Sunday, October 13, 2013

SLA Research and Language Teaching by Rod Ellis




Why I Read This Book
          This book isn’t pleasure reading, but I’m trying to make a commitment to read 10 pages a day of something related to professional development.
            My selection is somewhat limited out here in Cambodia, but this book was in the teacher’s resource center at my school, so I picked it up.
           
            This book was published in 1997, which may well mean some parts of it are out of date by now.  (I’m not an expert, but I get the impression Second Language Acquisition is a rapidly changing field.  When I read Understanding Second Language Acquisition by Lourdes Ortega, I was struck by how much of the research she cited was from within the past 10 years.)  However, as the selection in Cambodia is limited, I didn’t have a lot of other choices for professional development reading, so I went ahead and read this book anyway.
           
The Review
          Someone more knowledgeable than me will probably be able to detail exactly how the research has advanced in the last 15 years, and which parts of this book are now out of date.  I didn’t notice anything myself.  The theories and descriptions of language learning in this book seemed to me to be in line with what I had learned in my Applied Linguistics degree and what was in Lourdes Ortega’s book.   (The latter is actually not so surprising considering that Ortega frequently cites Ellis in her book, indicating he’s probably one of the major players in the field.)

            It’s an academic book, but it’s surprisingly readable.  Rod Ellis is one of those rare academics who can actually write well (a point worth highlighting, because it’s certainly not true of everyone—see HERE.)
            It’s not the easiest book I’ve ever read—I definitely had to put in some work to engage this—but once I got into the book it was interesting.  I enjoyed following Ellis’s explanation of how the mind processes second languages.  (Linguistics was not originally my first love, but when you spend all day teaching languages you can’t help but develop an interest in how your teaching is being processed in the minds of your students.)
            I don’t think I would recommend this book to the uninitiated—my understanding of this book was greatly helped by having already read Lourdes Ortega’s Understanding Second Language Acquisition, and so I was already familiar with the theories Ellis discusses. 
            And yet….I don’t know, Ellis writes in such a clear straightforward way that it might just be possible to struggle through even without a lot of background knowledge.

            Most of the chapters originally started life as separate articles all published separately in scholarly journals.  I believe this is a common convention for academics (to take articles they’ve previously published and then re-work them slightly and publish them in book form) and although all of the chapters are broadly related to the title of the book, SLA Research and Language Teaching, the shift in focus between chapters can be a bit jarring.  One chapter will give a broad view of the field, focusing on everything modern research can tell us to date about SLA and language teaching, and then another chapter will drop the reader into a detailed review of some obscure small study.  (To be fair, the smaller studies are reviewed not so much for the value of their results, but for the sake of giving examples of teacher conducted research.)

            To me the most interesting chapters were the birds’ eye view chapters, in which Ellis reviews the findings of second language acquisition, and then makes suggestions as to how language teachers might make use of this.

            There’s a lot of information packed in these sections, but I’ll just focus on things that caught my interest.

            The first question Ellis address is whether language teachers should teach grammar at all. 
            There is a school of thought (most closely associated with Stephen Krashen) that learners will acquire the grammar of a language naturally if they are simply provided with comprehensible input, or opportunities to communicate.
            In fact Krashen argued that actively teaching grammar will actually hinder the acquisition of a language, because it will cause learners to consciously monitor their speech which will interfere with the natural acquisition of the language.
            Krashen’s ideas are appealing for anyone who’s ever been bored by a grammar lesson, but unfortunately the research does not appear to bear out all of his claims.  Learners can acquire some simple grammatical forms naturally, but there are many more grammatical forms which are more difficult to acquire.  Students who learn language from comprehensible input alone (such as French immersion students in Canada) often do not acquire the more difficult grammatical forms.
            Moreover the research indicates learners can benefit from direct grammar instruction.
            So then, if grammar should be taught, how best to teach it?

