14.488, Disc: Re: Review:Applied Linguistics: Robinson (2002)

LINGUIST List linguist at linguistlist.org
Wed Feb 19 14:35:51 UTC 2003

LINGUIST List:  Vol-14-488. Wed Feb 19 2003. ISSN: 1068-4875.

Subject: 14.488, Disc: Re: Review:Applied Linguistics: Robinson (2002)

Moderators: Anthony Aristar, Wayne State U.<aristar at linguistlist.org>
            Helen Dry, Eastern Michigan U. <hdry at linguistlist.org>

Reviews (reviews at linguistlist.org):
	Simin Karimi, U. of Arizona
	Terence Langendoen, U. of Arizona

Home Page:  http://linguistlist.org/

The LINGUIST List is funded by Eastern Michigan University, Wayne
State University, and donations from subscribers and publishers.

Editor for this issue: Karen Milligan <karen at linguistlist.org>
To post to LINGUIST, use our convenient web form at

Date:  Wed, 19 Feb 2003 02:35:06 -0500
From:  "Ronald SHEEN (UQTR-Langues modernes)" <Ronald_Sheen at UQTR.UQuebec.ca>
Subject:  Re: 14.233, Disc: New:Review:Applied Linguistics: Robinson

-------------------------------- Message 1 -------------------------------

Date:  Wed, 19 Feb 2003 02:35:06 -0500
From:  "Ronald SHEEN (UQTR-Langues modernes)" <Ronald_Sheen at UQTR.UQuebec.ca>
Subject:  Re: 14.233, Disc: New:Review:Applied Linguistics: Robinson

Re Linguist 14.233

What follows is an update of what has occurred since my response to
Ana Llinares review of Applied Linguistics: Robinson (2002),
(Linguist 14.233) written on January 23. I submit the following
partly as a means of putting on record the positions adopted by
various applied linguists on this issue and partly as a means of
stimulating discussion of this crucial hypothesis.

First, as members may have noticed, there were no on-List responses.
However, the reviewer, Ana Llinares did reply to me as follows and
gave me permission to cite it:

"Thank you very much for your comments. In his chapter, Robinson does
not focus on how much grammar adults learn incidentally. The focus of
his study is on whether incidental learning (versus explicit learning)
is also affected by individual differences relevant to the demands of
the task.  Obviously, there is the assumption here that adult learners
learn "some grammar" incidentally. Having said that, I completely
agree with you on the lack of empirical evidence. To respond to your
question: No, the author does not provide any empirical evidence. In
general, all the chapters in this book claim the need for more
empirical work but most don't give any examples of this kind of
work. For example, neither of the two chapters on classroom studies
provide examples. They are very interesting quantitative studies but I
think there should also be a qualitative analysis of specific
examples. I should have highlighted this point much more in my review,
although I suggest, at some point, the need for more relevant

The author of the book, Peter Robinson, also replied to me but did not
give me permission to cite him on this List.  However, in his
response, he did mention studies on naturalistic learning such as the
well-known Zisa project.  However, from what I have read of such
studies, they do not demonstrate a general acquisitional process by
learners going through developmental sequences ending up with the
ability to produce correct native-like language.  In response to Peter
Robinson's reply to me, I sent him the following further inquiry on
February 11, to which I have so far received no response:

"Many thanks for your reply.  I will track them down providing I can
be sure that there is evidence therein of accurate spontaneous oral
production demonstrating the complete acquisition of a range of
grammatical features.  Is there such evidence?  I'm pretty sure that
the Zisa evidence does not contain such accurate spontaneous oral
production of a rance of grammatical features.  The again I read it
quite a while ago.  Further, I had the impression from what Pienemann
wrote to SLART a few years back that there was in fact no such
evidence which was one of the reasons that he had decided to devote
more time to developing the theory.  Than again, I could be wrong.

However, I think there may be a small misunderstanding.  I thought the
fact that I had mentioned "without pedagogical intervention" and the
fact that the context of my questions was the Long and Lightbown
positions on incidental learning, I was referring to classroom
learning.  In this context, I do not think that findings from
naturalistic learning can be used to justify advocacy for classroom
strategies unless in a complete immersion context.

