The Use of Learner Corpora in Teaching Foreign Languages: Source of Information About Common Mistakes - Literature review Example

 Over the last forty years, great strides have been made in terms of production of numerous corpora and user friendly software that is easily accessible to linguists, but these have not been widely used by teachers in language classrooms (Karin, 2009). O’Keeffe et al have defined a corpus as “a collection of texts, written or spoken, which is stored on a computer.” (O’Keeffe et al, 2007:1). It is not however, any old collection of texts, but must represent something because its merit will be judged on the basis of how representative it is of the subject of the corpus. Corpora could be general, such as the Bank of English corpus, or more specific, such as corpora comprising texts from a given academic discipline. (Hunston, 2008). Since corpora are stored on computers, this allows for very large amounts of text to be collected and analysed through the use of special software. Where language corpora are concerned, they can consist of both spoken and written texts, as well as multimedia elements such as video clips. One of the ways in which corpora can be used is for research purposes, to study how a lexical pattern or item is used.(O’Keeffe et al, 2007:3). According to Granger, there are four major disciplines that will impact upon learner corpus research, and each of these disciplines is vital in the optimal exploitation of corpus data; these four disciplines are corpus linguistics, linguistic theory, second language acquisition and foreign language teaching. (Granger in Karin, 2009:15). She also offers the view that the pedagogical use of this data, whether for immediate or delayed use, needs to be assessed on the basis of achievement of proficiency in accuracy, complexity and fluency, depending upon whichever target is desired. O’Keffe et al (2007) have set out some of the basic techniques that can be used on a corpus with standard software. One of these is concording, which is essentially the use of the corpus software to find a particular word or phrase, with the search word being referred to as the node. (O’Keeffe et al, 2007:8). Reading a text involves an element of trust, as Sinclair has pointed out, in that the reader trusts the text to have been composed in such a way that material presented earlier in the text will be followed and/or answered in subsequent text.(Toolan,2007). The use of a corpus based approach to the study of language reveals regularities of language that are difficult to intuit or to observe in texts as they are observed in daily life. Through the observance of elements such as concordance lines or lists of collocates, it is possible to determine the recurrence of co-occuring items.(Hunston, 2008). Semantic sequences may be observed in a general corpus, wherein in each case, a lexical word is used as the core item and the sequence will comprise patterns associated with each word and the number of phrase types occurring before the word. The prioritising of small words in identifying semantic sequences is crucial to textual meaning. Small words such as prepositions play a role in grammar patterns and the non random association between such patterns and groups of words that occur with them; hence they form an integral part of semantic sequences. Collocation refers to the co-occurance of words within a particular piece of text and it plays a role in the manner in which the text is interpreted. In discussing this role, Barnbrook(2007) has highlighted Sinclair’s views laying out two models. The first of these is the open choice model which is the “normal way of seeing and describing language” (Sinclair 1991:109, cited in Barnbrook, 2007). This grammatical model of language separates the lexical and semantic aspects of words from the manner in which syntax is organized. On this basis, sentences are formed by placing together appropriate lexical items which exist in pre-determined subsets of vocabulary that is based on features of speech. Some example combinations are: verbs, which would be used in a sentence through a combination of the verb and the auxiliary or tense. Hence the basic units of the open choice model are individual words that can be combined with high levels of flexibility albeit within certain grammatical constraints. One of the statistical approaches that are used in measuring collocation of words is collexeme analysis, which quantifies the degree to which certain words are either attracted to or repelled by “syntactically defied slots in grammatical patterns or constructions.” (Gries, 2008:404). The idiom principle is not based on choice of words but rather on pre-fabricated phrases, which may demonstrate internal variations based on lexical or syntactic elements or on the other hand, a pattern of co-occurence within a semantic environment. In considering the idiom principle, Sinclair points out that “the open choice principle does not provide for substantial enough restraints on consecutive choices.” (Sinclair, 1991:110; cited in Barnbrook, 2007:185). It is inadequate to capture the meaning of language, while the idiom