            One of the interesting findings of SLA research is that learners appear to always acquire the same grammatical structures in the same order.  (This is something I had previously encountered in my Applied Linguistics course, as well as in Lourdes Ortega’s book).  Below is an example of the order of acquisition taken from Ortega.
-ing
plural –s
Be copula
                                                
Be auxiliary
a/the
                                               
Irregular past
                                               
Regular past –ed
Third person –s
Possessive – ’s

            For example, learners will always master the -ing for of a verb before they can master the be auxiliary, and they will always master the be auxiliary before they master the irregular past, and they will always master the irregular past before they master the third person –s.  It doesn’t matter what first language background the learners are, it doesn’t matter if they are learning naturalistically or if they are in a classroom, and it doesn’t matter in what order the textbook tries to teach the grammar—for some reason, this is the order that all the learners must go through.
            Since the learner can not master the 3rd person –s until they have mastered all the previous forms, the temptation is therefore to say that the teacher should never attempt to teach the 3rd person –s until the learners are developmentally ready for it.  (In fact I myself have heard some language teachers suggest as much.)  But this, Ellis says, is impractical.  Among a number of other practical reasons he cites, for one thing learners develop at different rates, so if you have a class of 20 students, you’ll never get them all to the same stage of development at the same time.
            A sometimes overlooked point that Ellis emphasizes is that the order of acquisition applies only to implicit knowledge—the ability to use language in a free unstructured conversation.  However learners might have knowledge of much more grammar rules than they are able to use.
            The exact nature of the psycholinguistic constraints on language learning are still unknown, but the theory that makes the most sense to me is that the human mind is capable of storing in long term memory a lot of explicit knowledge about grammatical rules, but in real time conversation the mind can only focus its attention on a limited number of items at one time.  So in unplanned conversation it’s impossible for the learner to focus attention on all the grammatical rules at the same time.  When producing a sentence, the learner can only focus their attention on one or two grammatical aspects, and the rest will get neglected.
            Eventually, however, with enough practice, certain grammatical features will get practiced so much they will become automatic, and then the learner can say them without having to think about them.  And at this point they can shift their attention to focus on the next grammatical aspect in the order of acquisition.
            This is known as the interface position—the idea that explicit knowledge of grammar in the brain can, through practice, become implicit knowledge.
            Ellis rejects the strong interface position: “What appears to be happening in [theories of the strong interface position] is the equating of controlled processing with explicit knowledge and automatic processing with implicit knowledge” (p. 113)
            Ellis does, however, believe in the weak interface position—the idea that explicit grammatical knowledge by itself does not convert directly into implicit knowledge, but instead explicit knowledge of grammar may help the learner to notice these grammatical features in the input, and may also help the learner notice the gap between their own output and the language norms.  And this may help the learner intake and process these new forms.

            The point of all this, as far as Ellis is concerned, is that learners can still learn the grammar rules explicitly even if they aren’t ready to use them automatically in conversation.  If the learner has explicit knowledge of a grammar rule they are not developmentally ready to use yet, this will do them no harm.  And when they do reach the stage where they are developmentally ready to acquire the grammar rule, explicit knowledge will help them to acquire the grammatical form more quickly.

            Therefore Ellis suggests that learners can be taught awareness of grammatical rules even if they aren’t ready to use them. 
            Ellis suggests one way of doing this is to teach some grammar points for comprehension only—not expecting the learner to be able to use them in production, but making sure the learner knows how the grammar works so that they can understand it in the input.  For example Ellis suggests comprehension exercises where learners identify which picture goes with which sentence.

            Of course, Ellis acknowledges that for explicit knowledge to someday become implicit, learners will need plenty of time to practice free production, and this should also be an important component of any English course, but Ellis doesn’t go into much detail about this.

            Ellis’s theory on teaching grammar conflicts slightly with what I was taught in my CELTA (W) course.  According to the CELTA method, every grammar lesson should guide the learners from comprehension, to controlled production, to free production.  Ellis advocates only comprehension (and he also includes some forms of controlled production, like filling in the blank, as conscious raising activities) of new forms, and advises against having the goal of every lesson be free production of the new grammatical form.
            In fact, on page 90 Ellis says, “an assumption of much grammar teaching is that it is possible to lead learners from controlled practice to free practice and in so doing teach implicit knowledge (i.e. enable learners to make the necessary changes to their interlanguage so they can use the new structures when they are communicating). There is, however, very little clear evidence in support of the claim that practice makes perfect or that structures can be taught by taking learners through the text-manipulation/text-creation continuum.  In a review of studies that have investigated the effects of practice, Ellis (1988a) found little evidence to support the claim that more learners practise the better they become. In a study that sought to teach adjectival order in English nominal groups to Japanese college students by means of a series of activities strung out along the text-manipulation/ text-creation continuum, Tuz (1993) found that learners were unable to make the leap from controlled to communicative use of the structure.”  (p.90)