However, even if one takes data from naturalistic learning, are you
saying that there is data derived from spontaneous oral production
which demonstrates learners passing through developmental sequences
and finally producing formally correct utterances demonstrating the
complete acquisition of some grammatical feature. and this based
purely on exposure to CI and with no benefit derived from instruction
of some sort. If you do have a reference for such evidence, I'd
appreciate your giving them to me.  In my experience from reading such
research, transcriptions of such speech are conspicuous by their

As to classroom learning without pedagogical intervention, I'm pretty
sure there is NO evidence in the literature demonstrating students
passing through developmental sequences and ultimately producing
spontaneous correct speech.  Long and Lightbown certainly provide no
such evidence in support of their statements.  And from what you
write, it would appear that you do not provide such evidence either
based, that is, on the spontaneous oral production of learners who
have not benefited from ANY instruction.  If you have then, once
again, I'd like to read it.

My feeling is that classroom incidental learning resulting in accurate
systematic production is very much bereft of the support of empirical
evidence.  Without such clearly compelling evidence, I consider it
unwise for applied linguists to write as do Long and Lightbown.

I'm intriqued as to why you do not wish to enter into discussion on
this issue on the Linguist List.  As I have already written on several
occasions, I'm surprised at the complete absence of response from
authors of reviewed books to criticisms of what they have written.
Perhaps naively, I consider that we are all accountable for what we
write and the implicit refusal to respond to criticism is also a
refusal to be accountable.

The issue of classroom incidental learning without pedagogical
intervention is a simple issue.  If applied linguists have the
compelling supportive evidence, they have every right to base their
advocacies thereon.  Without it, they are doing a disservice to
classroom teachers for all they are doing is perpetuating myths.  The
last fifty years has produce so many such myths that one might expect
the perpetuators of myths to have learned from the past.
Unfortunately, Santayana's famous warning is as relevant as ever.

By the way, you may have more faith in grammaticality judgments and
guided sentence production than I do.  However, in my view, such
evidence alone is far from adequate to justify Long's and Lightbown's
position on incidental learning.  Ultimately, spontaneous correct oral
production is what best reveals underlying grammar.  Without it no one
has the right to make claims about what is and what is not acquired"

As further information on this issue, the recent response from Patsy
Lightbown in the current issue of Applied Linguiistics to my initial
response (also in the current issue of AL) is of interest.  Therein, I
objected to her generalisation that adults and adolescents are able to
acquire some features of language incidentally without pedagogical
intervention.  I wrote the following: 1.  Adults and adolescents can
"acquire" a second language.  This generalization is problematic
because of the lack of precision in the use of the term "acquire".  L
informs us that she uses it in the Krashen sense, i.e., the
unconscious process akin to that of first language acquisition and
that "Classroom research has provided additional support for the
conclusion that some features are acquired incidentally - without
intentional effort or pedagogical guidance.".

There is little doubt that exposure to meaningful language may result
in good comprehension skills.  However, whether it leads to accurate
language production is another matter and a crucial one for as Spada
and Lightbown (1993:208) point out, it is "spontaneous and free oral
production tasks which provide a more accurate reflection of the
learners' internal grammar.". This being so, Lightbown needs to make
clear whether she includes both comprehension and production skills
and which features are involved. In fact, Lightbown has so far
provided no empirical evidence to support the generalisation in terms
of accurate production skills.

In her response to this in the current issue of Applied Linguistics,
Patsy Lightbown now writes: "Sheen challenges the validity of the
generalization that adults and adolescents can "acquire" a second
language.  This seems to be due to a difference of opinion about what
"acquire" means.  Sheen apears to treat 'acquisition' as equivalent to
use language with a high degree of accuracy.  His focus is on the
failure of adults and adolescents to achieve 'accurate language
production' in the absence of instruction.  This is an issue that is
also of concern to me and that I dealt with elsewhere in Lightbown
(2000).  However, as I suggested there, accuracy as measured by
conformity, is only one way to assess 'language acquisition' .  When
learners use non-target forms such as 'he speaked' or 'three boy' or
"I don't speak very well English' they may be revealing internal
grammars which, while not identical to that of the native speaker,
have their own patterns of consistency.  These errors reflect the
learners' CURRENT LEVEL OF DEVELOPING KNOWLEDGE' (my emphasis) and
their current ability to use their knowledge of the L2.  That is, they
reflect the learners' acquisition of the language TO THIS
POINT. (Lightbown's emphasis)."

