Stephen van Vlack

Sookmyung Women`s University

Graduate School of TESOL

Approaches to Teaching Grammar

Fall 2006

Week 4 - Answers

Thornbury. (1999) Chapter 2 - Why teach grammar? pp. 14-28.

1. How can grammar be interpreted in the history of language teaching?

Grammar, for better or for worse, holds the central position in the history of foreign language teaching. This means that the vast majority of debate surrounding foreign/second languages has revolved around the role of grammar in both the language system itself and the acquisition of that system. This very fact puts grammar at in the central position of language and language teaching. As mentioned previously in this course, both teachers and learners give vehement lip service to their views which vary considerably as to the actual importance of grammar, but in the end their behavior, rather, indicates a firm belief in the prevalence of grammar as the overriding system among the various self-contained modules of language. The overall trend since the advent of systematized educational systems has been that grammar has remained paramount with occasional dips and rises in its perceived importance. This view may very well come from the simple fact that grammar is the only perceived aspect of language which fits into modes of institutionalized education.

            From an educational perspective, grammar teaching is in fact a matter of efficacy. Through the teaching of specific grammatical points, patterns, and rules teachers hope to help their students to learn the target language more efficiently. This is fine and, again, necessary from the aspect of efficacy, but the problem as I see it, is that grammar (because it is such a large part of language teaching) has come to be seen as the major component of language itself. Worse still it is handled as if it existed separately from all other aspects of language like lexis, morphology and pronunciation. This is strange to me, but it won`t be the first time that pedagogical perspectives have skewed the very nature of the issue they seek to explain. Thus, the development of formalized/institutionalized language education and study, which we must acknowledged goes hand in hand with the development of literacy (This is not a trifling coincidence), has had a profound effect on how the very nature of language itself is perceived. It seems then that we have really dug a huge hole for ourselves as (second/foreign language teachers, based on pedagogical concepts which probably do not necessarily reflect the reality of language, for ourselves from which it is hard to escape.

            What is interesting to think about is what people did before the establishment of national school systems and the resulting literacy. Literacy might very well be the key to the importance of grammar in more recent (the last 500 years or so) approaches. Our species, in its current form, has been around for 100,000 years (Cavalli-Sforza, 2000). In addition, humans have had access to language for between 150,000-50,000 years (ibid). This is a long time, especially when we consider that writing was invented only some 7,000 years ago (Haarmann, 1991). Once more, literacy as a general feature of human cultural existence is less than 2,000 years old. Now, the question is: what did people do before the advent of literacy to learn foreign languages? Probably, grammar was not an issue because the study of grammar is reliant on having a permanent record of the language so that pieces of the TL can be analyzed. The very concept of grammar is reliant on a literate mind to process it. Do you think these pre-literary people did not or could not learn second/foreign languages? Was there no contact between people prompting the learning of different languages? Actually there was probably more contact and more language learning going on in the past than there is today simply because there were more languages and there was more movement. Some researchers (Jane Hill (1970), for example) have claimed that pre-literate people may have certain advantages over literates as relates to language learning. Literacy, it has been claimed, (Ong 1982) actually alters the way people learn and even think. Brain organization changes or is reshuffled to make room for the new set of knowledge. So, grammar might not have been central until very recently. Grammar is a relatively new concept, it turns out, through its association to and reliance on the new technology of writing. From its connection to writing we can also see how grammar works best when limiting itself to written and not spoken forms. Grammar is of, about, and for writing.

2. What are some of the arguments in favor of grammar-focused instruction?

The arguments listed here can be placed in a continuum based on the basic underlying belief they seem to hold about the nature of the system of grammar in relation to language itself.

The sentence-machine argument

I think we all understand and are familiar with the basic crux of this argument. The idea is that grammar forms a framework into which words are inserted. Knowing the words without the framework does not guarantee that the person can use the language. It`s a simple and often-used argument, but it is also a vast oversimplification of how grammar/language works. In fact, we can say that it is a very structuralist view of language.

            One problem with this argument is that it simply doesn`t work in the real world. Not just any lexical item can be inserted into a slot designated for its particular category. For example, take a simple set of slots: N,V,N John squeezed the hedgehog. Does this sound good to you?

How about this?

(1) James was pinned down by the debris. (He can`t physically move.)

(2) James was pinned down by Mary. (He lost an an argument.)

(3) James was pinned down by gunfire. (He can but shouldn`t leave his shelter.)

(4) James was pinned down by work. (He can`t leave his desk.)