principle takes into account patterns of co-occurrence, or the collocation between words and phrases, taking into account the grammatical structures and semantic environments within which they occur. The frequency of co-occurrence of two or more linguistic variables is often the most frequently used statistic in corpus linguistics (Gries, 2008). For instance, some words and phrases are used more frequently than others and their meanings as set out in the dictionary, may not necessarily indicate the meanings that are supplied “by introspection” (Sinclair, 1991:112, cited in Barnbrook, 2007). Rather the meanings that are given the first place in the dictionary may be relatively uncommon in the large corpora, because the meaning of the more frequently used words may sometimes be explained only in terms of usage, thereby suggesting greater applicability of the idiom principle. The efficacy of the idiom principle also appears to be corroborated in another study carried out by Erman where the objective was to ascertain whether pause frequency and pause duration could provide information about the size of the linguistic units stored in the mental lexicon (Erman, 2007). Linguists now generally agree that the occurrence of linguistic units is of multi word size in the mental lexicon rather than single items. Erman’s study sought to examine whether the distribution and duration of pauses in linguistic units would support the hypothesis that multi word constructions are stored in long term memory and retrieved in their entirety, which would make the occurrence of pauses rare, or brief if they do occur. Speakers store strings of words or prefabs and also store the matching contexts and situations. The results from the study corroborated the hypothesis, suggesting that prefabs are stored in long term memory and retrieved in wholes (Erman, 2007). Another hypothesis in this study was that the ability for multi-word storage increased with age, so the study also sought to determine the differences that existed between adolescents and adults in terms of the distribution of pauses. The results on the study however were inconclusive, because adolescents paused much less than adults overall. In a study of usage of the word “really” in a corpus of academic lectures and academic book review articles, the patterns of usage were compared. The objective of the study was to compare whether and to what extent the frequencies, meaning and usage of the word “really” varied in register across spoken and written academic discourse (Diani, 2008:298). The data collected in the study revealed that in terms of frequency, usage was much higher in the spoken as compared to the written word. But among the three written corpora that were used, i.e, the LIBRA corpus, the HEM-Economics and HEM-History, the usage of “really” was much higher in the former which comprised review articles, while usage was much lower in the other two written corpora due to the perceived danger of hyperbole and exaggeration. On the basis of analysis of a parts-of-speech tagged corpora, Hudson(1994) found that the percentage of words in written English that are nouns appeared to be consistent at around 37%, after noting that “the super class the includes pronouns as well as common and proper nouns is clearly what most linguists would call simply ‘nouns’.(Hudson, 1994:332 cited in Hardie (2007). Hudson uses some of the parts-of-speech frequency data as the metric of variation in the broader analysis of multiple dimensions of variation in both spoken and written English. He suggests that it could also be used with other languages, because while this norm for writing may vary from language to language, it could form a useful basis of a uniform counting method that would give very similar results for speech and for writing. Hardie (2007) however has set out some of the major objections to analysis based on POS ratios for wide super-ordinate categories such as Hudson’s noun. The first objection is that superordinate categories such as noun, pronoun, etc cannot be derived from groups of POS tags in a straightforward and uncontroversial way. The second objection is that since the figures derived for these super ordinate categories will depend on the POS tagging scheme used, absolute percentage figures are an artefact of the tagging scheme. Thirdly, it is difficult and problematic to compare two datasets that have been tagged with different tagging schemes and more so, in making cross linguistic comparisons (Hardie, 2007:72-73). Semantic sequences occur in corpora, with the lexical or grammatical words forming the core, with an association between grammar pattern and meaning. (Hunston, 2008). According to Sinclair (1991, 2004, cited in Hunston, 2008: 291), individual words occur in identifiable longer sequences and when they are described in terms of collocation and colligation, they form a unit of meaning. Moreover, the unit of meaning has an emotive or evaluative implication which may be recovered from an observation of the context in which the word/s occur (Hunston, 2008). O’Keeffe et al (2007) provide some examples of how collocations ascribe meaning; for