2/19. Critique to Noreen's chapter
For this post I am going a little different and I'll focus on
the draft from Noreen Lape (Chapter 4). Based on my previous readings in the field of SLA, I will critique and give some specific feedback to the author of
this chapter. Essentially I express my reactions, and I point out things I didn't understand or that confused me in the chapter.
1)
I get very confused with the wide variety of terms used in pages
1-6:
English
native speakers, English language learners (ELL), non-native speakers, second
language writers, foreign language writers, non-English FL writing, EFL
writers, ESL writers, English-medium/native language writing.
This seems a lot to me, and then later you end up focusing on
SLA strategies while talking about “FL writers. I would suggest to be
consistent with the terms and use only L1 and L2 writers, L1 and L2
instructors, etc. I would avoid the terms “native” and “non-native”, because
sometimes they can become controversial. Specifically,
the terms native and non-native divide learners in two totally different groups
that may not necessarily exist (e.g. are L2 heritage and bilingual speakers
natives or non-natives?). More importantly, it creates a dichotomy that is
highly hierarchized, since the native-like language becomes the referent for
the non-natives. This ideal of “nativeness”, however, may not necessarily work
in the L2 learning context (and issues related to identity may come into play
here).
Overall,
I would say that SLA research uses the terms “natives” and “non-natives” when
there two very controlled groups of participants, L1s and L2s, and when the
topic being researched is directly related to the issue of being a native
speaker or non-native speaker.
2)
When citing the responses from the tutors, it would be nice to
specify where that data is coming from. Maybe just say something like
“extracted from Interview data” or “extracted from tutors’ journals”.
3)
I am not very sure who is supposed to be the audience of this
chapter, but as someone in the field of SLA I felt somewhat offended in the
second half of page 3 and first paragraph of page 4. There are pretty strong
statements such as:
-
“FL
writing tutors do not have strong set of foundational principles upon to build
a pedagogy”
-
“its
scholars [in FL research] have not always conversed with their counterparts”
-
“FL writing scholar do not even converse with
each other”
I feel these statements are not completely true, and I am not
sure how much they add to the chapter. First, I am not sure there is a
significant difference between SLA and FL writing when it comes to L2 research.
Second, the statements seem to be supported by only one reference (Reichelt et al.). It would be more reliable to include
references from other peer-reviewed articles . Third, I am not sure why you are
problematizing not having a common framework across all languages. There is a
common framework in SLA: It’s common to examine the same issues across the
different languages and evaluate the extent to which these issues are
generalizable to all languages. Overall, I think I
would try to re-orientate this part from a more positive perspective, for instance talking about the need for more
research in L2 tutoring, or talking about the added difficulty of writing in an
L2, which involves mastery of both the language itself and the L2 writing conventions.
4)
In page 5, the models “writing-to-learn” and “learning-to-write”
are definitely very commonly used in L2 research. Are these models the same as
the pedagogies pointed above in (a) and (b) respectively? If that’s the case, I
think I would make it more clear, since the 3 pedagogies may not sound as
familiar to everyone.
5)
On page 6, I would revise the sentence “some FL writing truncate
the writing process into two steps (composing and editing) […]”. Again, that
sounds a little strong and vague. Are we talking about courses? students?
researchers? I think any L2 scholar would agree that the L2 writing process
goes beyond just 2 or 3 steps.
6)
From pages 7 to 12, I think I am missing a section that encompasses
or helps us understanding
why you are discussing the “Output Hypothesis” and the “Interaction
Hypothesis”. These two hypotheses,
together with the “Input Hypothesis” (Krashen, 1982) and the Noticing Hypothesis
(Schmidt, 1995) are all processing-related theories. They all focus on the
“processes” and use cognitive concepts to explain how interaction and learning
are linked. I think I would like to see a broader picture before you actually
start focusing on Output and Interaction Hypothesis. It could be a small section, but by including it you would
be setting the readers’ mindset into a Cognitive-based approach.
Also,
for the processing-related hypothesis, my understanding is that these cognitive
processes happen in the head of the learners. So I understand that “instructors
can prompt students to employ SLA techniques [processes?]”. However, I am not
so sure about the ‘techniques’ [processes] being “essential tools for a FL
writing tutor” (p.7). When you started mentioning the hypothesis I thought we
were talking about cognitive processes that should happen in the mind of the learner
writers. But then when you go and apply it to the
mind and the conversations of the tutors, it feels a little forced. I think I
would try to at least make it smoother. Maybe give the real definitions first,
and then make the transition to how you apply it to tutors? For instance, “hypothesis
testing” occurs when learners test their knowledge (hypothesis) and discover
that their utterances are not being understood. It would be nice if you could
give a solid and extended definition first (based on learners’ cognition) before
you actually say “FL writing tutor feedback can initiate hypothesis testing”.
Comments
Post a Comment