This is a report of a session conducted at the IATEFL conference in Harrogate in April 2006
The Listener as Lemming: Listening Strategies Re-visited John Field
John is a major researcher into the listening skill, and this was a session which provided much food for thought.
Are listening strategies problem-free? Could strategies lead students into trouble? Are they more influenced by top-down strategies? Or are they more influenced by the co-text – what they’ve just heard?
Skills and strategies:
Skill – a process which an L1 learner practices
Strategy – a technique which an L2 learner would need to overcome gaps.
Not wonderful definitions; plus there’s overlap
L2 learner – a) limited L2 knowledge b) limited experience of performing in L2
L2 learners have 2 main strategies
1 = aimed at learning 2 = aimed at overcoming problems.
Today we’re looking at strategy 2
- Avoidance (smile and nod)
- Achievement (try and make sense of bits you’ve understood)
- Repair (could you say that again?)
Graded L2 materials assumptions:
- L2 listener listens like an L1 learner
- Many L2 listener problems are caused by limited vocabulary and grammar
- Because of this, L2 listener has to focus attention to the input and pays no heed to the wider meaning
“Thee solution to gaps in understanding is more knowledge of grammar and vocabulary”
– I challenge this.
An important point that John makes: Context use is not being used to deepen understanding, but to compensate gaps in understanding.
Research experiment: a paused listening in which learners were asked to write down the last 3/4 words they’d heard. The words were unpredictable; the text was authentic and naturalistic; the pauses could not be predicted.
They often did not remember the words, but they wrote down the meaning/ideas. This suggests the importance of bottom-up strategies in listening
A lot of bottom-up information is missing in listeners until they are a very high level (advanced). Learners actually understand depressingly little.
According to Stanovich’s 1980 Interactional Compensating Hypothesis:
We use more bottom-up skills in L1 in a comfortable listening environment (e.g. a conference centre)
In a local pub we use more top-down strategies (external information)
Field is not so sure that this is true.
It is claimed that there are 3 components of strategic listening
- quality and quantity of intake
- degree of confidence about intake (and so I can make hypotheses)
- accuracy of inferencing (some are better at guessing than others)
However, research shows that …..
Potential problems 1: novice listeners
- hears an unknown word, interprets it as a known one and builds a false hypothesis about rest of text
- hears a known word, changes it to another to fit a pre-established schema
Potential problem 2
- Risk avoiders – don’t compensate
- Risk takers – don’t test their assumptions
Potential problem 3
- Forms a hypothesis on sentence A which is then contrasted by Sentence B. They choose sentence A.: time and faith have been invested in hypothesis A even if B sounds more plausible
Text played: I found out that the thud was the cat. Learners wrote (instead of thud, an unknown item) sound
sun in the cat
frog and the cat
They were not using top-down strategies, but were making lexical hypotheses. Listening materials (and teachers) tell learners how they should infer meaning from context. In reality, they prefer to match it with another word in their lexicon.
Students heard butcher and wrote putcher puccha pootscha
Arab students over-compensating
Sts heard butcher’s shop in the middle of the High Street and wrote bookshelf in the …
They were making an approximate match.
So learners make approximate matches. This resembles L1 listening in many ways. L1 listening is not precise, either. John gave an example from that very morning. He was looking for the conference centre, which is situated in King’s Road. He asked a passer-by King’s Street, please? Reply: East Street? The listener knew of East Street, and that was the nearest match he could make.
- L1 listening is made up of ‘goodness of fit’. The listener weighs up different pieces of evidence about what has been said.
This happens in reading, too.
T /-\ E C /-\ T
In word 1 we read the symbol as H, whereas in word 2 we read the same symbol as
A. Goodness of fit.
So L2 listener brings in:
- grammatical probability (I am waiting [for] )
- lexical frequency (prefer book to hook)
- frequency of chunks
- knowledge of the existence of words
- phonemes in the input
Of the above, if you were learning a language, which of the above would you be more confident about using? Probably 1 and 6.
We should explicitly teach strategies e.g.
After usual strategies (gist, intensive etc):
Stage 3: write down words you heard – discuss – no answer from teacher
Stage 4: write down more words
Risk takers and risk avoiders are taken into account.