instance in terms of the verbs turn and go, go is associated with a more negative context. This means that people can go insane, blind or bald but do not turn insane, blind or bald. Similarly, some textual features operate on the principle of repulsion, where meaning is constructed in a manner opposite to collocation; for instance cheerfully happy is not used while blissfully happy is appropriate(Renouf and Banerjee, 2007). Such information is useful to a learner trying to understand the subtleties of the English language and to teachers when they try to explain why certain words go together while others do not. A study by Liu(2008) examined the frequency and usage patterns of English linking adverbials across five registers (spoken English, academic writing, fiction, news writing and other writings). It found that while the overall use of linking adverbials in fiction is lower than speaking and academic writing, its use of the sequential linking adverbials is actually much higher. This suggests that in teaching journalism English, it may be advisable to focus on sequential and summative linking adverbials, such as in the meantime, in short and meanwhile. This study also found a generally low use of summative adverbials which suggests that the present focus on these adverbials in teaching material and in the classroom may be questionable. In another study which reported on the corpus based investigation of English similes, which conform to the basic pattern of (as) adjective as noun group, with the noun group being followed by prepositional phrases (Moon, 2008). Since the basic structure of the as­-similes is criteria, there is only a limited potential for syntagmatic variation. The meaning of a simile may generally correspond to the meaning of the component adjective, although this may not necessarily be the commonest or core meaning; but some similes may also realize more than one meaning. A reliance on corpus data in the investigation of idioms, proverbs, similes, etc have been criticized on the basis that they may not provide an accurate representation of language use because they may contain the wrong kind of data.(Moon, 2008). Moreover, even large general corpora may not provide adequate amounts of information about the distinctions between regional varieties of languages. While the web may provide an alternative source of massive amounts of data, it may not necessarily be representative of actual patterns of usage. Qi and Ge (2007) carried out a lexical study on word frequency and text coverage of 570 word families from Coxhead’s AWL (Academic word list) in medical research articles. Their findings suggested that (a) academic vocabulary which has high text coverage and dispersion throughout medical research articles, is an important set of word items (b) the AWL does not adequately represent all the academic words frequently used in medical articles and (c) the different coverage of academic words suggests that academic words serve some rhetorical functions in academic texts. Rayson(2008) has taken up a case study approach as a means for the comparison of corpora, using automatic tagging software, which assigns parts of speech and domain tags. The corpora selected were the UK 2001 general Election Manifestos of the Labour and Liberal Democratic parties. A macroscopic analysis carried out on both these texts using a web based tool, Wmatrix to implement the proposed method and to show the relative merits of key words and key semantic domains. This study presented the complete method for extending the keyness technique from key words to key semantic domains, which represents a way to combine elements from both corpus based and corpus driven paradigms within corpus linguistics. Use of Corpora for academic purposes: Corpora are now used routinely in producing dictionaries and other reference materials, as well as language teaching materials, while also serving as a mode for investigating the ideological and cultural assumptions that underlie language.(Teubert and Krishnamurthy, 2007). Cooper (2008) found that a bilingual dictionary is a valuable linguistic resource because it records the differences in the segmentation of semantic space between two languages and therefore the difficulties in producing faithful translations. But they are still a valuable resource. Attempts are even being made to make the process of translation of language the subject of artificial intelligence, so that some translations can be carried out automatically using the Internet. But as Beebe et al (2009) point out, if this is to be effective, then a balance is required between training and education, because translation not only requires the academic knowledge and familiarity with a language; it also requires effective training that emulates the workstations of professional translators. (Beebe et al, 2009:2) A study was carried out to examine how the integration of corpus and contextualized lexico-grammar into the teaching of foreign and second language affected the outcomes (Liu and Jiang, 2009). The study involved 244 participant students learning English as a foreign language and as a second language, at one