Stephen van Vlack
Sookmyung Women`s University
Graduate School of TESOL
Developing Bilingualism
Fall 2007
Week 7 - Answers
Hamers and Blanc- Chapter 6: Neuropsychological foundations of bilinguality
Franco - `Second language recovery in bilinguals` and `Paradoxical recovery of a language`
1. What is housed in (what is the purpose of) each hemisphere of the brain? (HB pp. 136-137)
When we talk about the brain in this context, for the most part we are talking about its cortical structures. The cortex is where higher level cognitive functions, present exclusively in mammals and most highly developed in human beings, are believed, in large part, to be carried out. We know that the cortex is divided into two sections or what are called hemispheres. The hemisphere that the vast majority of people use the most in relation to language is the left hemisphere. Based on this many researchers have posited that the left hemisphere is the dominant hemisphere for language. But things are not that straight forward. We also know that there is a cross over (corpus callosum) between hemispheres of the brain and the rest of the body. This means that the left hemisphere of the brain controls, for the most part, voluntary movement and sensations on the right side of the body and the right hemisphere controls the left side of the body. Now since most humans are right handed (the right side of their body is dominant over the other) this correlates well with the fact that the left hemisphere of the cortex is also dominant in relation to language. (This very strong brain-body connection is interesting when we bring language into the picture. That the left hemisphere of the cortex has been shown to be dominant for both linguistic and corporal control lends credence to the concept of embodiment which we briefly discussed in the Introduction to Linguistics class.)
What about people who are left handed? If the left side of the bodies of left-handed people is dominant, then the right hemisphere of their brain should also be dominant. As it turns out his is not generally the case. Only 30% of left handed individuals are also right brain dominant. So, it seems that the left hemisphere of the brain truly is the dominant side in relation to language and this does not seem to be affected overly much by hand preference. But does this mean much to us, probably not!
The left hemisphere is the hemisphere that has been claimed to house most of the functions and units of language. According to Bogen (1969) the left hemisphere handles information processing in an analytical, sequential, logical, propositional, and convergent manner. The right brain, on the other hand, handles information in an intuitive, divergent, holistic, parallel and appositional manner. Let`s line these up so we can see which feature compares with its counterpart.
Left Right
analytical holistic
sequential parallel
logical intuitive
propositional appositional
convergent divergent
It is hard to say if Bogen is really right in his assessment of the two hemispheres. His research was carried out more than 30 years ago, which is a very long time in brain research. Since his study we have come to understand the workings of the brain much better, but of course, not completely yet. Still, some researchers (Corballis & Morgan 1978; Corballis 1991) believe that there is not such a huge difference between the two sides of the cortex. They believe that each hemisphere basically functions in the same way, but do so at different times. This is similar to the ideas of Green (1986) and is certainly worth looking into especially when we think about the big mess we find when really trying to find differences between hemispheres. For a more recent review of this rather sticky and often misunderstood issue see: Ornstein, R. (1997). The right mind. San Diego: Harcourt Brace.
In the end it must be acknowledged that the densest band of neurons in the entire brain (the corpus callosum) works to connect the two hemispheres. The corpus callosum is really the super highway of the brain. It serves to carry more information better and faster than any other part of the brain. From an evolutionary point of view, such an important and expensive expanse of connections would only develop if it were necessary. Once more, if the corpus collosum is severed or passage across its expanse is inhibited by drugs there is a further connection between hemispheres in the cerebellum. While this is a lower level connection physically, functionally and evolutionarily it is none-the-less a connection and one which substantially pre-dates the corpus collosum. What this tells us is that the two hemispheres are designed to work together. Studies designed to try to find highlight differences are ill-informed of the basic interconnectivity of the two hemispheres.
The two main types of techniques used for assessing the working of the brain have been: A. behavioral measures and B. anatomical/physiological measures. In the behavioral measures the researcher works by observing the patient and their behavior and by giving them a barrage of various tests all of which are specifically designed to measure the workings of their brain. These tests feed different information to different hemispheres and have the subjects try to work out specific linguistic tasks. In the anatomical/physiological measures, the brain is either studied after the subject has died or is currently undergoing surgery or nowadays they have machines which can measure how the brain is working even in normal, brain-intact subjects.