My Own Teaching Philosophy
          Ellis notes that a language teacher’s practices and philosophy are seldom the results of dogmatically applying research, but rather the result of the teacher’s experiences.  And this has been true for me as well—not only my experiences as a language teacher, but also my experiences as a language learner.  (This is perhaps something Ellis neglects to consider—how someone’s own experience learning a foreign language can affect their teaching.)
            I’ve always believed that grammar was an essential component of learning a foreign language, in part because of my experience in Japan.
            Most of us in the JET program arrived having little or no Japanese, and were immediately thrown into an immersion foreign language environment.  I had friends who seemed to miraculously pick up a consider amount of Japanese just by constantly immersing themselves into the language, but this didn’t work for me.  Attempts to pick Japanese up through exposure or through conversation just gave me a headache and made me feel tired.  I felt like my brain was drowning in an ocean of Japanese language that I didn’t understand.
            What helped me was consciously studying and memorizing grammar rules out of books, because this gave me a structure on which I could try to hang all the new words and phrases that I encountered and also made me less intimidated by the deluge of Japanese I was constantly exposed to.
            For this reason I always believed it was beneficial to teach my students grammar.
            However, despite believing that grammar instruction has some value, I spent 5 years working in the Japanese public school system, which is notorious for producing students who learn all the rules of English grammar, but still can’t speak a word of the language. 
            (As Ellis says in his book, “A good example of [learners who are only taught the explicit rules of grammar]is high school students in Japan.  After six years of studying English, much of which is taken up with the teaching of grammar, many of these students leave school with no procedural ability to communicate in English.” p. 75-endnote 10)
           
            The students in Japanese public schools are so shy and self-conscious about their English ability that they are very reluctant to speak the language, and so consequently never acquire procedural ability.
            One of my current colleagues once said that teachers are always reacting to their last disaster.  If you have a class which is low on listening, you become obsessed with teaching listening for months afterwards.  You continue this until you have a class that is low on reading, and then you become obsessed with teaching reading.
            This was true of me, and while working in the Japanese educational system I became obsessed with just trying to get students to speak and not to worry about grammatical accuracy when they spoke. 
            (I still maintain this hang-up to a large degree.  A senior teacher at my school recently suggested to me that I should be more aggressive about correcting students’ grammar during speaking activities, and I replied that my experience in Japan had made me very reluctant to correct students during free production activities.)

            However, my next job in Japan was at an English conversation school, where the only focus was on conversation and there was no grammar instruction given.
            Here I observed the opposite problem.  Many of the students who had been attending this conversation school for many years had achieved a high degree of fluency in the language, but had severe problems with grammatical accuracy.
            It is as a result of this experience that I am a firm believe that grammar can not be learned simply from exposure to the language, or even from communicative activities or negotiating for meaning. 
            Every language school has at least a few adherents of Krashen, and I have a couple co-workers who would like to do away entirely with the grammar component of our language courses, but my experience has made me a firm believer in teaching grammar.
            That being said, I like Krashen’s theories in so far as I can regard them as another tool in my toolbox.  I don’t like the restrictive aspect of Krashen—the idea that a language teacher shouldn’t teach grammar—but I do like to use his methods as a supplement to grammar teaching, and for this reason I like the idea of providing the learners with lots of comprehensible input. 
            Ellis mentions some of the reasons why Krashen’s ideas have become so popular with language teachers is because Krashen’s theories are very easy to understand, and they are very easy to adopt.  And I probably count myself in that category.  The idea of simply providing students with lots of comprehensible input is not only appealing in its theoretical aspects, but very easy to do practically.
            (Ellis also cites Krashen as an example of a theory which “may be seriously flawed and yet still give rise to useful proposals.  For example, Krashen’s Input Hypothesis has been subject to considerable criticism (e.g. Gragg 1984, White, 1987). However one of the pedagogic proposals derived from it—namely that teachers should provide learners with plenty of opportunities to read through an extensive reading program—has arguably had a beneficial effect on L2 programs for ESL learners in the USA.” (p.104))

            For this reason, I’ve become a convert to the extensive reading idea.  I still supplement it with traditional grammar teaching, but my hope is that by the time I get around to teaching a certain grammar point, the learners have already encountered this point many times before in the input because of extensive reading, and therefore the explicit grammar instruction will just be a clarification of the rules on a structure they are already somewhat familiar with.
           
