The Advantages of Using Corpora in Language Teaching
Introduction
Corpus linguistics is a relatively new discipline which originates from the second half of the twentieth century when the first machine-readable corpora were compiled. “Corpus” (plural, “corpora”) refers to a body of texts (usually electronic) selected for purposes of computer analysis. These texts can be drawn from a specific area of language use, or they may be drawn from many, depending on the purpose for which a particular corpus is compiled. With the evolution of technology, the existence of computers enabled linguists to not only store huge amounts of texts but also to analyze these natural language samples in a way which had not been possible before. One of the best known uses of corpora in language is concordancing, which was first developed by Tim Johns in the early 1980s, which led to the development of ‘data-driven learning’ (DLL). The “corpus revolution” has moved from language description to language teaching and learning. Nowadays, many people use corpora in language teaching because of its advantages.
The advantages of using corpora in language teaching
There are many advantages of using corpora in teaching languages. According to Federica Barbieri (cited in “Corpus What?”), corpus linguistics has much to offer to language teachers. Actually, both teachers and students can benefit from corpora. Corpora are generally used as a source of information about a particular language in language teaching.
Corpora enable teachers to teach authentic language or the way a certain language functions in real life to students. For example, teachers can explain to students that people in real life usually use the word, “gonna” instead of “going to” due to certain contexts and reasons. The authentic uses of language can be found out easily by using electronic corpora. These corpora can be quickly analyzed using a concordancer which is a software that searches for patterns in language such as the use of a specific word or phrase. Sometimes, parsers (a type of software) can be used to analyze the grammatical patterns in language, but these tend not to be commonly used in the classroom, except perhaps in linguistics classes. If desired, teachers can ask their students to use a concordancer to carry out students’ own searches of the corpus, and develop their own idea about the uses of language. In other words, for example, when students are uncertain about the use of a lexical item, they can use concordancer to produce a concordance of the item in question. Besides, language rules which are based on observation of actual use can be taught to learners. This is what is called “data-driven learning” or DDL by Tim Johns, though he mainly had in mind that students themselves would develop ideas about language use by examining real language data (cited in “Corpus What”). For example, students may be uncertain about when to use the terms ‘results’ and ‘findings’ in writing research papers. The use of a corpus would enable the students to see when and where and how these two terms are used.
Another advantage of corpora in language learning and teaching is that intuitions about language and language use can be tested across a range of corpora by teachers and students. In other words, the use of language in a variety of registers and genres can be compared by using either different corpora, or by selecting within the corpus. The findings from one corpus can therefore be verified or disproved by comparing with another corpus. For example, intuitions about the use of language in spoken as against written language allow the verifiability of the intuitions. Svartvik (1992) and Biber (1995), stress the importance of this factor in which they claim that for results to have any meaning, the results must be able to be verified (cited in “The Methodological Background…”). Thus, corpora offer the possibility of verifying results from the samples of the real language usage which this is helpful to teachers in language teaching.
Many areas of language can be explored by using corpora such as historical language change, vocabulary, syntax, text, spoken language translation issues, register, variation, lexicography etc. But in the classroom most teachers are interested in matters of direct and immediate concern such as the use of lexical items and patterns in grammar, particularly in relation to register. For this reason, in the classroom most teachers find that the use of concordancing software is relevant. As mentioned above, concordancing software is software that can quickly search through a corpus and find every instance of the term being searched for, which could be a particular lexical item. The results of the search are displayed in a context of a user-determinable number of words to the right or left of the search item. Thus, the student or teacher can see how the word is used in context. The searches for common (high frequency) words tend to result in the production of many hundreds of contexts, which can be analyzed by teachers widely and thoroughly (this is suitable to be done with intermediate or advanced language learners). It is therefore sometimes best for teachers to search for words which are creating a problem because they are of low frequency and that their analysis will be narrow. An example of this kind of approach is Tim Johns’ ‘kibitzers’, which are directed at advanced ESL learners, and which can be found on his home page.
Corpora can be a source of case studies of specific language features that perhaps are the teachers’ areas of interests as well as major or specialization. Because corpora contain (or should contain) a large number of words from different types of registers of language use, they enable students and teachers to discover differences in language use according to register, and, for example, between formal and informal uses of language. Corpora are also a resource of language research. With corpora, teachers will be observers of language use as well as researchers of a language. For example, corpora help teachers investigate metaphor clusters in certain texts like economic texts, business texts and so on. Time in studying and researching language students’ needs can be saved with the help of corpora.
Data in the corpora can be used for creating teaching materials. According to Tan (2003), corpora benefit both teachers and learners because it involves material design which becomes a source to teachers to provide linguistic information and also a resource for CALL (Computer Assisted Language Learning). Corpora which provide new insights into the way language operates help language teachers in designing and preparing lesson plans and with materials for their classes. For example, teachers can make use of simple concordancing packages like WordSmith or MonoConc. They can select the language items from these packages that they find they need to teach their students based on the learners’ needs. These packages also offer information about frequency of use of linguistics features which later can help teachers make decisions about priorities in terms of teaching materials.
