Learner Corpora And Language Teaching

The book under review, Learner Corpora and Language Teaching, is edited by Sandra Götz and Joybrato Mukherjee, introducing the latest research in learner corpus studies and its value in language teaching. The volume is a collection of research papers originally presented at the 12th Teaching and Language Corpora Conference in Giessen in 2016, on the theme of combining language learning and teaching with the utilization of corpora. This volume is well organized with four thematically defined sections, namely, New learner corpora and tools, Written learner corpora and language teaching, Spoken learner corpora and language teaching, and Learner corpora and language teacher education. Despite a long history of corpus-based teaching, the specific learner corpus is rarely used in teaching. Each section contains a number of research projects in which the use of learner corpus materials and methodologies is essential. Authors are established or young scholars from Europe and Asia, and the language explored ranges from English to Japanese. In the situation that ‘learner corpora are still at a periphery of language teaching’ (Flowerdew, 2012), a distinctive feature of this volume is that it focuses on the value of learner corpus, both spoken and written, in the application into language teaching.

The papers collected in this volume present important contributions to methodological innovation in learner corpus studies. Section I starts with the introduction of a newly compiled spoken corpus, The Trinity Lancaster Corpus (TLC) as a case to be studied by Dana Gablasova, Vaclav Brezina and Tony McEnery (p. 7), who are trying to fill the gap between learner corpus and its direct application in language pedagogy. As each corpus material in TLC is marked based on the band of A, B, C, and D, with A indicating the highest performance and D a fail. So proficiency in spoken English can be compared to benefiting both students and teachers to understand successful communication. Furthermore, the corpus features various sociolinguistic characteristics of a corpus, like age, gender, education, learning experience. All these variables could be studied to explore their effects on language learning.

It is found in the first case study that with increasing L2 proficiency, the disagreements tend to be accompanied by more politeness markers to form more complex politeness strategies. This is the evidence indicating which resources could be more suitable to speakers at each level proficiency levels and corresponding pedagogical intervention may be appropriate for a targeted group of speakers. Another finding in the second case is that the advanced L2 speakers adjust their use of (un)certainty markers according to linguistics settings, while with several learners applying the same approach irrespective of the demands of the task. This suggests that intervention at both lower and higher levels of proficiency may be necessary even for advanced speakers. The third case study states a relationship between the overall performance of the learners and their listenership strategies. This indicates that signals of active listenership can be taught to elevate learners’ overall understanding and awareness of the conversation principles. These three cases demonstrate explicitly how findings from learner corpora can be directly used in materials design and how examples from these corpora can be integrated into classroom exercises.

As the second part of Section I, an automated tool for rating EFL essays is introduced by Olga Vinogradova, with a case study of a learner corpus named Russian Error-Annotated Learner English Corpus (REALEC). The corpus is compiled with two genres of graphical materials and argumentative essays in English by first- and second-year Bachelor students at different departments in four cities in Russia. Errors in the corpus are annotated manually by experts and performed in BRAT, an open-access user-friendly web-based tool. This paper described how to use BRAT thoroughly, from error identification to error classification, and error correction to annotation display. With the computer technology of automatic item generation (AIG), the annotated errors of the corpus are input into RETM-REALEC English Test Maker to generate test questions. These two pools of questions would provide a large number of choices for examiners to choose from. After being edited by instructors, questions are grouped into four types of tests. Both correct and wrong answers together with expected corrections would be automatically displayed in the interface of Moodle after the tests. The result shows that editing generated questions is of great necessity due to various special cases. The author also rationally analyses to what extent this process is useful for EFL instructors.

Section II of the volume focuses on the application of learner corpus in investigating the overuse or underuse of certain linguistic patterns in English by non-native speakers. Some of them select keywords as an indicator to analyze the use of the English language by native and non-native speakers, like the word ‘TAKE’ in Albert Biel’s paper (p. 51), and countable nouns, prepositional phrases, verbs and general adverbs in Pascual Perez-Paredes & Maria Belen Diez-Bedmar’s (p. 101); and some take syntactic feature as a parameter to calculate log-likelihood of learners’ English to native English, like linking adjuncts in Meredith D’Arienzo’s research (p.75), and direct quotes in Leonie Wiemeyer’s (p.129). Albert studies the use of ‘TAKE’ in German learners in learner corpus ICLE and comparable American English Speaker Corpus (LOCNESS), finding that non-native speakers do not overuse simple verbs due to restricted vocabulary as generally considered. Similar to the methodology as Albert adopts, Pascual Perez-Paredes and Maria Belen Diez-Bedmar take multiple keywords to analyze learner language complexity, and it is concluded that noun phrase is of great interest in identifying achievements in language acquisition. Except for words-based study, scholars also take advantage of syntactic features in their studies. Combining quantitative and qualitative methods, Meredith argues that EFL learners do perform universal variation in that both over- and underuse linking adjuncts appears in the study. And pedagogical interventions are proposed for Italian learners to acquire cohesive devices. Leonie studies the intertextuality of L2 learners, suggesting that in spite of the familiarity of direct quotation and conventions, teaching should enhance learners' knowledge in length, syntactic integration, and purpose of direct quotes.