Chinese university and two U.S. universities. A wide spectrum of information was collected in the study, including teacher lesson plans and journals as well as student reflection papers and corpus search projects. On the basis of the data collected in the study, it appeared that the integration of corpus had a beneficial effect, resulting in improvements in student command and understanding of lexico-grammar and enhanced skills of discovery learning (Liu and Jiang, 2009). The researchers have concluded that pedagogical implications are that teachers will have to carefully decide whether and to what extent they wish to incorporate corpus into teaching. If they do use it, they need to pay attention to providing training and modelling for students to reduce their difficulties. Effective writing requires the use of complex grammatical structures, but this is an aspect that is poorly defined in linguistics and literacy. In a study carried out with students from years 10 and 11 in a UK school, the findings suggested that this complexity can be addressed by viewing grammar as a series of options that can be manipulated to maximize the potential for meaning in text.(Rimmer, 2008). The use of corpora for the description of language is beneficial; but one of the limitations identified by Gilquin et al (2007) in the use of corpora in teaching English for academic purposes is that researchers mainly use native corpora, but have largely neglected the use of learner corpora, which could provide valuable potential for insights in this area. Learner corpora comprise electronic collections of spoken or written data from foreign language learners, which is a valuable source of data, especially when they are error tagged (Kraif,et al, 2007). Behrens (2009) examined various sampling methods used in child language acquisition corpora, such as (a) diary studies (b) large sample studies and (c) longitudinal studies and suggests that a combination of research methods is likely to be most effective provide adequate provision is made for triangulation of data (Behrens, 2009:xvii). Child language acquisition corpora can provide valuable insight into the processes whereby language skills are acquired. Learner corpora comprise continuous discourse and not a set of decontextualized sentences, containing productions in their entirety, both the correct and incorrect uses. Discourse studies of language are generally qualitative and corpus linguistics finds wide applicability because it can compare the linguistic characteristics of discourse from different spoken and written registers.(Biber et al, 2007:10). Historical English language corpora may also be useful because they are publicly available and when taken together, can be used to investigate language variation over time (Nevalainen et al, 2008:3). One of the advantages of learner corpora lies in the large database of learner data that is available for empirical study, on the basis of which valuable insights may be gleaned. One of the problems however that is inherent in the extraction of data from databases is access limitations, but Gabrielatos(2007) has proposed an accessible measure to extract a wider range of data by refining a query to also include relevant terms that draws upon techniques used in information retrieval combined with the creation of a corpora from the world wide web. Yet another advantage of learner corpora lies in the fact that such data can be subjected to analysis using a wide variety of analytical tools that allow a better analysis of the underlying lexicogrammatical signatures within such corpora.(Gilquin et al, 2007: 322). The errors inherent within the data can form the basis for a more comprehensive understanding of patterns of lexicogrammatical use, through the use of error tagging systems. One example of an error tagging system is that devised by Nicholls (2003; cited in Granger, 2007:256) to annotate the Cambridge Learners Corpus, containing data from a wide range of learner populations. This is a three tier system specifying the error domain (the form, grammar, lexis, etc), the category of the error (for instance, whether tense, gender or number) and the word category (adjective, noun, verb, etc), which offers tremendous potential in teaching of English as a foreign language.(Meurnier, 2007). One of the findings that has been emerging from learner based corpora is that some of the linguistic features that are characteristic of learner language are common to learners with a background of different mother tongues, while other features appear to be characteristic only among a particular linguistic groups. This suggests that the former may be developmental while the latter may be peculiar to a particular linguistic group (Gilquin et al, 2007: 322). Learner corpora have formed the basis for learner corpus informed CALL (Computer aided language learning) programs, which are remedial exercises that specifically target the identified difficulties. (Granger, 2007). Such computer based programs also offer higher levels of flexibility. If learner corpus research is to contribute to the value of such programs however, there may be a need to address some significant methodological issues, such as using a larger corpus, a clear classification of texts and more systematically sampled sub-corpora.