            As for how to teach grammar: although I’ve been trained in the CELTA style, my own experience is very similar to Ellis’s research—when teaching a new grammar point it’s difficult to get the students to make the jump from controlled production to free production. 
            This is especially true when dealing with students whose motivations are low (for example, a class full of 13 year-olds who are not in the classroom by their own volition, but by their parents’ directive).  And I’ve found with teenage learners most of the free production or communication gap activities I set up for them quickly fall apart, and they just revert back into their native language.  So for practical reasons as much as anything I have been finding myself doing lesson plans that are very similar to Ellis’s theories—I try and raise the students’ awareness of grammar rules, but don’t force them to produce new grammar points they aren’t ready for.
            Therefore the proposals contained in Ellis’s book were not very different from what I was already doing, however, it was still nice to get some theoretical justification for what I was already doing.

Notes, Nitpicks, and other Addenda

*******Group Work*******
          As Ellis mentions, teaching grammar rules is only one part of a language curriculum. For these grammar rules to become procedularized, students will need an opportunity to practice them in conversation.
            In foreign language settings, this is not always easy to do.  If you have a class of 20 students, not all the students are going to get adequate time to talk with the teacher.
            The solution in most classrooms then is group work, where the students are given a communicative task, divided into groups, and then practice speaking with each other.
            This is standard procedure at almost all language schools, including my own.
            Ellis, however, cites some disturbing research about group work.
            On page 52, when talking about why French immersion students in Canada never develop beyond “a very defective and probably terminal classroom pidgin”, Ellis writes: “The reasons advanced for the failure of immersion in Hammerley’s eyes are the fact that immersion learners spend a large amount of time interacting with other interlanguage speakers, the impossibility of creating a ‘natural sociolinguistic language acquisition setting’ in the classroom, the tendency of learners to transfer structures from their L1, and the lack of motivation to advance to higher levels of proficiency once learners become functional.” (p. 52)

            And on page 51:
            A further problem of communicative classrooms is that much of the talk which learners hear come from other learners.  This interlanguage talk may encourage fossilization, a point which Prahbu (1987) has argued forcefully.  In short, although much can be done to make a classroom communicative, the resulting environment may not be conducive to successful grammar acquisition, because the input learners receive is impoverished, because they resort to their L1, and because opportunities for certain kinds of output are limited.
           
            For someone who relies a lot on groupwork, like I do, this is a bit discouraging to read.
            But what’s the alternative?  Ellis never addresses this.

          There is, however, at one bright spot for group work.  On page 243, Ellis cites a study that says: “linguistic forms that learners of L2 French negotiated in small group-work were subsequently used independently by individual learners.  One of Donato’s main points is that whereas no individual learner initially possessed knowledge of the forms in question they were able to establish them collectively. Another is that learners can successfully acquire new knowledge through the scaffolding provided by other learners (i.e. they did not have to rely on expert others.)

************The First 35 pages******************************

            On the whole, I found this book quite interesting and readable.
            The first 35 pages, however, just about did my head in. 
            In the first 35 pages, Ellis examines the relationship between language teachers and SLA researchers.  It’s a very thorough review of the literature, in which he recounts just about every theory on what the relationship between teachers and researchers should be.  (There have, apparently, been several academic papers published in which various theories have been proposed to explain the relationship between SLA and language teaching.)
            10 or 15 pages of this would have been fine, but after 35 pages I thought all these various theories were just going round and round in circles around the same concepts, and that all of these theories were just in the air anyway.
            The fault, of course, is mine.  Ellis is not writing pleasure-reading, but an academic review of the literature, and it’s my fault for not being able to keep up.
            I only mention it in case other readers, like me, get discouraged.  If you can make it through the first 40 pages, the rest of the book gets into a lot more concrete research findings, and becomes a lot more interesting.  So it’s worth persevering through the first chapter.

**************Connections with Chomsky’s Political theories************************

            That being said, the first chapter does contain one or two interesting points.
            Here’s a quote from page 23 of the first chapter:
            Action research originates in the work of Kurt Lewin in the United States….Levin was concerned with decision-making centred around changes in practice in the work place. He was interested in what effect involving workers in the decision making process (the research) had on the factory production (the action). His approach is exemplified in his experiment on the Harwood factory in Virginia.  Lewin was able to show that when change was imposed on workers by management, production dropped substantially, that when representatives of the workers were involved in researching the change, production initially dropped but later recovered and that when all the workers participated in the decision-making, production rose markedly after only two days.  The study demonstrated the practical benefits of involving actors in decision-making. More importantly for Lewin, it demonstrated the need for and the advantages of democracy in the work place.
            I’ve often heard Chomsky say that what little research there is supports the contention that manufacturing would be more efficient if workers were in charge of production.  (For example this interview here: Every bit of evidence that exists (there isn't much) seems to show, for example, that workers' control increases efficiency.)
            I suspect this might just be the research Chomsky has been referring to.
           
