Delphi module

2.3.3.2
The importance of interaction

Swain argued that what the immersion students lacked were opportunities to 'output' more, ie to express themselves in more and more complex tasks. You may remember that Krashen's Input Hypothesis played down the significance of oral production (see section 2.3.2), emphasizing only comprehensible input. Swain argues for an Output Hypothesis, suggesting only through more production tasks would post-intermediate learners be able to try out more complex grammatical structures and get more feedback on their attempts. This in turn would 'push' their grammatical development in the second language.

At more or less the same time as Swain's investigation, Michael Long (1983) was proposing similar modifications to Krashen's Input hypothesis. Krashen seems to put the listener (the learner) in a rather passive role: it is the speaker who adjusts language to make it comprehensible. And Krashen was not very precise in clarifying what kind of adjustments to input might be significant to the acquisition process. Long suggested in his Interaction Hypothesis that a significant factor might be opportunities for learner and speaker to interact and 'negotiate meaning', rather than just the speaker's adjustments of vocabulary and grammar.

Reflective task 18

Imagine you are giving directions in your native language to an intermediate level non-native speaker. How would you ensure that you get the message across? Now imagine the roles are reversed and you are the non-native speaker: what could you do to try to ensure that you get the message?
Based on Long's Interaction Hypothesis, Pica et al (1987) compared the comprehensibility of two types of native speaker input. Two groups of non-native speakers had to follow instructions from the native speakers in order to complete a task involving physical placement of objects (eg put the umbrella drawing next to the toy dog at the back of the picture). In both groups, the input was delivered to the non-native speakers individually:
- Those in Group A received input that was modified by the speaker beforehand to try to ensure that it was 'comprehensible' for the level of the learners involved. There was no interaction between the NS and the NNS.
- Those in Group B received input that was not pre-modified by the NS, but interaction between the NS and the NNS was allowed: both could seek confirmation and clarification.
Which group do you think demonstrated higher comprehension levels? Why?

Click here to view Commentary.