The benefits that can be gained from the use of corpora depend very much on what the corpus being used contains. This allows very great flexibility in the use of corpora, because they can be designed to with the task at hand in mind. If the teacher and students are concerned with the use of language in IT, then it is a relatively simple matter to produce a corpus containing language from IT-related materials, such as magazines. Similarly, many corpora contain large amounts of written data, as compared to spoken data. A particular corpus may be used to examine the differences between how language is used in speaking and in writing. A teacher wishing to use a corpus has, therefore, to be very carefully selective in choosing the corpus. On the other hand, with the easy availability of materials on the Internet, it is possible for teachers to compile their own corpora. Teachers of ESP (Teaching English for Specific Purposes) may wish to download articles from relevant journals, and compile a corpus. Students’ assignments, for example, can be scanned into a corpus, and subsequently analyzed. The problem is not so much the compiling of the corpus, but the subsequent analyzing of it that takes time and can be onerous. Clearly, though, a teacher or learner is unlikely to be interested in determining every feature of a text – it is far more likely that the concern will be with an immediate problem, and for this kind of activity corpora provide a swift and accurate answer.
It is common that second language acquisition is not without learner errors. Thus, corpora that contain samples of the written work of non-native speakers (second language learners) can help language teachers to know exactly how to prevent their students from making the same errors as well as what kinds of mistakes their learners tend to make. For example, the interference of certain elements of native language use in second language acquisition such as the inclusion of the word, “lah” in Malaysian English or the tendency of Malays to say, “give benefit (to something)” instead of the correct use of English, “to benefit (something).” The data about learner errors can assist teachers to gain an impression of the needs of their students in language learning. Besides, corpora help teachers categorize and identify learner errors and their tendency to make mistakes in learning a language. This means teachers are empowered by using corpora. Thus, corpora help teachers discover how they can help their students in language acquisition and this can make language teaching more effective.
Corpora can help second language teachers (especially non-native speakers) to understand the language they are teaching better before they can utilize and apply their knowledge to students. This can be related to a kind of corpora which are native-speaker corpora. Native-speaker corpora assist non-native teachers to teach a variety of language like the one used by the native speakers. This means, this type of corpora give a more accurate description of a language. Thus, teaching a language will be more effective when teachers refer to native-speaker corpora.
Corpora provide additional details about the frequency of particular features in specific contexts. Also, corpora offer a great number of instances of particular features of a language. One advantage of using corpora is that the frequency of patterns can be expected to reflect real language use. For example, many Malaysian speakers say to ‘discuss about’ by analogy with ‘to talk about’. Thus, with corpora, teachers can always check the use of certain words etc and not simply think they are correct just because they are widely used. The use of a corpus would quickly reveal that actually the patterns in English are ‘to discuss’ but ‘to have a discussion about.’
Samples in corpora can help teachers to devise tasks that include a wide range of features. For instance, students can be assigned to find verbs, phrasal verbs or expressions containing verbs that combine with the noun, “islamization.’ This obviously can result in the learner-centered approach of teaching language. Also, a variety of information in corpus samples can provide materials for homework tasks.
Teaching a language will be more systematic by using corpora as a source of reference, meaning teachers will start with certain areas of a language appropriate to students’ level of proficiency and will continue with other aspects in the next levels. Besides, as language evolves, with corpora, teachers can teach new things about a language. For example, teachers can discover evidence of principles of grammar in a particular context or register which cannot be found in older grammar books. For example, there are major differences in what written-language based grammar books say about language use, and the actual spoken language data. Language teaching then can be updated. Thus, students can learn a great deal from corpora – if they are properly guided by their teacher.
Conclusion
In conclusion, teaching a language will be easier for a teacher with the existence of corpora. It is easier in the sense that the teacher can confirm his or her intuitions about language use, and provides support for the development of materials. From the students’ point of view, it can provide immediate evidence of language use, and language patterns. However, the benefits to be gained are entirely dependent on the quality and relevance of the corpus being used. Teachers should therefore not rush into the use of corpora without ascertaining the value of the corpus being used. If teachers wish to use corpora they will find that their use can be very fruitful, but they have to be very certain of why they are using a corpus, what they are using it for, what they intend to find, and how the findings will be used in their teaching to the benefit of their students. Only if these questions can be satisfactorily answered, should the teacher use corpora in the classroom. Besides, the use of corpora is more suitable for intermediate or advanced language learners and not the beginners.
References:
Corpus What? Retrieved February 25, 2007 from http://www.tesol.org/s_tesol/secetdoc.asp?CID=963&DID=4244
Gabrielatos, Costas. (2005). Corpora and Language Teaching: Just a fling or wedding bells? Retrieved February 26, 2007 from http://tesl-ej.org/ej32/a1.html
The Methodological Background: British Traditions of Text Analysis, Correlative Register Analysis and Corpus Linguistics. Retrieved February 28, 2007 from http://users.utu.fi/micnel/thesis/Chapter5%20.html
Tan, Melinda. (2003). Language Corpora for language teachers. Retrieved February 26, 2007 from Journal of Language and Learning. Website:
http://www.shakespeare.uk.net/journal.uk.net/journal/jllearn/1_2/tan1.html