            But what does any of this have to do with SLA research and language teaching?  Ellis says: “There is, of course, a dual application of Lewin’s model of action research to teaching.  One is that researchers interested in changing classroom practice need to work with teachers with a similar interest in researching change. The other is that teachers need to work with learners in negotiating the activities they will engage in. (end note page 37).
            The latter application is more relevant to my situation, and it’s a reminder to me to try and integrate my students into most of the decision making processes. (And for what it’s worth, my own anecdotal evidence confirms this.  Classroom enthusiasm seems to be much higher when I let the students vote on options instead of just imposing activities on them.)

************Direct versus Indirect Requests***************************

            Near the end of the book, Ellis describes a couple of smaller studies as an example of classroom research (one conducted by an outside observer, Ellis himself, the other by the teacher of the classroom.)  These are meant to serve as examples of how teachers can participate in research themselves either by gathering information or by testing out theories of SLA.
            And, judged solely on the criteria of examples of ways classroom or teacher research could be conducted, I think the examples serve their purpose.
            But if we look at the validity of the research itself, I have some nitpicks.
            The first example of research is a quantitative study following how two children in an ESL class learn to make requests.  Ellis notes that over time the children’s grammar becomes more sophisticated, but he also notes that in spite of their increasing grammatical sophistication both children prefer direct requests to indirect requests.  Ellis notes that this is in contrast to the native speaker norm, in which indirect requests are preferred.
            The whole time I’m reading this study, I’m thinking: “But that’s not because they’re English is unsophisticated—that’s because they’re children.  Children are always more direct than adults.”
            At the very end of the study, Ellis does acknowledge this: “This conclusion needs to be treated with caution, however…The study has …provided no baseline data from native-speaker children in a similar classroom context.” (p.195). However I still felt like he was burying this at the end of the study, when it really should have been a red flag throughout the entire discussion.

*****************Using the Past Tense to Describe a Picture*********************

            The other classroom research study dealt with 3 students who were given the task of describing a picture in order to practice the past tense. 
            Again, buried away in the notes is a little detail which should perhaps have gotten greater attention.
            Jim Lantolf reports that in his current research both native and non-native speakers found it difficult to sustain the use of the past tense when telling stories based on pictures, even when they were given a cue like “last weekend.”  (p.217 endnote 4)
            Actually, I’ve caught myself slipping up on this sometimes.  When I’m attempting to model the past tense to my students by using a picture, I often find myself slipping into the present tense despite myself.  Even though I tell the students that the picture took place in the past, after I get a few sentences into the story my mouth starts slipping into the present tense— there’s just something about describing a picture that deludes your brain into thinking the events are happening right now, because you are looking at the picture right now.
            This could mean that describing a picture is not the best way to elicit the past tense, and could screw up the results of the study right there. 
            Or maybe not.  Ellis claims that he’s had better luck using a similar task: “I found that the majority of learners when given such a cue attempted to adhere to a past-tense sequence when describing the main narrative events.  (Although it would still be interesting to know with what degree of success these attempts were achieved.)
            The focus of the past tense exercise was on teacher correction.  Or to be more precise, teacher student negotiating for meaning, because the teacher never overtly corrected the students, she just pretended not to understand when the students failed to use the past tense.  The difference?  It is extremely difficult to bring about a focus on a specific linguistic feature while at the same time maintaining true communicativeness.  Once learners realize that the task is intended to provide such a focus, they are likely to stop treating it as an opportunity to communicate and switch into a ‘learning’ mode. One way in which this can be prevented is if the focus is induced methodologically by means of requests for clarification directed at utterances containing errors in the feature that has been targeted.” (p. 216)
            Yes, but if the teacher pretends not to understand every time a learner uses the present tense instead of the past tense, isn’t that a little bit too obvious?  Wouldn’t the learners quickly realize that the exercise is meant to focus on the past tense, and still go into “learning” mode?

