Stephen van Vlack

Sookmyung Women`s University

Division of English Language and Literature

English in the Era of Globalization

Fall 2007

Week 6 - Answers

Kecskes and Papp 2000, Chapter 2

1. How do the authors propose the L2 to affect the L1?

Rather than proposing that foreign language exposure might have an effect on L1 forms themselves, as we know is often the case in exposure from L1 to L2 or FL, Kecskes and Papp propose that a FL might affect cognition and as a result of such affects will impact on how the L1 is used in cognitively demanding situations like reading and especially writing/composing. It is also important to understand that in this proposal language and cognition are indelibly intertwined. One exists with the other. The basic idea is that language use is controlled by cognition or conceptual structures. This is the basis of all language use and to be able to use language effectively we need to be able to control more and more varied conceptual structures. In his proposal they are claiming that learning a foreign language allows us to awaken, if not exactly learn, certain conceptual structures which have lay dormant and that these conceptual structures will then be able to heighten elements of language use. In this sense Kecskes and Papp are being extremely cautious. It is highly possible that a second or foreign language might affect a first language in a variety of ways and certainly on cognition but not only on cognition but on meaning. This is an idea we will return to later in this course. To understand exactly the hypothesis of the current research we need to first understand the nature of written speech and how it might vary from spoken language.

2. What is written speech and how is it developed and how does it work?

Written speech is basically a fancy expression for writing which is used to highlight the primacy of speech and the fact that written forms are developed from spoken forms. In this view, writing differs from speaking in that it is not reactive, it is, at worst, a delayed reaction and, at best, a highly planned and carefully constructed instance of language use. The chief difference between speaking and writing, then, is seen as the level of planning, the level of control and the thought which goes into, or really should go into, each written utterance. Written speech varies not only in the way the forms are put together, but also in the forms themselves. It exists, therefore, on two different levels: the planning level and the form level. It is important to recognize that these two levels are not mutually exclusive, they are constantly interacting with each other. In basic neurocognitive theory, one skill facilitates another related one based on processing. Thus, the better one is able to plan (this comes as a product of practice), the more complex their utterances are going to be. Likewise, the more one is able to put together complex utterances with ease, the greater they are going to be able plan their utterances. It is widely acknowledged in literacy circles that the quality of the forms used and planning do actually go together.

            In a regular naturalistic language learning cycle, written speech is developed as an abstraction from spoken forms. We take the basic ideas of the spoken forms (the underlying conceptual structures) and extend them from the concrete to the abstract realm. One of these chief methods of extension is through fusion and linking. We know that written speech is much more efficient than speaking in that more meaning is contained within a shorter span of language (a smaller number of words) because more complex grammatical/conceptual connections can exist between the different units. By fusing concepts and linking these concepts linguistically we are able to create much more complex ideas in a shorter linguistic space in the written code. Maybe a simple example will make this clearer.

(1) Unbelievable though it may seem because of his sweet demeanor, Harry is the man who did kill and eat my dog at my recent fancy dress ball.

(2) I had a fancy dress ball. It was not very long ago that I had this fancy dress ball. To this fancy dress ball I invited a man named Harry. I liked Harry. I thought that Harry was a man of sweet demeanor. We all thought Harry was such a sweet man. He wasn`t. At the fancy dress ball he killed my dog. After that he also ate my dog. I think it is unbelievable.

            Looking at these examples above we can contrast the efficacy of (1) against the strange, ineffectual ramblings of (2). We can see how by carefully planning at both levels our utterances, we can write (create written forms) of a much more effective nature much more effectively than we can actually speak. But to do so requires more cognitive control. To be a good writer of a language one needs to have control over many different formal elements of that language. This is basically the hypothesis or proposal that Kecskes and Papp are making for this study. They are claiming, and in actually a very conservative way, that learning a foreign language helps us master or relearn some of the formal elements of our native language. This comes about, they claim, because foreign languages are taught in a formal way.

            When we learn a foreign language we don`t really learn how to speak directly and we`re really not generally exposed to spoken forms of that language. Rather, we are really introduced to written speech, albeit in the mode of actually speaking. When we study grammar formally in a foreign language situation we are not really learning the grammar of spoken language (which varies dramatically from that of written forms as you saw from the simple example above) but we are really studying the grammar of written speech. This is why this transfer seems to occur.

