Written by: Philip Kerr, e-mail:
А teacher trainer and materials writer, whose projects include the coursebook series Inside Out and Straightforward. He lives and works in Brussels.
In a presentation at the 2008 Exeter IATEFL conference, Catherine Walter and Michael Swan suggested that the use of a number of familiar reading tasks in the L2 classroom may be a waste of time. Similar claims have been made by Scott Thornbury among others. The tasks coming under attack include skimming, scanning and guessing the meaning of lexical items from their context, and form the basis of reading lessons in the majority of published coursebooks (e.g. Headway) and are central to the presentation of the teaching of reading skills in the majority of teacher training manuals (e.g. The TKT Course).
The arguments employed by critics of these tasks are powerful and draw on a large body of research (see Reading in a Second Language for an excellent review of this research). Their arguments include the following observations:

  • L2 reading comprehension ability correlates very strongly with lexical knowledge. Classroom time, especially at lower levels, is better spent helping learners with vocabulary acquisition than it is devoted to tasks such as skimming and scanning.
  • Comprehension difficulties in L2 arise, not from problems with skimming and scanning, but from gaps in the learners’ knowledge of or familiarity with linguistic features of the text. Skimming and scanning will not compensate for these gaps.
  • Good L1 readers will use particular strategies (e.g. skimming and scanning) when appropriate. They will transfer these strategic approaches to L2 texts when they have a language threshold which allows them to do so.
  • Good readers do not typically guess the meaning of unknown words from context, because they do not need to. Research does not suggest that practice in inferring meaning from context leads to gains in reading comprehension

Perhaps the most eloquent commentary on skimming and scanning is the complete omission of these terms from the index of Grabe’s Reading in a Second Language. It would seem that our well-established classroom routines for reading lessons are in need of re-evaluation. Until learners reach a level somewhere around B2 with a lexicon of a few thousand word families (enabling them to decode about 80% of ‘everyday’ texts), we should, it seems, be focussing on activities that foster word recognition and on programmes that encourage extensive reading.
However, before we consign skimming, scanning and inferring to the dustbin of discredited ELT techniques, it is a good idea to pause for thought. Firstly, because the recent history of ELT warns us against reaching hasty conclusions. Explicit grammar instruction, translation and dictation are but three of the many classroom activities that have been discarded, only to be brought back into the frame of critical acceptability. Secondly, as Grabe points out, the classroom implications of reading research need to be tested in particular contexts.
There are at least four strong reasons, in my view, why skimming, scanning and inferring may be justified in some ELT contexts. These have little, or possibly nothing, to do with the development of learners’ reading skills. However, I would tentatively suggest that we may be doing the right sorts of things for the wrong reasons.

  1. A central part of a teacher’s job must be to promote her learners’ motivation: the motivation to learn English, in general, and the motivation to read texts in English, in particular. Among the many theories of learner motivation, there is general agreement that motivation contributes to success, and that success can lead to enhanced motivation. Skimming and scanning tasks in most coursebooks are, on the whole, neither particularly time-consuming nor particularly difficult. For a low level student, the sense of achievement that can be derived from successful task completion is something that we cannot afford to ignore. Well-designed skimming, scanning and inferring tasks (i.e. that are neither too easy nor too hard) can and should provide a very positive sense of achievement.
  2. Readers interact with a text to decode and construct meanings. Well-designed skimming, scanning and inferring tasks can help them in this process, and such tasks might be seen as part of a communicative methodology for language acquisition (rather than as tools to develop reading skills).
  3. It is generally accepted that a useful way of teaching language is to break it down at times into discrete items. It is also generally accepted that these items are best presented in context, and that means within a text. Thornbury suggests that we should treat classroom reading texts primarily as vehicles for language presentation. In reality, this is probably what happens most of the time. The skimming and scanning activities are often very short, dealt with quickly, before moving on to the ‘meat’ of the lesson: the language focus. Whilst these tasks may not offer much payback in terms of reading skills, they seem to offer a reasonably economical way of encouraging students to notice the context of the language that they will subsequently study.
  4. Many students in ELT classrooms may have little or no interest in becoming fluent L2 readers. They do, typically, have a desire to perform well in examinations, and it is common for English language examinations to have a reading component. In both international exams (e.g. the Cambridge First Certificate) and local exams (e.g. the Polish school-leaving test, the ‘Matura’), it is common for the reading component to include tasks which will be best performed if candidates adopt strategies of skimming, scanning and inferring. Classroom practice of these strategies can therefore be seen as psychological and strategic training for the exams.

The arguments of Walter, Swan and Thornbury are, I think, of considerable importance. The approach to reading skills in both coursebooks and teacher training courses (especially of the short, intensive variety) is in urgent need of a rethink, but I hope that the reasons I have listed above are sufficient to demonstrate that we should be wary of jumping to hasty conclusions.

  • Grabe, W. 2009. Reading in a Second Language. Cambridge: Cambridge University Press
  • Soars, L. and J. Soars. 1986 – 2009. Headway. Oxford: Oxford University Press
  • Spratt, M., A. Pulverness and M. Williams. 2005. The TKT Course. Cambridge: Cambridge University Press
  • Thornbury. S. 2006. The end of reading?
  • Walter, C. and M. Swan. 2008. Teaching reading skills: mostly a waste of time? in Beaven, B. (ed.) IATEFL 2008 Exeter Conference Selections. Canterbury: IATEFL