Save your time!
We can take care of your essay

• Proper editing and formatting
• Free revision, title page, and bibliography
• Flexible prices and money-back guarantee

Place Order

As a paralleled content, Section III is on the topic of spoken learner corpora and language teaching. Compared to written corpora, spoken corpora is another mode of language output. Will EFL learners be affected by these two different modes in terms of language accuracy and error rate? Mariko Abe finds that third-person singular - s may be problematic for Japanese learners as it does not improve in the less time-pressured written production mode compared to spoken mode. To test the expression features of German English learners, Anna Rosen takes 18 smallswords as indicators to compare learners with and without studying abroad experience. The result supports the previous study that learners benefit from the exposure to natural language environment. Besides linguistic features, spoken corpora together with the recordings contain more information on paralinguistic cues, like speech rate, tone, etc., which are potential resources for scholars to explore. In the research of Tomas Graf and Karin Puga, they study speech rate and edge tone respectively based on a spoken corpus, both of which explore possible and novel approaches for spoken corpus research as well as for spoken language assessment.

Learner corpora can be used to provide negative evidence for language learners to identify common and persistent errors, meanwhile, learner corpus data used in Data Driven Learning (DDL) activities can increase future teachers’ abilities to notice and evaluate errors. Section IV links learner corpus with teacher education to complement the gap between popularization of corpus and few trained teachers in corpus linguistics (Mukherjee, 2004). Marcus Calie (p. 245) reviews previous findings on corpus literacy and present data, sorts out a research and teaching perspective of corpora use, finding that the former remains under-addressed in language teacher education. Therefore, the author proposes an added value of learner corpus, that is, learner corpus could narrow the distance between teacher education and their career reality.

As a whole, the volume presents diverse approaches to deploy learner corpus in language learning and teaching. It explores multiple purposes that scholars could realize with the use of learner corpus. To be specific, diverse learner corpus study patterns start to emerge from the processing and analysis of the corpus data used. One major pattern is driven by distinctive linguistic features between advanced and lower learners, between L2 learners and native language corpus, which results in further exploration of how to improve language learning accordingly. These features include, but not limited to, lexical markers (Dana Gablasova, Vaclav Brezina and Tony McEnery in Section I, Pascual Perez-Paredes & Maria Belen Diez-Bedmar in Section II, Mariko Abe, Anna Rosen in Section III), syntactic structures (Albert Biel, Meredith D’Arienzo, Leonie Wiemeyer in Section II). Except for linguistic feature in common among spoken and written corpus, it is noted that scholars are researching on paralinguistic features of a spoken learner corpus. Such features of fluency, tone and speech rate, are undoubtedly vital for language learning and use. And these studies (Tomas Graf, Karin Puga in Section III) expand the use of learner corpus above traditional word analysis. Another pattern is about designing the teaching intervention process, e.g. exam questions, teacher training method, based on learner corpus (Olga Vinogradova in Section I, and Marcus Calie in Section IV).

Another shining point of the volume covers a comprehensive range of corpus types and learner corpus applications in the perspective of learners and teachers. Both spoken and written learner corpus studies are discussed in the volume, while with a deficiency of an overlapping content of Section I and Section III, in terms of the topics on spoken learner corpus. And readers may get confused about unclear classification of Section I and Section IV, in that the second paper in Section I is about how teachers could use learner corpus to design exam questions, which should fall into the category of Section IV on the topic of training teachers to apply learner corpus as teaching tools in general sense. In addition, authors in the volume also introduce lots of innovative corpus tools and learner corpus resources which are compiled in recent years, which inject new nutrition for the development of learner corpus studies.

Except for the mentioned advantages and convenience of learner corpus in language teaching, it is also critically demonstrated that several issues in the pedagogical application of the learner corpus. One is the inability of a corpus in presenting all paralinguistic cues when transforming spoken language into written transcriptions. Therefore, it is necessary to clarify what features of spoken communication will be captured in the transcript. And these features will play a role in the analysis (Adolphs & Knight, 2010). The other issue is about the variable controlling in the comparison of the different corpus. In an ideal situation, the corpus to be compared with should bear similar, if not the same, features of data in all respects. Otherwise, the comparison would be unreliable.

As a whole, Learner Corpora and Language Teaching is a highly informative and illuminating collection of latest studies on learner corpus and its application in language learning and teaching. Its innovative perspective, explorative efforts, rich learner corpus resources, critical thinking make it a truly valuable book for teachers, researchers and language learners. The research approaches, fresh corpus resources and inspiring findings are paving the way for future studies.