(Pendar and Chapelle, 2008) In investigating the use of English for academic purposes, corpus-linguistic methods differ from other research based paradigms in that they focus upon the “co-text of selected lexical items in academic texts.” (Gilquin et al, 2007:320). Corpora can be useful in several ways; computer corpora can be analyzed both qualitatively and quantitatively using software tools such as WordSmith Tools 4. The word lists linguistic corpora provide information on the frequency and distribution of both single words and word sequences used in different corpora, which allows for an automatic comparison of which part of the vocabulary is salient within a given corpus. Concordances may also be helpful in the analysis of text, to determine the preferred co-occurences and grammatical structures. Examining the lexical cohesion in the text can contribute significantly to deriving its attendant meaning, because it examines the way lexical items relate to each other to create continuity in the text.(Morley in Flowerdew et al, 2009: 5-22) The UVaSTE corpus is an open access corpus comprising 3,000,000 words of scientific and technical English, which is accessible to the research community through the Internet. This corpus is enriched through the availability of additional information about the text type and purpose, subject matter, year of production, genre, author and geographical variety of the text as well as the origin of the documents.(Fuertas, Olivera, 2008:73). The documents date back up to 25 years and is subdivided into two sub corpora – informational and learned research texts comprising 75% and the balance comprising informational reporting and instructional administrative documents. Although the corpora is currently localized in use, it has potential for adding to the amount of knowledge that can be added to professional activity through the use of such specialized corpora. These corpora can also be used for concordances in the classroom and provide a better understanding of the grammar implications in teaching English as a second language. Another important trend that has developed in recent times is the use of non sexist English, where “person” may be used frequently to replace “man” which would have been used earlier. It is therefore necessary to teach second language students the niceties of lexical gender because this also has social and ideological implications.(Fuertas, Olivera, 2008) Gender sensitivity must be made “an overt and deliberate part of ESP teaching”. (Fuertas, Olivera, 2008:78). Elements such as these may make a difference in the ability to acquire advance language skills, because the ability to master tough movement constructions could help students avoid the less explicit, semantically opaque, loose-fit variants in their use of language.(Callies, 2008). Researchers have pointed out that corpora can be used effectively in teaching English for academic purposes because “the value of corpus work lies in the fact that it can replace both instruction with discovery and refocus attention on accuracy as an appropriate aspect of learning.” (Krishnamurthy and Kosen, 2007). While some researchers are in favour of using corpora for research purposes, their use in the classroom however, is considered more problematic. For example, it requires a high level of motivation and curiosity about language from students, which is generally missing because students may prefer teachers to analyze corpus data and explain it in summary (Krishnamurthy and Kosen, 2007: 358). If language teaching is to take place effectively, learners must use corpora effectively and this also involves the selection of the appropriate corpus. But another problem in developing corpora is effective classification of academic subjects, which has always created problems because of the lack of consensus among academic institutions and librarians with the absence of a standard classification system. A study by Liu and Jiang(2009) also highlighted how some challenges in effective integration of corpus, such as the limited use of search engine would require advancements in technology. In designing an effective EAP corpus, one of the problems that must therefore be addressed is the lack of classification, which could be addressed by localising the corpus to a particular EAP application. Krishnamurthy and Kosen (2007) have suggested that universities in the UK could adopt a joint classification system using a bottom-up approach, with modules as a bottom level category with programmes, departments and broad academic groupings as a broad level category. At the lower levels, individual modules would form their own subcorpora that can be used for ESAP purposes. EAP corpora must also contain the appropriate references and quotations from academic texts in order to be valid and viable for use. A comprehensive EAP corpus should also contain texts that are produced by students at all levels. Data from one of the largest corpora, i.e, spoken British English was used