Stephen van Vlack
Sookmyung Women`s University
Graduate School of TESOL
Second Language Learning Theories
Spring 2007
Week 13 - Answers
Johnson, Chapter 9: Building a new model of second language acquisition
1. Why is such a new model, the dialogical model, needed, according to Johnson?
According to Johnson we need this new model because there seems to be a giant rift between the current prevailing theoretical models of second-language learning and the reality of second-language learning. This rift, or contradiction even, should be obvious to us when we consider that one of the salient features of SLA as a process is its failure. This alone should tell us that something is seriously wrong.
For Johnson, the reality of second-language learning is the context in which the language is learned. She feels strongly that current models of second-language learning focus has generally continued to be simply on internal states when in reality there needs to be much more focus on the external state of the situation as a way of affecting or determining, to a large extent, the internal state. None of this should come as great surprise to us because this has been her rallying call from the very beginning of the book. It also seems to make perfect sense if we factor in just a fraction of what we know about human cognitive systems and the role of perception in building those systems.
It should, though, be understood that in trying to heal this rift between internal states and external states her model is not going to form a nice clean link such as we investigated somewhat earlier with things like information processing models. The bottom line is that the dialogical model which she puts forth is not compatible with the current models. In this way she's taking a similar stance as the models which she hopes her model will someday supersede. She's on the far end of the nature and nurture spectrum. For her it is obvious that language learning is predominantly nurture based. She doesn't really mention anything about the actual cognitive systems which might be needed to support the system she advocates, but she mentions, following Vygotsky and Bakhtin, a bot of what they need to learn and a little of how. So, in essence this new model she is advocating is a bit incomplete. There is no trait model to back her up, or is there?
2. What is the crux of the new model?
Her new dialogical model is basically a combination of Vygotsky's sociocultural model and Bakhtin's model of heterodiglossia. Again there is nothing striking new or shocking here because this is certainly the direction that she has been moving in throughout the book. In describing the model itself she doesn't really integrate any new information. All she does is explain again the necessity of having such a model and what such a model might mean to the field of SLA. In a way, she brings it all together, but introduces nothing new
3. How does she suggest we teach the new model?
It is in this section of the chapter where she gives us the most pertinent information about her new model. It is also important to point out that some serious questions arise about not only the efficacy but also the necessity of some of the teaching which she advocates in relation to her new model. Principally she claims that in or to teach this model we have to try to create a sociocultural environment in the classroom which mimics that of the target language or that of the community outside in which the target language is to be used. In reality it is not certain that this is applicable to most subsequent language learning situations and might not help us too much in Korea where English is not used for a wide variety of functions in Korean society. English primarily exists as a haze embedded within a strikingly dominant Korean matrix. Johnson seems to be entirely fixated on, and therefore limited to, second (not foreign) language learning and the idea that subsequent language learners need to become more like native speakers of the TL or need to move towards speaking in the target language culture. For a global language like English I am not sure if this is possible or even desirable, at least not for us here on the fringes of the English speaking world. But she also has some valid points in this regard.
An assessment of the local environment regarding what the students might be expected to do with English is a good idea and is something which local language researchers need to become deeply engaged in if we are to ever teach our students appropriate language functions. A complication in Korea, and not only Korea, is that the external situation is changing rapidly. In the period from which a child starts English education in Korea to the time she finishes the use of English in Korean society changes quite a bit. This means that we as teachers need to be sociologists as well as experts in language learning and teaching. This is no mean feat.
Johnson does give some interesting and quite valid ideas about trying to get the students to develop their voice through English. This correlates with a lot of bilingual or multilingual research (see Grosjean 1982, Nicol 2001) about how to such individuals actually store and use their languages, so she's on the mark here. Research has shown that bilinguals are not the same as monolinguals. Rather than simply have two sides to them which reflect their two languages what they wind up actually becoming is in integration of these different systems. This is a very interesting area of research and I think it's good that she mentions this because it is not mentioned by people in standard second language learning models, in fact it's impossible in a generative-based model. How this is going to happen is going to be little more difficult, especially in a place like Korea. But we're supposed to create a classroom environment which mimics the target language social environment then, for English, which social environment do we pick. Our students are young and we have no idea where the future will take them so it is hard to decide on what to do to build them for the future. This is a bit of a problem particularly for people who may never actually enter a native speaker social environment as most of our students won`t or a social environment solely with native speakers of the target language. We should be aware, as Johnson does not seem to be, that the number of non-native speakers of English is many time greater tan the number of native speakers and which the latter is shrinking the former is increasing rapidly.
The idea of genres fits in nicely with an ESP approach and I think that it can complement such approaches very well so that's one other interesting area which she advocates in relation to teaching.
