Crystal and Davy (1975) have published transcripts of unedited audiorecorded
conversations, representing standard, educated colloquial English
usage. They admit, as might be predicted from the arguments which Brown
puts forward, that they are "unclear as to how data of this kind can best be
used in a teaching situation" (p. x). A paradox is whether informal language
can be formally taught or tested. Such conversational English is important as
it is different from the language presented to learners in most textbooks.
Usually students are exposed to formal varieties, although informal conversation
must provide some kind of baseline for a description of English, if
only because of its massively common occurrence. Crystal and Davy do not
suggest teaching the productive use of such a variety of English; they
propose a policy of exposure to increase the receptive skills of discrimination
and comprehension. Similarly, Brown (1977, p. 156) does not approve of
teaching foreign learners to produce assimilated and elided phonological
forms, but only developing students' listening comprehension of such forms.