3. Briefly outline the design of the experiment.

The experiment has four different variables in its design.

            First is the different situations in which the subjects are placed - immersion, intensive and control. While we are all aware that immersion programs have been touted as effective situations for second or foreign language learning it might at first seem very strange that an immersion situation would result in better usage patterns in the first language. We make this assumption because, as is also present in this study, the subjects are actually exposed to less of their first language in an immersion situation and we would normally assume, in our teacher-fronted world, less instruction to mean less learning. We might therefore have initially hypothesized that the immersion group subjects would actually lose certain skills in their native language, in this case Hungarian while making the most improvement in the TL (French). We would possibly have initially hypothesized the best effects for the intensive group because not only are they getting a large amount of input and instruction in their first language but they are also getting a quite large or as the name implies fairly intensive amount of instruction and input in the foreign language, in this case either Russian or English. Control groups are always control groups we expect little to happen to them.

            Second is the timing of the different tests - X, nine months later, 12 months later

The timing here is the simple type of timing that we would find in a standard longitudinal type study. In such studies, there needs to be a certain amount of time between testing instances for development to occur and be measured. If there is no timing between the different tests then there might not be any measurable differences simply because the subjects have not had enough time to acquire any new skills in any measurable way.

            Third is the tasks - familiar essay, unfamiliar essay, picture-based story FL + L1, advertisement. The different tasks here should be familiar to most of you because many of you have experienced them in your GEP course or on the MATE test itself, with the exception of the final test, responding to the advertisement. This last task is a little bit strange and the authors themselves acknowledge this fact. It would seem that the final task would require or, under normal circumstances, call for a more informal type of writing and that most writers would not include very many subordinate clauses or sophisticated grammar in such a task. I`m wondering why this was put as the final task, because at this point we would assume that we would want a task that would call for a possibly large number of complex structures. This could be a certain flaw in their design, or at least it makes the results a little bit harder to interpret.

            Fourth is the evaluation of the tasks - Frequency Index, Unusual Subordinations, Loban Index, Loban number. What all this basically boils down to, fancy and complicated though it may seem, is the idea that complex language requires more cognitive control (planning and awareness of the language). Based on this, all these different measurement tools are designed to evaluate different types of complexity in the forms used by the subjects on the different tasks. Each one has a slightly different focus so that the researchers can see which different aspects of complex structure seemed to be easier or less easy for different subjects to produce on different tasks. As we might assume, the results of differently designed tasks will result in the production of different types of complexity from the subjects. It is hard to say at this point if the experiment design is possibly a bit overly complicated, but, in the end, the results do emerge and rather clearly.

4. What are the results of the experiment?

The results showed that exposure to a FL does seem to affect usage in the L1, even where the students are getting a smaller amount of formal instruction in the L1. It seems that FL exposure activates passive knowledge in the L1, which is beneficial in writing. This is pretty much what they expected to find and in fact they do find it, but of course a certain amount of exposure is required for these results to occur. The control group did not have these kind of affects so, as we find with general bilingual development, a threshold level of exposure seems to be required for positive aspects of subsequent language instruction/learning to come into play. These results also point to the specific notion that the different languages a person might have can be interrelated. This is a very important conclusion and is one which will be discussed at great length more next week.

5. How would you interpret the results? Does your interpretation vary from that of the authors?

There are many different ways of actually interpreting these results. One of the key issues in this study is the age of the participants. The success of the study is really related to the participants themselves. If this study had been done with people who were three years younger or three years older the results might not have been the same, or at least not quite as uniform as they were (they were using rather small numbers in each group). Younger subjects who had not yet completed some of their basic cognitive development might have actually shown some negative effects on their Hungarian as a result all the increased exposure to other languages and diminished exposure to their own first language as in the immersion group. Older learners who had, to a certain extent, already mastered some of the specific cognitive elements related to composition might not show strong results because they don`t need to learn those aspects. So the results are not quite as simple as they might appear. They are however compelling in that we can see that a well developed and carefully constructed bilingual program, or a program of intense foreign language instruction, provided at the right time in a learner`s development will have beneficial effects on not just one language system but on all the different language systems contained in the learners. This is really nice for you as people who are language learners and bi-/multi-linguals. Again, this is an idea we will return to later in this course.