This response of Lightbown is also implicit in the position adopted by
Robinson and others.  However, it brings us back to THE ISSUE.  That
is the following.  The hypothesis of "incidental learning without
pedagogical intervention" obviously entails learners' using incorrect
forms at some stages in the learning process.  However, the argument
then stands or falls on the availability of empirical evidence
supporting the reality of developmental sequences.  Long (2000)
implies that such evidence exists but provides none.  Robinson does
likewise but again provides none.  Lightbown does likewise but in
providing none, does so in spades.  In Schmitt (2002) reviewed by
myself on this List, Spada and Lightbown contribute a chapter
including hypothesising on the reality of developmental sequences in
question formation.  However, they provide no evidence in support.
Nor do they address the fact that many learners begin by forming
questions on the model of "WH + subject + verb" (What your brother
like?") and continue to use such forms for as long as ten years thus
demonstrating evidence of fossilisation.  In doing so, they implicitly
provide evidence of the ABSENCE of developmental sequences.  This is
also implicitly absent in Lightbown et al. (2002), an account of a
six-year study of comprehension-based learning in New Brunswick,
Canada.  Not only do the authors fail to provide not a single
transcription of what any student was able to produce, they treat the
issue of developmental sequences with a deafening silence.

Thus to end this discussion as I started it,
just as other myths have been created by proposing seductive
hypotheses WITHOUT supportive empirical evidence, the same is now
occurring in the case of "incidental learning without pedagogical
intervention".  It will continue to develop and prosper if anonymous
reviewers and editors continue to publish the claims without insisting
on the provision of supportive empirical evidence.

The issue is of crucial importance to the applied linguistics world of
advocacy of teaching strategies because the acceptance of the validity
of the hypothesis implicitly justifies a "focus on form" as advocated
by Long and others and implicitly rejects the necessity for an
approach based on a focus on formS as advocated by DeKeyser (1998) and
Sheen (2003).  However, any objective evaluation of the available
evidence going back to the comparative studies of the 60's justifes
only one conclusion which is the following:
In teaching programmes in which accuracy of production has some
importance, an enlightened "focus on formS" approach is a necessity
for ALL features of grammar.

Ron Sheen.
Department of Modern Linguistics,
University of Quebec in Trois RiviE8res,
Quebec, Canada.


DeKeyser, R.M. (1998). "Beyond focus on form: Cognitive perspectives
on learning and practising second language grammar" in C. Doughty &
J. Wlliams (Eds.) Focus on Form in Classroom Language Acquisition,
(pp. 42-63) Cambridge: Cambridge University Press.

Lightbown, P. (2000).  "Anniversary article: Classroom SLA research
and second language teaching".  Applied Lingustics, 21: 431-462.

Lightbown, M. P., Halter, H. R., White, J. L. & Horst, M.  (2002)
"Comprehension-Based Learning: The Limits of 'Do It Yourself' ".
CMLR, 58: 427-464.

Lightbown, P. (2002)  "Response to Sheen"    Applied Linguistics 23-4.

Long, M. (2000) "Focus on form in task-based language teaching" In
Language policy and pedagogy: Essays in honor of A. Ronald
Walton. R. D.  Lambert & E. Shohamy (Eds.) 179-192. Philadelphia: John

Long, M. H., & Robinson, P. (1998). "Focus on form: Theory, research
and practice" in C. Doughty & J. Wlliams (Eds.) Focus on Form in
Classroom Language Acquisition, (pp. 15-41) Cambridge: CUP.

Schmitt, N. (Ed.) (2002) An Introduction to Applied Linguistics. Edward
Sheen, R. (2002)  A Response to Lightbown's (2000) Anniversary
Article.  Applied Linguistics, 23-4.

Sheen, R. (2003) "Focus in form a myth-in-the-making" English Language
Teaching Journal, 57: (pages forthcoming).

LINGUIST List: Vol-14-488

More information about the Linguist mailing list