david-hutchinson.ca

Resources for applied linguistics and language teaching

Language Acquisition Research

I’ve been doing some review of second language acquisition, and one thing has struck me as I’ve read about some of the research.  The way that a lot of the research is done seems problematic, especially the research which attempts to answer how language is acquired.

Much of this research follows the format Pretest – Treatment – Post-test, with treatment differing between the experimental and control groups.  The treatment is usually some kind of instruction, opportunity to practice, or other exposure to language, especially a particular grammatical structure.  Even this is a big shortcoming of such studies, as they often equate language acquisition with acquisition of morphosyntax.  The pretest – treatment – post-test format is pretty standard, but the interpretations made are sometimes a bit flimsy.  Either they will find that the experimental group outperformed the control group (as a group!) and conclude that the treatment leads to acquisition, or they’ll find the experimental group didn’t outperform the control group (as a group!) and conclude the treatment does not lead to acquisition.  Perhaps I’m overgeneralizing a bit, and they will usually hedge a lot, but that’s basically it.  I think they’re oversimplifying language acquisition a bit, and here are my two reasons.

First, they make assumptions based on the group.  If the experimental group outperforms the control group (p< .05!), or on the other hand if it doesnt, they make conclusions about acquisition in general, assuming that all learners will acquire language in a similar way, thus for example assuming that either all learners or none will benefit from grammar instruction.  Although this may be somewhat true in the case of L1 learning, or for very young learners, there is a lot of evidence that older learners and adults learn a second language in different ways, that they have different learning styles and that some tend to prosper more in a structured environment with explicit instruction, whereas others prosper more in an immersive environment.  Most acquisition research has ignored this.

Another problem is they often assume that the effects of the treatment should be manifest immediately.  But I think it’s completely reasonable that instruction will have some beneficial effects some time after, possibly well after, the instruction took place, as long as other prerequisites of acquisition are met (sufficient input and opportunity for output).  For example, if the third person singular (eats instead of eat) is taught too early for the learners to begin using it right away, the instruction may nevertheless have a beneficial effect later when they are ready to learn the structure.  Of course, some studies include delayed post-tests, but delayed post-tests are rarely sensitive enough to give reliable results; and how are the experimenters to know when to place the delayed post-test?  If it is too early or too late, it may not have a chance to give positive results.

Anyways, it’s obviously a bit too easy to criticize research in a field like Applied Linguistics, since so much of it is flawed and it’s really quite impossible to get firm results that are expected in the hard sciences, just by nature of what is being studied.  So I’m reluctant to play the armchair critic; but what I want to say is that basically, interpretations of experimental research need to be made very cautiously, and I try to avoid strong claims about the nature of language acquisition.

If asked my own view, which comes from a composite of experimental research, my own experiences (as both teacher and learner), listening to others’ experiences, and intuition, is that acquisition of a second language comes through both comprehensible input, output (both of these being necessary conditions), and that language instruction and focus on form is usually beneficial, although with the caveats that it depends on the learner, and that inappropriate instruction without input / output can be detrimental.

 

Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Post Navigation