It is not imperative that you know all about this, for example, what a WADA test is, but a certain amount of familiarity is necessary with the procedures and some of the methods of these different studies.
2. The confusing array of language recovery (following a trauma which causes aphasia) can best be described by Pitres` Law. What does this law claim and how does it work ? (HB pp. 140-142; Fabbro, 117)
Pitres` law is not really a law. There are no real laws, especially in relation to the brain, there are only tendencies, some stronger than other, but they are still only tendencies. What the tendency of Pitre`s law basically says is that when a polyglot receives a brain trauma which results in aphasia, then it is their most familiar language which will reemerge first. This seems to make perfect sense, especially when we consider how activation levels works in the brain. We must also realize that while Pitres` Law occurs above the chance level (Fabbro (2000) says 70%), there are still many cases where the most familiar language is NOT recovered first. There seems to be many reasons for why this idiosyncratic recovery happens. It is interesting to read about such things in medical reports and Fabbro (2000) has vast sections of his book devoted to such phenomena. For us, however, the most important thing is to try to find patterns behind selective/idiosyncratic recovery. They can give us clues on what we can do as teachers to try to make the language we are trying to get our students to learn to be more robust and enduring.
3. What are the major factors that affect selective recovery of a language in a polyglot? (HB pp. 142-146; Fabbro, 120-126)
The key word to focus on in the term selective recovery is the word selective. The use of this term implies that the patient themself plays an active role in determining which language they will recover first. Certainly this is often the case in such instances, but not in all. Some of the most salient features of selective recovery come form the patient`s language acquisition history. This is, of course, a very broad term and includes many different variables, such as how long they studied their languages, in what context(s), their feelings toward the language, and even the teaching methodology used. In reality, language recovery, as far as we can determine from the discussion in both readings, is a result of both the patient`s psycho-affective state as well as physiology. It should come as no surprise that different languages are located in different places in the brain, and this includes many different subcortical regions, in general and in different degrees. I am certainly against making too much of localizations in the brain and we know too little about the encoding process to make any definitively meaningful claims out of this, but there must be something to it. If different languages are indeed housed on different areas of the brain, as they sometimes seem to be, then there must be a reason or reasons.
Thus, if we think of a coordinate bilingual, forms from two language are supposed to be linked but linked in the brain does not mean adjacent or touching the way it would in our physical world. Synonyms or translation partners are obviously strongly linked but they do not have to be housed anywhere near each other in the brain. Remember this.
One recently attested location difference (which is also mentioned in Fabbro (2000)) claims that some languages are seated deeper in the brain. They have more elements housed and linked more in the mid brain and even the brain stem than in the cortex. This is an extremely interesting idea and one which has very important repercussions in the field of language learning and multilingualism. A detailed discussion of these interesting issues is not province of this course and will be taken up in greater depth in the Human Learning and Cognition class.
4. What are the three main bilingual-specific behaviors and what can we find out about them by looking at polyglot aphasia? (HB pp. 146-147; Fabbro,127-133 )
The three main bilingual-specific behaviors, according to Grosjean are translation, code-switching and code-mixing. Grosjean calls these bilingual-specific because they are not used by monolinguals, but I think he is wrong. Monolinguals may or may not engage in such behavior on a regular basis depending on the types of situations they are put and how much they know about their own language. When a monolingual register shifts, this is the same as a code switch but on a slightly different level which is so subtle one might not be able to always perceive of the switch. Monolinguals who are able to use many different synonyms of a word or synonyms of forms for a single function could be said to be using a process akin to translation. So, I do not think these are skills limited to bilinguals although a bilingual would need to do them on a slightly different level than a monolingual. Once more, because easily identifiable codes are involved, the process is readily visible in bi- and multi-linguals while it may be invisible or more opaque in monolingual language usage where one basic code is applied.
