In the third part of this debate, Scott Thornbury offers tips on how he believes L2 reading should be taught.

What are the implications of this view? To me they are the following:

  • We should be teaching the language of second language reading, not the skill of it. The longer we waste time teaching reading skills at the expense of language, the longer we postpone the transfer of skills that depend on there being the critical mass of language that is necessary to facilitate such transfer;
  • At lower levels, therefore, we should use texts, not to teach reading, but for the teaching of grammar and vocabulary, and as models for writing, and as springboards for discussion;
  • Such texts should not be used superficially (eg. for skimming and scanning purposes) but should be subject to intensive reading and close analysis, so that their potential as linguistic data is maximised;
  • At higher levels we should choose texts which encourage the transfer of L1 reading skills, and we should also promote out-of-class, extensive reading, as a source, primarily, of incidental vocabulary acquisition;
  • We should explicitly target reading skills only for those (higher-level) learners who are poor readers in their L1. (With the corollary that it might be better to teach them these skills in their L1, if possible).

Further implications of the above might be:

  • Coursebook writers should label texts not as READING but as TEXT (except at higher levels where the transfer of specific reading strategies is being targeted, but see below).
  • Coursebook texts do not need to be as long as they are in order to be exploitable for language teaching purposes. A 50-word text is likely to have enough vocabulary and grammar packed into it to keep your average low-intermediate class busy for a couple of hours at least. And for the purposes of discussion, or as models for writing, short texts do just as well, if not better, than long ones. Most coursebook reading texts are in fact simply time-fillers, and if used solely in order to “practise” skimming and scanning, are about as pedagogically useful as watching an episode of The Office with the sound off.
  • Getting rid of READING texts from the beginner to intermediate syllabus would effectively free up hundreds of hours of classroom time, as well as substantially reducing the length (and cost) of coursebooks.
  • At higher levels, the kinds of texts that prompt learners to transfer their L1 reading skills, that they want to read in the same way as they read in their L1, are those that are intrinsically motivating, i.e. that engage the learners’ needs and interests. Since few coursebook reading texts succeed in doing this, it might be better to eliminate reading texts from higher level books as well, and substitute them with tasks such as:
    “Find a text on the Internet about a famous person you admire/a place you would like to visit/a sport or game you used to play....etc. Print it off, bring it to class, and be prepared to share it with your classmates.”
  • In class, the task might be – in pairs – to read one another’s texts and ask each other questions about them in order (a) to clarify any language difficulties, and (b) to explain and justify the choice of text. Then they report on their partner’s text to the whole class. Again, by eliminating reading texts from high-level coursebooks, classroom boredom time is reduced, coursebook length is slashed, and whole tracts of Amazon rain forest are spared. Is this the end of reading as we know it?

© Scott Thornbury

What do you think? Have your say on the Forum.