as the basis to explore lexical grammatical aspects of progressive forms with future time reference, such as verb preferences, co-selection of adverbs, subject types and negation (Neselhauf, 2007). This analysis revealed that future time progressives in British English are patterned to a large extent and the actual language use runs counter to the claims that are found in traditional grammatical descriptions. For example, the analysis showed that the claim that only motion verbs generally occur without an adverbial nearby could not be confirmed. Moreover, the study also yielded some valuable insights for instance, that in accounts of lexical grammatical patterns, the notion of frequency that is being used must be clarified. It is not sufficient to merely identify a pattern or tendency, but also to examine whether the pattern is a general one or specific to the construction under investigation.(Neselhauf, 2007). For a valid analysis of such patterns, the terms used in the definition of such patterns must also be precisely defined. Moreover, evidence suggests that there is a marked preference for syntactic structures where dependencies are short (Temperley, 2008), which would also aid in retention in student memory. Yamasaki (2008) points out how certain collocations and colligations are associated with the discourse functions of unspecific anaphoric nouns like problem, reason, idea which are referred to as labels. In spoken English, the use of such nouns occurs in copular construction while speaking, when adjectives are included as complements. As a result, there is a difference in collocational and colligational patterns between speech and writing and an awareness of such differences would be helpful to students learning English. Gilquin(2007) has also highlighted certain unidiomatic collocations that tend to recur in learning writing. Oral corpora form a useful method for teaching and learning languages, because native learning can be utilized in the acquisition of second language, while spoken language analysis can be used as an aid in reflective practice where training for language education teachers is concerned.(DeCock, 2007). The concept of transfer from the native language plays a major role in the acquisition of English language skills. Transfer may be defined as “the forms and meanings and the distribution of forms and meanings of [individuals’] native language and culture to the foreign language and culture.(Gilquin, 2008). Contrastive analysis of the two languages is the very basis of language transfer studies because structural congruity or partial similarity between the languages may be necessary for language skills transfer to occur, although this may not necessarily be adequate in itself.(Cosme, 2008). Effective academic teaching of English as a second language or a foreign language will therefore require an understanding of the role played by transfer in acquisition of language skills. Meuniur and Granger(2008) have pointed out the importance of phraseology in second language acquisition, firstly because native and learner corpora plays a vital role in the extraction and description of multi-word units; secondly the directionality of collocation must determine which words must be taught with their respective collocates and lastly, the role of memorization in learning phraseology cannot be underestimated. (Meuniur and Granger, 2008:15-17). Hence, teachers must accord due importance to phraseology and the manner in which it is taught in order to garner effective results and second language learning outcomes. Conclusions: In conclusion, it may be noted that the focus of corpora linguistics is on deriving meaning from texts. The meaning lies in the discourse which is self reflective, because it talks about language use. Examining the lexical cohesion in the text may be useful in deriving its meaning. In teaching foreign languages, the use of learner corpora can be a valuable source of information about common mistakes. It is also a valuable tool for teachers to understand how transfer takes place within the context of the student’s native language, as well as syntactic structures in English, so that the learner outcomes are improved. Bibliography: Barnbrook, Geoff, 2007. “Sinclair on collocation”, International Journal of Corpus Linguistics, 12(2):183-199 Beeby, Allison, Patricia Rodríguez Inés and Pilar Sánchez-Gijón (eds.) 2009. “Corpus Use and Translating: Corpus use for learning to translate and learning corpus use to translate”, Benjamins Translation Library 82 Behrens, Heike (ed.), 2008. “Corpora in Language Acquisition Research: History, methods, perspectives”, Trends in Language Acquisition Research, 6. Biber, Douglas, Ulla Connor and Thomas A. Upton, 2007. “Discourse on the Move: Using corpus analysis to describe discourse structure”, Studies in Corpus Linguistics, 28. Callies M. (2008) “Argument Realization and Information Packaging in Tough-Movement Constructions - A Learner-Corpus-Based Investigation”, IN Gabrys-Barker, D. (ed.). “Morphosyntactic Issues in Second Language Acquisition Studies”, Clevedon: Multilingual Matters Cooper Martin C.,2008. “Measuring the