In any case, the evidence from polyglot aphasics is very interesting here. It shows that these special skills (behaviors) may be housed separately in the brain. That means they are housed apart from other (monolingual) language functions. The evidence for this is quite numerous. In one case, for example, an English-speaking professor of classics incurred a head injury which deprived him direct access to English. He could use English only after he had first phrased the intended utterance in Latin. What he was doing was translating from Latin to English, a skill he was very well versed in due to his long-term position as a Latin teacher. Eventually he did regain English but only due to a concerted series of increasingly difficult translation activities (sad and funny). In other cases, aphasics have been observed to code-switch and code-mix extensively between two languages, one of which they could not form single utterances in. A woman, speaker of German (L1) and French (L2), could not speak German after she woke up after her injury. When addressed in German, she spoke only French and claimed that she knew little German, but when speaking at length she was observed to code-switch regularly between French and German.
This is important because it provides us our first real evidence that language is composed of different and, in a way, conflicting skills.
Albert and Obler make a lot of nice snappy conclusions about polyglot aphasia. For us as English language teachers trying to instill a degree of bilinguality into our students, we need to think about how this information can be used to guide us in a more effective way of teaching. Based on this, one of the most important conclusions we can find is that language recovery is based on affective factors. Simply put, this means that polyglots initially recover the language that they like the most, that has the most pleasant associations, regardless of their level of proficiency in that language. Now, it might seem like a lot of this (getting the students to form pleasant associations with our TL) is out of the teachers` control and it might be, but there is also a lot that is in your control. The basic idea is that if any of your students wind up getting whacked in the head someday you want them to be speaking English when then wake up. Well, how can we get our students to love English, to form pleasant associations with it? This means that you need to go out of your way to give your students meaningful input. Give them the tools they need to feel confident enough to use the language and be successful in it. They need to be able to experience the joys of life (not exclusively the pains) through English.
Another important thing that Albert and Obler mention here is that languages learned as children are able to be recovered through hypnosis. This means that we never actually forget what we learned or even were exposed to passively. We simply forget where we have placed all that scattered information in the maze of our brain if the information goes largely unprocessed. This is both good and bad news for teachers. It means that our students really do remember what we teach them (expose them to), but it also means that we have to be very careful about how we present information. Information (and by information here I mean TL language input, NOT erroneous facts about the TL) must be presented in such a way that it is easy to be remembered. Link it to real life and real uses. Make it emotionally directly relevant for the students Try your best to get them to use this information for real purposes. These two main effects are complimentary. By increasing the one, you are increasing the other.
There does not seem to be any concrete evidence for lateralization differences in monolinguals and bilinguals despite the fact that people have tried very hard to find them. Bilinguals seem to be just as left brain dominant as monolinguals.
There is a wealth of evidence which clearly shows that both languages of a bilingual are housed in the same hemisphere. The only possible exception to this is a speaker who has minimal competence in the L2, and this would be a temporary developmental difference. As their competence increases more of their processing in the L2 will be in the left hemisphere. All the evidence supports left hemisphere dominance in relation to language. Think about why this might be!!!! It has something to do with language processing!
5. How does the bilingual experience effect the onset of lateralization? (HB p. 151)
Early bilingual experience seems to speed up and even increase the effects of lateralization. It is, however, very hard to say what kind of difference this really makes. It must also be noted that an early onset of literacy (conceptual thinking) also causes lateralization to occur earlier. Since it`s first mention by Lenneberg all the way back in 1967, lateralization has been used to explain all sorts of differences between child and adult learning, particularly in the area of language, but is anyone really sure of its effects or even if its reality. In more recent publications (Thelen 1989), we see that the effects of lateralization are a bit sketchy. Since everything in the brain is deeply interconnected, what great difference would lateralization have? That is the question we have to answer. More evidence that refutes Lenneberg`s version of lateralization is that adults (people way past the stage of full lateralization) who have incurred injury to a part of the brain after an initial period of loss are usually able to recover most, if not all, of the lost material. Lenneberg predicted this would not be able to happen. In the same vein, when children (who have not completed lateralization yet) incur the same type of injury they go through the same type of recovery as adults. Such evidence impels us to call into doubt the major premise of what lateralization and its supposed importance in providing biological evidence for the CPH (which the evidence from bilinguals refutes anyway).
