Functionality and Tools

Tools are used to find out something about a language. Therefore the relevance of properties of a language must also be taught. This coincides with new aims in language pedagogy. In these concepts learners are to discover the language by themselves. As mentioned before the learner follows an explorative approach to language learning. Programs used in this context are e.g. concordancers, which can help to find examples of language usage in "authentic" texts. This does not only support the learning of lexical aspects but also of grammatical and even of semantical aspects.

Note that the keyword "tool" does not refer to applications like a webbrowser or a wordprocessor, which can also be used in language learning. In order to improve "simple" applications such as concordancers the addition of CL-technology may be helpful. One example is the project TATOE, a concordancer able to use the output of the morphological analyser Morphy. A morphological analyser can not only provide the input for further analysis tools but can be used directly in order to identify the morphological structure of words as it is done e.g. in Glosser. Foreign electronic texts can then be read and understood more easily. The tagger might be able to classify the results or at least provide some more information about the structure of the found items. There are a number of tools for the analysis of spoken language possibly helping the language learner to improve his pronunciation. However I am not aware that this has been done in larger experiments. This might be a case where the learner needs an expert to explain to him the output of frequency-analysers and the like.

However one general problem is that learners generally do not need to learn abstract linguistic knowledge about a language in order to perform communicative actions. But as CL-tools usually deliver this kind of knowledge there are only very little areas where this makes sense.

Finally there is the area of style checking via assessment techniques such as "Latent Sematic Analysis" (LSA). In this case the tool may be used not for analysing language examples for the learner but to analyse learner language itself. There have been some experiments where texts written by learners are compared to a prototype text written by a teacher and thus evaluated and rated, see e.g. LSA at CU (Summary Street project). The site also provides a number of introductory texts on LSA.