis also reveals that the differences between L1 and L2 language systems may have resulted in some common errors and the so-called Chinglish. On the one hand, the linguistic habits and thinking habits of L1 possibly interfere with writing in L2. On the other hand, students unfamiliarity with the English language, including grammatical rules, semantic structures, and correct usage of words and phrases, also contributed to reduced efficiency or miscommunication of writing in L2. Based on the above-mentioned findings, suggestions are made for students on methods to improve L2 acquisition and application.

OPUS corpus toolkit for ensuring intelligent translation (case study of L1 and L2 texts of English-Ukrainian film discourse)

The article explains the concept of “translation memory” and defines it as a computer database where segments of texts of different L1 discourses are represented, as well as equivalents of these segments in L2. Computer-Aided Translation, Machine Translation and Parallel corpus toolkit are outlined as the main types of translation memory. In particular, Computer-Aided Translation is considered as the process of translating L1 text to obtain L2 by using specialized computer software. In this way, the human factor plays one of the most important missions in the process of performing Computer-Aided Translation, because the L1 text is subjected to three types of processing: pre-, inter- and post-editing. Machine Translation is viewed in a narrow sense as the process of translating a text from L1 to L2, that is performed by a computer in whole and/or in part, and in a broad sense as a branch of scientific research, that is in the focus of Linguistics, Mathematics and Cybernetics, and aims to build a system that implements Machine Translation in the narrow sense of this concept. Parallel corpus toolkit is a database with a set of L1 and L2 texts, that contains a large number of texts of different discourses, issues and topics. In addition, the attention is paid to the OPUS corpus toolkit as one of the translation memory types, which ensures the efficiency of the process of intelligent translation and is currently a free corpus system in the public domain, which contains corpora of texts from L1 and L2 to L3...Ln from numerous Internet resources and is constantly updated. The tested resource capabilities of the OPUS corpus tool have proved their effectiveness in the process of verification of one-, two-, and three-component L2 lexical constructs on the example of L1 and L2 text fragments belonging to film discourse.

Non-selective language activation in L2 lexical inference and text comprehension: Comparing skilled and less-skilled readers

Non-selective language activation refers to the automatic co-activation of L1 and L2 information. In L2 reading, the activated L1 information can be utilized to different degrees to facilitate lexical inference and text comprehension. The current study examined the contributions of L1-L2 translation and lexical inference to text comprehension. Hierarchical regression models showed that in general, lexical inference contributed to text comprehension over L1-L2 translation. The results indicated that L2 learners did not use activated L1 information mechanically. That is because successful lexical inference incorporates learners’ ability to strategically utilize contextual information and integrate word meanings to update the context. The study further classified the participants into two groups using k-means cluster. Among the less skilled group of participants, L1-L2 translation was related to both lexical inference and text comprehension. However, lexical inference was not significantly related to text comprehension. Among the more skilled group of participants, lexical inference predicted text comprehension only after school, grade to start English learning, and L1-L2 translation were controlled for. The results of the two groups demonstrated that while L1 information was utilized in both groups, strategic and effective usage of information in two languages differentiated skilled L2 readers from less skilled L2 readers.

Integrated writing pedagogy in the EFL classroom

The following article discusses something which is frequently perceived as nothing more than homework by many instructors of English as a Foreign Language (EFL) for its time-consuming nature: second language writing. The first section outlines some issues that students of English often experience when taking English-language classes. Here, references to specific cultures are made, but the problems can also be universally considered. The subsequent section describes two specific approaches to writing instructors can adopt in their own classrooms and evaluates their effectiveness. The third and final section details how instructors may go about borrowing aspects of the two approaches to form an integrated approach aimed at helping students to overcome the problems they face in composing L2 texts. Here, procedural outlines are included so that instructors can take this integrated writing pedagogy into their own EFL classrooms.

Understanding the complexity and dynamics of web-based pre-class peer discussions experienced by L2 readers: a qualitative case study

ABSTRACT This qualitative case study investigated L2 readers’ semester-long experiences with web-based pre-class peer discussions (WPPD). Data were collected from a university L2 reading course and included field notes on 21 students’ web-based learning activities and in-class learning activities, interviews with these students, and their reflections. Through a qualitative analysis of these data sources, the study revealed that the L2 readers felt that the WPPD created an engaging peer-learning platform for their pre-class learning. The students also felt that their WPPD, buttressed by their instructor’s in-class assistance, gradually facilitated their pre-class navigation of a new text, as exemplified by their questions and answers about the deep meaning of the text. The study concludes that despite the L2 readers’ individual differences regarding their need for teacher assistance, they perceived WPPD as useful for their pre-learning activities motivating them to engage in peer learning and develop subject knowledge, especially in L2 contexts, where understanding the deep meaning of L2 texts at the content level is emphasized.

Comparing the effectiveness of study strategies on comprehension, retention, and learning from L2 English texts

This study compared the use of the study strategies rereading, highlighting, and note-taking on the levels of comprehension, retention, and learning from L2 texts. Nineteen intermediate students of English (L2) participated in the study. In phase 1, they studied three expository texts in English, each with the support of a different strategy. After reading, they answered an immediate recall and true/false statements. Phase 2 took place a week after and comprised delayed recalls and a critical writing task. Prior to data collection, participants received instruction on study strategies to ensure they knew about the strategies to be worked with. Results of immediate recalls pointed to rereading as an effective strategy to comprehension; highlighting was related to higher true or false scores. Regarding retention, good performance in the delayed recalls was associated with the highlighting and note-taking conditions. Thus, the effects of rereading did not endure delayed tests, providing evidence for the ineffectiveness of this strategy for retention compared to highlighting and note-taking. Results from the critical writing task demonstrated that the task fostered elaborative inferencing, although the number of explicit text references was small. A link between highlighting and learning is hypothesized.

Genre effect on Google Translate–assisted L2 writing output quality

AbstractThis study investigates and compares the quality of Google-translated texts (GTTs) across writing genres (narrative, descriptive, expository, and persuasive) with EFL student–generated texts (SGTs), and explores students’ attitudes toward Google Translate (GT) output. In a mixed-methods design with a computational approach to text analysis, this study utilizes multiple data sources, including 328 written texts, written reflections, four attitude questionnaires, focus group discussion, and individual interviews. Forty-one Arabic-speaking undergraduate students majoring in English language and translation at a Saudi university participated in this study. They engaged in eight computer-mediated writing sessions by responding to tasks using English, then responding to the same tasks using their first language (L1). Subsequently, they utilized GT to translate L1 texts into English and compared SGTs with GTTs. The findings show that GTTs have higher literacy levels and richer content in the persuasive and expository genres, and higher style levels in the narrative and descriptive genres. Moreover, the comparison between SGTs and GTTs reveals that GTTs have higher literacy levels, better style, and richer content in the descriptive, expository, and persuasive genres. Meanwhile, the students hold positive views on the general quality, grammatical accuracy, and provision of lexical alternatives in GTTs across genres. Pedagogical implications are discussed.