6. How does the age of acquisition effect cerebral strategies? HB (pp. 151-153)
As has been pointed out in the book, children (early bilinguals) seem to use their left hemisphere more. Adults (late bilinguals), in turn use the right hemisphere more. This of course, is related not only to how they process the language but the environment in which the language is learned. Children often learn in a much more natural environment, even if it is in a classroom (see number 9 below). Again we have to ask ourselves what the big deal is. Is the right brain bad? Is it unnatural? Some people would say yes, others would say that it is important to try to activate both sides of the brain and set out to use activities and processing which does both.
7. How does the relative amount of competence in the bilingual`s L2 effect hemispheric uses? (HB pp. 153-154)
We know from the theories of Krashen (1982) and especially form McLaughlin (1990) and Bialystock (1990) that there are differences between learned (controlled) and acquired (automatic) material. This can directly effect the use of different hemispheres. It has been posited that learned material (controlled) processing is related to right brain functions while acquired material (automatic) processing is associated with left brain functioning. We also know that, especially adults, move from controlled to automatic processing. In this way we can equate proficiency level with cerebral preference. Adults in the beginning stages will use a lot more right brain processing. This is supported by evidence from timing experiments. Bialystok conducted a series of timing experiments and concluded that the main difference between controlled and automatic processing is the time and ease it takes to retrieve the material. If the controlled material is stored in the right hemisphere and processing is in Broca`s area which is in the left hemisphere then we have a clear explanation why it takes longer to achieve such material.
8. How might learning context effect hemispheric preferences? (HB pp. 154-155)
By context, we mean the location of the learning. There are two main contextual situations for language learning; in the classroom or out in the real world. Based on what we know about the generalities of right and left brain processing it is not hard to imagine what has been posited in relation to environment and processing. Generally speaking, students who learn a language formally in the classroom tend to use their right hemisphere more than those who learn their language naturally in the real world. But we need to remain skeptical as sharply conflicting results have been reported. One of the reasons for this would be the methodologies that are used to teach language in the classroom vary a lot and are based on different cerebral processes. In addition, different learners have different preferences as to how they like to use and think about language regardless of the environment of language use.
In the end, because the right and left hemispheres are so closely connected via the corpus callosum (which acts as a bridge between them), it is hard to say if there is a such a huge difference between right and left processing.
9. How do differences in scripts relate to hemispheric preferences? (HB pp. 155-156)
We know that there are two kinds of scripts (writing systems): logographic (Chinese) and phonemic (Korean). Evidence from studies shows that logographic systems like Chinese involve more right hemisphere interaction, while phonemic systems do not include any effects from the right brain. One reason for this might be due to the memorization effect that is necessary in reading Chinese characters (Liu 1995). In any case it is simply an interesting aside for us to note the differences between such systems because the effects on literacy are not small.
Kecskes and Papp (2000) Chapter 6
1. What is language distance?
There are two, possibly three, types of language distance depending on how much credence you give to Kellerman`s hypotheses of perceived linguistic distance.
Linguistic/Typological distance
Linguistic or typological distance refers to the actual structural differences between two languages. For the most part this has been mostly studied at the level of syntax simply because syntax has been the dominant area in linguistics for the last 50 years. It is also the area of linguistics which is easiest to study because it is quite concrete. When looking typologically at different languages we can describe them basically according to way in which grammaticality is encoded. Some languages like English or what are called configurational languages. In these configure a languages grammaticality is encoded in the word order. Such languages are said to have grammatical word order. In such languages, like English, even though we change the order of the sentences we have to do so under highly constrained conditions. Other languages have what is called pragmatic word order. These languages are not configurational in that the different elements in a sentence can come in any possible word order. This is possible because in these pragmatic word order languages the different elements are morphologically marked for their functions. Thus, the sentence `the farmer plows` can be rendered in Latin either as:
Agricola arat or Arat agricola with no meaning difference. It should be noted that there is a direct correlation between how much morphology a language might have and whether that language is going to be configurational or freer in its word order. Thus, languages like English which has little inflectional morphology and Chinese which has no inflectional morphology are going to have to be configurational languages, while languages like Finnish, Russian, and Latin which all have a large amount of inflectional morphology are more pragmatic based in how they use word order.