***********Confusing section 1***********************************

            And while I’m nitpicking, here is one more point I found confusing.
            On pages 88-89, Ellis talks about the options of teaching new grammatical structures.  There is, of course, the old fashioned way of just directly explaining to students how the grammar rule works, which is referred to as “explicit instruction.”
            Or, learners can just be exposed to examples of the new grammar, and learn the rule by themselves.
            There are two ways of doing this.  There is “input flooding” which Ellis defines as “learners are simply exposed to sentences or texts containing the target structure but nothing else is done to draw the learners’ attention to it.” (p. 88)
            And then there is “input enhancement” which Ellis defines as “efforts are made to increase the prominence of the target structure in the input.  This can be achieved by either doctoring the input itself (e.g. using bold print to highlight the structure) or by setting some task that requires learners to attend to the structure (e.g. asking questions that will lead the learners to pay careful attention the structure.)” (p.89)

            Then on page 89, Ellis cites the following study:
            Williams (1995b) compared the effects of input enhancement and explicit grammar instruction on the acquisition of two structures (participal adjectives such as ‘boring/bored’ and present passive.) In the case of input enhancement, the learners (who were enrolled in an intermediate ESL university writing class) were exposed to written texts containing an artificially increased incidence of the target forms with the target structures italicized. In the case of explicit instruction, the learners received the same input but in addition they received metalingual explanations and corrective feedback on their own attempts to use the structures.  Both the input enhancement and the explicit instruction groups did better than the control group in tests of both structures. The explicit instruction group did significantly better than the input enhancement group in learning participial adjectives, but the differences was less evident for the present passive.  One tentative conclusion of this study was that input enhancement may be effective by itself for teaching complex structures like the passive but works better in conjunction with explicit instruction for easier structures like participial phrases, a conclusion that also reflected the results obtained by Robinson (1996.)
           
            Now this really had me scratching my head.  To me, this seemed completely counter-intuitive, and I would have appreciated some more explanation.  Wouldn’t you think that it would be the easier structures that would be picked up by input enhancement alone, and that only the more difficult structures would need explicit instruction in addition to input enhancement?
            I half wonder if this isn’t just a printer’s error, especially since it’s not at all clear to me that participial phrases are easier to learn than passive structures, and furthermore the very next sentence seems to contradict this it.
            To continue the quotation. the very next sentence reads:
            These studies, then, suggest that input flooding, used by itself, may work best with structures that are relatively easy…”
            So, I think this is a contradiction, right?  Although he does switch terminology from input enhancement to input flooding, so I’m not sure if he’s using the terms interchangeably now, or what.

***************Confusing Section 2******************************

            Similar to the compliant above, there was another part of the book on page 85 that seemed to me to be contradicting itself.  When talking about the effectiveness of implicit instruction versus implicit instruction, Ellis writes:
            Hammerly (1975) found that it depended on the grammatical structure.  Explicit instruction worked best when the material to be learnt was relatively simple, but implicit instruction was more effective with complex rules.  Psychological studies (e.g. Reber 1976, N. Ellis 1993) have also found that explicit rule instruction is effective if the rule to be learnt is complex and if the grammar information is supported with examples.
            Ellis connects these two sentences with “have also”, which implies continuity between them, but don’t these two sentences contradict each other?

            On the whole, I found Ellis’s book very readable and engaging.  But the two examples above really had me scratching my head.  I re-read both passages several times, and they still don’t make any sense to me.

********************Overgeneralizing and Interlanguage**************************

            As I mentioned above, on the whole this book matched up very well with Lourdes Ortega’s Understanding Second Language Acquisition.  There was, however, one point of difference I found:
            When talking about form-focused (grammar) instruction, Ellis says:
            There is, however, some evidence to suggest that form-focused instruction can have a deleterious effect. Lightbrown (1983) found that Grade 5 learners (aged 10 to 11 years) of French in Canada over-learnt progressive –ing as a result of teaching. They overgeneralized it, using it in contexts which required the simple form of the verb, which they had used correctly before the instruction.” (p.58)
            Ortega also talks about the phenomenon of over-generalization, but Ortega says that it isn’t necessarily a bad thing, but just a stage in how the interlanguage of a learner restructures itself after acquiring a new form.  After first overgeneralizing, the learner can reshape their interlanguage and use the form more accurately in the future.
            (Although actually later in a different section of the book, on page 108, Ellis returns to the same study and said that later the overuse of the progressive –ing did in fact later decline.)

Link of the Day

1 comment:

Joel said...

Update:
It turns out that I was mischaracterizing Krashen's views in this post. Krashen is not opposed to some focus on grammar, as I discovered after reading his book.

http://joelswagman.blogspot.com/2015/03/the-natural-approach-by-stephen-d.html