Swain (2000) represented output within the
sociocultural perspective by using the term “collaborative dialogue”. In
opposition to interactionist approach, which focuses on individual cognitive
attempt for comprehensibility of message, the sociocultural approach emphasizes
the collaborative effort for co-construction of meaning. As Donato (2004)
stated, ” collaboration is a powerful concept that moves us beyond reductive
input–output models of interaction and acknowledges the importance of goals,
the mutuality of learning in activity, and collective human relationships” (pp.
299–230). Thus, the sociocultural approach stresses the dialogic and dynamic
nature of peer–peer interaction. Collaborative dialogues can be generated
during reading, speaking, writing, listening and even grammatical activities.
Through collaborative dialogue, learners help each other reciprocally during
the process of interaction. This is extension of sociocultural perspective by
claiming that, in peer-peer interaction, learners can act as both experts and
novices during collaboration (Kowal and swain, 1994).
Collaborative dialogue and L2
There are several studies, which argued that
peer-peer dialogue mediates learning. For instance, Storch (1999) found that
collaboration had a positive effect on overall grammatical accuracy of students
during completing form-focused activities such as, a cloze exercise, a text
reconstruction, and a short composition.
Likewise, DiCamilla and Anton (1997) analyzed
the discourse of five pairs of Spanish L2 learners who were working on a
writing assignment. They produced collaborative dialogues during their joint
interaction. In particular, they indicated that repetition helps students in
recognizing features of language. It also results in providing necessary
assistance by peers, themselves, to solve problems of lexis, spelling, verb
form, and etc.
In line with above- mentioned examples,
Tajeddin and Jabbarpoor (2014) investigated the effectiveness of individual
versus collaborative focus-on-form instructional tasks on the acquisition of
English inversion structures by EFL learners. Results showed that collaborative
task showed more success over individual task implementation. In other words,
they concluded that collaborative language production can increase grammatical
acquisition. This study will also examine a grammatical feature (i.e., English
conjunctions), which is a controversial area for EFL learners.
Language- related episodes and L2
Collaborative dialogue can be
defined in terms of language-related episodes (LREs). An LRE is defined as
”any part of a dialogue where students talk about the language they are
producing, question their language use, or other- or self-correct their language
production” (Swain, 2001, p. 287). There are a lot of studies which have
analysed the significance of language-related episode in L2 learning. (e.g.,
Kim and McDonough, 2008; Kowal and Swain, 1994; Leeser, 2004; Swain, 2000;
Swain and Lapkin, 1998, Watanabe and Swain, 2007; Williams, 2001).
Swain and Lapkin (1998) analyzed
the dialogue of two grade eight French immersion students as they
collaboratively wrote a story. They used LREs as their unit of analysis. By
using the results of pre and posttest, they demonstrated that some LREs were
the basis of learning.
It is worthy of mentioning that several
factors such as proficiency, nature of the tasks, context, and even
psychological factors can affect both occurrence and resolution of LREs and the
LREs and proficiency differences:
Proficiency level of the learners
is one of the most important variables that can affect LREs (e.g., Abadikhah
and Mosleh, 2011; Leeser, 2004; Williams, 2001). When teachers assign learners
into different pairs or groups in classroom, the role of proficiency becomes
Different studies in the realm of
proficiency revealed contradictory results. Based on Van Lier (1996), more
proficient learners will be more successful by teaching less proficient peers;
however, other studies asserted that the more proficient learner may repudiate
the input or feedback, which comes from the less proficient one during
collaboration (Yule & Macdonald, 1990).
Williams (2001) investigated the effect of proficiency level on LREs.
She investigated whether English L2 learners from different proficiency levels
varied in terms of the quantity and quality of LREs during collaborative
activities or not. She found that as proficiency of learners increased,
occurrence of LREs increased as well. Furthermore, the learners were able to
correctly resolve their LREs more frequently as their proficiency increased.
Leeser (2004) accepted this result by
indicating that the greatest number of LREs was produced by high-high pairs,
while they wanted to carry out a dictogloss task. Their collaborative dialogue
was also analyzed in terms of the type (lexical or grammatical) and the
resolution of LREs (correct, unresolved, or incorrect). While correct
resolution was the most frequent outcome across dyad types, low–low dyads were
more likely to leave LREs unresolved.
On the contrary, Watanabe and Swain (2007) focused
on individual learners who interacted with both more and less proficient peers.
They wanted to investigate how frequency of LRES changed when four core
participants interacted with eight higher and lower non-core participants. The
results showed that core learners produced more LREs with a higher level peer,
during writing stage. However, core learners produced more LREs with lower learners,
during the noticing stage. So, this study neglected previous results which
showed that growth in levels of proficiency will always lead to more LREs, by
stating that that the pairs who produced both the most and the least LREs were
core-low pairs during noticing stage.