About ten years ago, I taught for just such an
organisation. Following a change in local
management, out went the textbooks and in came a
huge amount of work for the teachers, who were
asked to develop task-based syllabuses for every
single course. Understandably, there was a lot of
resistance to the wholesale upheaval of a system
which seemed to have been working well for the
preceding 70 years or so but within a few months, the
new, task-based approach had to be put into practice.
The overhaul went so far that even preparation
courses for external exams had to tow the task-based
line so, instead of working through a coursebook and
following the well-known ‘PPP’ (presentation →
practice → production) approach, exam students
were involved in activities such as firstly telling each
other anecdotes, afterwards reflecting on their
performance and receiving feedback from peers
before having a second attempt, this time recorded
and assessed. As a teacher of a number of exam
courses, I had my doubts about the effectiveness of
such an approach for helping students obtain the
certificates and grades they needed (for all ‘four
skills’ plus grammar and vocabulary). The school
management were also clearly worried that if passrates for external exams went down, the reputation of
the language centre would be damaged and, as a
consequence, student numbers would significantly
decrease. Nevertheless, the task-based experiment
was allowed to continue and, in fact, there was no
decrease in the pass-rate at all – quite the opposite,
as it happens.
I was impressed by this result, but also somewhat
mystified as to the success of lessons which
appeared to me to be somewhat unstructured,
haphazard, out of my control and in which students
had simply chatted a lot together. In addition, the
courses had even been more fun to teach – surely
this was just a blip! However, exam pass-rates
continued to be high, indicating that the task-based
approach seemed to be working.
Why Should TBLT Work Any
Better Than Other
Approaches? The Theory...
Some years later, I had the opportunity to investigate
the question of why a task-based approach should be
successful, as part of an MA in TESOL. Finally, my
questions would be answered!
Now teaching at a university, I decided to analyse a
series of lessons in which fifteen C1-level students
research and discuss a controversial issue, much as
in the context of an academic seminar.
In terms of the structure of the lessons, I opted for a
more ‘deep end’ TBLT approach, with students first
researching a topic (the Israeli-Palestinian conflict),
formally discussing it together as a class, before
listening to C2-level (in this case, native) speakers
perform the same task and noting down any useful
language that occurred. Following this, one week
later, students repeated the same discussion, without
notes and having received absolutely no language
input from me at all.
To measure if there had been any improvement in
students’ performances, both whole-class discussions
were recorded and the content analysed using
software, to determine how much of the native
speakers’ language the students had taken on board
and re-used. The results were interesting, showing
that students had re-used approximately 16% of the
new phrases and part-phrases (four- to eleven-word
chunks which students had not used in the first
discussion) used by the native speakers. In other
words, without any explicit language instruction and
using a purely task-based approach, students had
acquired and used a significant amount of useful
language, appropriately and correctly in almost all
cases.
So, why did this extremely “hands off” approach
work? To answer this question, it is worth taking a
look at the theories that underpin TBLT. First, there is
Krashen’s (1982) ‘input hypothesis’, which states that
learners acquire language better when it occurs at a
level just above their own (i+1), as in the example
above. Second, Schmidt’s (1990) ‘noticing
hypothesis’ maintains that if learners identify the
language they need themselves, then they are more
likely to acquire it, as they clearly did in the
aforementioned research.
9