In the preceding chapter we looked at ways to describe errors. In this chapter we focus on the classification of those errors. We do so by reference to three criteria: modality, medium and level. Modality refers to whether the learners' behaviour was receptive or productive. Notice that we prefer 'receptive' as a label to 'passive' since language processing is never passive: on the contrary, 'passive' processing calls for considerable mental activity and interpretive involvement on the receiver's part. Medium indicates whether the language produced or received was spoken or written. Was the learner operating, when the error was made, with speech sounds or with written symbols? Taking modality and medium together, we are able to specify which of the 'four skills' the learner was operating at the time of the error: speaking, writing, listening or reading. In addition to this we want to specify on what level of language the learner was operating at the time he or she erred. We recognize just three levels of language: the levels of substance, text and discourse. If the learner was operating the phonological or the graphological substance systems, that is spelling or pronouncing (or their receptive equivalents), we say he or she has produced an encoding or decoding error. If he or she was operating the lexico-grammatical systems of the TL to produce or process text, we refer to any errors on this level as composing or understanding errors. If he or she was operating on the discourse level, we label the errors occurring misformulation or misprocessing

130 Errors in Language

errors. These six types are then differentiated further according to the medium of each. The result is the classificatory system of errors with 1WELVE categories shown in Figure 5.1.