Semantic Distance between Languages from a Statistical Analysis of Bilingual Dictionaries”, 1-33, Journal of Quantitative Linguistics, Volume 15 Cosme C, 2008. “Participle clauses in learner English: the role of transfer”, IN Gilquin G., Papp S. & Díez-Bedmar M.B. (eds), “Linking up Contrastive and Learner Corpus Research “, Amsterdam, Atlanta : Rodopi, pp. 177-198. Diani, Guilani, 2008. “Emphasizers in spoken and written academic discourse”, International Journal of Corpus Linguistics, 13(3): 296-321 De Cock S.,2007. “Routinized Building Blocks in Native Speaker and Learner Speech: Clausal Sequences in the Spotlight”, IN Campoy, M. C. and M. J. Luzón (eds.) Spoken Corpora in Applied Linguistics . Bern: Peter Lang, 217-233. Erman, Britt, 2007. “Cognitive processes as evidence of the idiom principle”, Internaitonal Journal of Corpus Linguistics, 12(1): 25-53. Flowerdew, John and Michaela Mahlberg (eds.), 2009. “Lexical Cohesion and Corpus Linguistics”, Benjamins Current Topics, 17. Fuertas-Olivera, 2008. “Pedagogical application of specialized corpora in ESP teaching: the case of the UVaSTECorpus”, Scripta Manent, 3(2): 68-81 Gabrielatos C.,2007. “Selecting query terms to build a specialised corpus from a restricted-access database”, ICAME JOURNAL Computers in English Linguistics, 31. Gilquin G, 2008. “Combining contrastive and interlanguage analysis to apprehend transfer: detection, explanation, evaluation”, IN Gilquin G., Papp S. & Díez-Bedmar M.B. (eds), “Linking up Contrastive and Learner Corpus Research “, Amsterdam , Atlanta : Rodopi, pp. 3-33 Gilquin G. (2007) To err is not all. What corpus and elicitation can reveal about the use of collocations by learners. Zeitschrift für Anglistik und Amerikanistik 55(3): 273-291. Gilquin, Gaetanelle, Granger, Sylviane and Paquot, Magali, 2007. “Learner corpora: the missing link in EAP pedagogy”, Journal of English for Academic Purposes, 6:319-335. Granger, Sylvanie, 2007. “Integrating learner corpora and natural language processing: A crucial step towards reconciling techological sophistication and pedagogical effectiveness”, ReCall, 19(3): 252-268. Gries, Stegan Th, 2008. “Dispersions and adjusted frequencies in corpora”, International Journal of Corpus Linguistics, 13(4): 403-437 Hardie, Andrew, 2007. “Part-of-speech ratios in English corpora”, International Journal of Corpus Linguistics, 12(1): 55-81. Hunston, Susan, 2008. “Starting with the small words”, International Journal of Corpus Linguistics, 13(3): 271-295. Karin, Aijmer 2009. Corpora and Language Teaching, University of Gothenburg, Studies in Corpus Linguistics 33, viii, 232 pp. John Benjamins Publishing Company Krishnamurthy, Ramesh and Kosen, Iztok, 2007. “Issues in creating a corpus for EAP pedagogy and research”, Journal of English for Academic Purposes, 6: 356-373. Liu, Dilin and Jiang, Ping, 2009. “Using a corpus based lexicogrammatical approach to grammar instruction in EFL and ESL contexts”, The Modern Language Journal, 93(i) 61-79 Liu, Dilin, 2008. “Linking adverbials”, International Journal of Corpus Linguistics, 13(4): 491-518 Meunier F. and S. Granger (eds), 2008. “Phraseology in Foreign Language Learning and Teaching”, Amsterdam and Philadelphia: Benjamins Meunier F, 2007. “Reprint of 'The pedagogical value of native and learner corpora in EFL grammar teaching'”(2002). IN Teubert W. and R. Krishnamurthy (eds) Corpus Linguistics: Critical Concepts in Linguistics. Volume 2. London and New York:, Routledge. 119-141 Moon, Rosamund, 2008. “Conventionalized as- similes in English”, International Journal of Corpus Linguistics, 13(1): 3-37 Nesselhauf, Nadja and Romer, Ute, 2007. “Lexical grammatical patterns in spoken English”, International Journal of Corpus Linguistics, 12(3): 297-333. Nevalainen, Terttu, Irma Taavitsainen, Päivi Pahta and Minna Korhonen (eds.) The Dynamics of Linguistic Variation: Corpus evidence on English past and present O’Keefe, A. McCarthy, M and Carter, R, 2007. “From Corpus to classroom: Language use and language teaching”, Cambridge: Cambridge University Press. Pendar N. and C. A. Chapelle, 2008. “Investigating the promise of learner corpora: methodological issues”, CALICO Journal, 25(2):189-206 Qi, Chen and Guang-chun, Ge, 2007. “A corpus based lexical study on frequency and distribution of Coxhead’s AWL word families in medical research articles (RAs)”, English for specific purposes, 26: 502-514. Rayson, Paul,2008. “From key words to semantic domains”, International Journal of Corpus linguistics, 13(4): 519-549 Renouf, Antoinette and Banerjee, Jayeeta, 2007. “Lexical repulsion between sense related pairs”, International Journal of Corpus Linguistics, 12(3): 415-443 Rimmer, Wayne, 2008. “Putting grammatical complexity in context”, Literacy, 42(1):29-36 Temperley, David, 2008. “Dependency length minimization in natural and artificial languages”, Journal of Quantitative Linguistics, 15(3):256-282 Teubert W. and R. Krishnamurthy (eds) Corpus Linguistics: Critical Concepts in Linguistics. Volume 2. London & New York: Routledge. Toolan, Michael, 2007. “Trust and text, text as trust”, International Journal of Corpus Linguistics, 12(2): 269-288 Read More