Perceived Linguistic/Typological distance
It is important not to under estimate how people`s perceptions of language are important. Reality is really just how we perceive it and if we are truly convinced of something as being real it`s just as real regardless of how true it may actually be. All learners of a language, even those with just a passive or passing knowledge of that language, have strong beliefs/perceptions of how that language works and how different it might be from their own language(s). Everyone goes into language learning with some idea of how hard they think the process might be. These preconceived and often faulty notions have a huge effect on the learning process. They affect not only how much a learner might try with the language but will also affect the different strategies they will use in trying to learn the language. We know from our discussions in this class that different learning strategies will actually lead to different types of organization in the brain (remember coordinate versus compound types of organization). So this perception is extremely important.
Social distance
Aside from just the actual structural forms of the language learners also think about the social distance between their own language or languages and the new language which are there trying to learn. Looking at this we cannot separate a language from its actual speakers. The impression the actual speakers of a language make on the learners is essential in determining how much energy and what type of a learner might invest in the learning process. So, it is about motivation. Some people learn languages because they are more similar and familiar while others are enticed by the exotic. Aside form the question of motivation there is the reality of cultural differences. Languages with a more similar cultural base will be easier to learn because there will be more similar concepts which may be easier to eventually co-join in a CUCB. Languages which a much more dissimilar cultural base will have more dissimilar concepts which may not be able to be co-joined in a CUCB. Remember he development of a CUCB is a long and arduous process which requires intensive exposure to the target language (TL). More cultural differences will make this process that much harder.
2. What are some of the different aspects of language distance?
Here we just want to reiterate some of the things mentioned above with a deeper look at some of the linguistic differences of different languages. We can basically say there are three and one of these was discussed at some length above.
Configurational nature of the language = Morphological systems = Word order
Here, again, we are looking at whether a language has grammatical word order or pragmatic word order. As we saw in class with the Turkish examples agglutinating languages like Turkish are more often than not non-configurational because they have a complex system of morphemes which mark words for their specific functions. The functions, then are not marked by the word`s place in the structure as we find in configurational languages like Korean, Chinese, Japanese, or English.
Lexical differences
Differences between the word store two languages might have, regardless of the underlying concepts from which they are derived, have a huge affect on the learning outcomes. Obviously, languages that share a large number of word forms (cognates), such as English and French, Korean and Japanese, will be easier to learn than languages which do not share such forms.
Metaphorical distance
This is similar to what was discussed above in relation to concepts and cultural differences, but focuses more on Lakoff`s (1987) idea of idealized cultural models (ICMs). The basic idea behind ICMs is that cultures take certain metaphorical structures (anger = heat, for example) and build not only language but thought structures out of these. Native speakers of English use hundreds of expressions related to this one ICM on a regular basis. These ICMs reflect the way people think and they vary from language to language, culture to culture. It is also true that ICMs will be more similar in languages which rely on a more similar cultural base. So metaphorical distance really underlies (in a hidden way) so much of what must be learned in a new language.
3. How might language distance affect language learning in multilinguals?
In a very general sense, languages that are perceived as being more similar whether linguistically or socially (and better yet both) will be more easily learned than languages that are not seen as being so close, but there are grammatical variables as well. It is, generally speaking, easier to move from a more marked (GWO) to a less marked (PWO) grammatical system. Then of course there is also the metaphorical structure of the two languages and the learner`s level in being able to use these new metaphorical structures is very much reliant on their overall proficiency in the L2. In the previous week we talked about transfer and how transfer was strongly affected by perceived similarities between languages. People who believe that the languages they are learning are more similar will transfer skills much more easily than people who believe that the two languages are more different. If your think about the English learning situation in Korea we know that the schooling is all based on spotting differences. Hence, no transfer and no happiness.