4. How does she suggest we test the new model?
At first glance trying to test her new model might seem like a very difficult task indeed, but with modern technology in might not be nearly as hard as it may seem. The bottom line behind her testing paradigm is that we need interaction in particular and possibly unrelated instances which have been set up carefully according to specific concerns revolving around the test-taker`s needs. Is interesting to see how the MATE actually falls, or rather has the potential to fall, fairly closely into her testing paradigm, with of course the interaction missing, but with some inventive computer programming I think we could compensate for some of these problems. In short, I think that her way of testing actually shows a lot of promise and is certainly not beyond the realm of possibility even with today's technology.
5. What are some of the positive and negative attributes of this model?
There are both some strong positive and also some rather negative attributes (well, attributes hard to realize) of the dialogical model which Johnson (2004) proposes. Starting first with the positive attributes, what we can say is that many of the basic premises of this model seem to be, based on lots of different recent evidence, actually correct. Really, the dialogical model does seem to more closely adhere to how people actually learn and not just language. This is obviously an immensely strong attribute. In addition, as we saw in class it seems that we can successfully link this dialogical model with other pre-existing models which cover different areas. These other models would be connectionism which would function as a basic explanation of how learning would and could actually occur in the brain as well as cognitive linguistics which would function as a theory of state which, as we have mentioned, is sorely lacking in this dialogical model. Once we bring all three of these together we have a model which is, in reality, much stronger than the current models which are not very well integrated and often contradict each other on small and large issues. In addition another strong attribute is, I believe, the testing model which she proposes. Looking at current models of second language acquisition we realize that this elusive idea of competence is extremely hard to test. And even if we believe that we have a test which can measure our students competence, this test often is not able to assess performance. Looking at a test model which Johnson proposes we would be able to have tests which cover both competence and performance, particularly as these are not distinct in her model.
Having pointed to attributes, and some very strong positive attributes at that, it must be acknowledged that there are some negative attributes to Johnson`s model. The first thing that critics would mention about this is the lack of formalism in relation to language about the model. As is mentioned above, though, by integrating connectionism and cognitive linguistics models into this dialogical model we are able to make a very strong model and critics would hopefully be stifled by this. On a different level I think there's one major problem with this model and that is the teaching of the model. This is not because students necessarily have different expectations about what they need to learn and what they want to learn. It is based on the fact that learning and social environments is really based on need. Our students have no idea how they need English.
We have a major problem here in Korea and in most of the world where people are trying to learn English; needs are not clearly defined or adhered to in teaching. It is clear that neither students nor their teachers know exactly why they need English. A quick survey of Korean society would show that few people indeed need English for much of anything. We certainly don't need it to function in Korean society. The only need it for is when we are engaging outsiders, of which there are few in Korea. Yes, certainly people need to engage the outside world, but for a rather limited range of purposes. Yet we teach general English with the vague goal that we want our students to be as close to native speakers as possible. This seems not only strange and counterproductive but also rather impossible under the circumstances. Learning, and language learning is no different, is about linking different sets of information. In a naturalistic learning situation people do this through social interaction and social discourse. Interacting socially is, however, very hard.
It takes people their entire life to learn how to interact in different ways for different purposes. Interaction is a dynamic thing which keeps on changing depending on age and little refection is given to this beyond elementary school. How people interact, with whom and for what purposes changes dramatically across a lifetime and this works differently for people of differing socio-economic status. Once more, people only expel the energy to interact efficiently when they have a need, either intrinsic or extrinsic. A lost of what interaction comes down to in the end is the quest for useful information. In Korea this becomes a particular problem because Korea is a highly developed information based society. Koreans have been very particular and even more successful about developing strong information power bases in Korean. Any relevant information form English is quickly and irrevocably integrated into Korean through Korean cognitive processes. Unlike many other societies where people must use English to acquire a lot of necessary information, Koreans can interact with and claim information through Korean. This means that Koreans really don't have a great need for English. In a situation where somebody can use either their L1 with their L2 as a means of extracting necessary information, then is obvious that the L1 will be used over the L2. This is natural because it generally takes much less effort. This is a major problem for a model of teaching which focuses on building a society in the classroom when there is a stronger, better, and much more useful society right outside the door. This is one of the main differences in teaching English in a ESL and a EFL environment and of course Korea, as one of the few monolingual countries in the world, really is an rather extreme example of a very strong EFL environment.
It seems, then, that we would need to rethink a few of Johnson's teaching ideas, which are overly simplistic and centered on ESL, to see how we could try to get them to work better in an EFL environment and particularly one like Korea where the information we need to function in the world is readily available in Korean.
References
Grosjean, F. (1982). Life with two languages. Cambridge, MA.: Harvard University Press.
Nicol, J. (Ed.) (2001). One mind, two languages. Oxford: Blackwell.