(4a) James was pinned down by sorrow. (He is too depressed to move.)

All these sentences (1-4) have the same basic structure and even make use of almost the same sequence of words but the meaning of the verb in all of them is clearly different. Hmmmmm.

How about this? John thought the hedgehog or John put the hedgehog

These are also problematic but for a different reason. Verbs have what is called valency or selection restrictions which allow them to choose only certain types or amounts of objects, but even here the (seemingly) same verb can have a variety of valency restrictions.

(5) The elephant walked (dapperly).

(6) The elephant walked its dog.

(7) The elephant walked around the yard.

(8) The elephant walked on its hind legs.

(9) The elephant walked its dog around the park.

There is lots of variety there as regards valency. Thus, the sentence-making machine reflects only the most basic aspects of language and we find that employing such a machine generally and quickly leads one into a grammatical/usage quagmire. Also, at what level are the slots awarded? How does the machine work? Modern linguistics, would argue that this sentence making machine is part of UG but we don`t need to teach UG overtly. In fact, the argument goes we probably can`t. Yikes!

The fine-tuning argument

            This is also a simple argument which seems to be wholly reliant on the argument above: knowing ways of constructing different sentences (in the form of generalized structural patterns, phrase structure rules) will allow learners to make more complex sentences and, thus, create more subtle shades of meaning. Yes, this might be true in writing in some way, but is generally not true of speaking. Structures, in and of themselves, have only a general, more functionally-based meaning at best (see Fillmore, 2003; Goldberg, 1995 for a highly theoretical discussion of this) and the potential meanings of any empty structure is virtually infinite. Lexical items are needed to reduce the ambiguity of the structure. Thus, fin-tuning is really a matter of lexis tempered by structural patterns. Looking at the fine-tuning argument from a grammar-machine perspective makes little sense in the reality of language use and creation.

            I would actually say that this argument works much better for language perception, and particularly in some types of reading where one could expect to find structures complex enough to warrant such an approach. Looking at this argument from the perspective of language perception as opposed to production we can form a link between it and Schmidt`s noticing hypothesis, a central part of the advance organizer argument.

The fossilization argument

            Yes, fossilization is a reality that must be dealt with, but this argument is totally reliant on the belief that classroom learning invariably gives, or at least offers, well-rounded, complete instruction in all aspects of the TL. Sadly, I think not. In fact, classroom instruction might actually lead to more fossilization if prescriptive grammar is the norm in the process, which it will be to different extents. In effect, as well, the fossilization argument is really one more related to feedback than actual language instruction. The thing we need to remember is that foreign language teaching is always an equation of the balance between teaching about the language (grammar) and teaching the language (actual forms in context). The more of one we do the less we can do of the other.

The advance organizer argument

            This argument only works for people who have certain learning styles and are applying certain strategies. If there is one thing that we have learned from all the research done on the area of SLA it is that second language learning is variant and slippery. It is very hard to make generalizations that apply to everyone. This might be because we have been asking the wrong questions, or because different individuals are really using different processes to learn the TL. In any case, this argument might be valid for some, but is certainly not valid for all. Don`t get me wrong, I believe Schmidt has something good here but we need to remember that advanced organization will work differently for different students. It can only be an aid to the entire language learning process. The advanced organization is only effective in the more holistic and natural follow up. If the follow up never materializes then the whole is for naught.

The discrete item argument

            Above I identified grammar as reliant on the technology of writing. Well, this argument sees it the same way. The role of technology is to make a certain task easier to perform; to render aid in doing something. Now, by breaking down language into units and assigning them roles in linguistic creation, we are using the technology of grammar to help students understand more quickly and easily. That`s the intention in any case. In reality, this is a very hard thing to do. Language by its very definition is infinite as well as ambiguous and thus cannot be described in that way, at least not in a straight forward way. The only way to do this correctly at all is to do it very generally and situationally and then we have substantially weakened the original intention. Oh no!

The rule-of-law argument

One problem - All rules are eventually broken and in language this happens quickly. Language is a slippery fish. This is similar to the argument above, but runs contrary to the nature of language in its actual use. More dangerous is that, based on this argument, we are tricked into thinking that grammar rules provide the basis of language in general, which may not be at all true. Too much emphasis on rules which may not and probably do not exist in the minds of native speakers in any substantive way may result in non-native speakers actual using vastly different cognitive mechanisms for their second language than they use, or indeed native speakers use, for their first language.

The learner expectations argument

Lame! Does the doctor ask the patient how they might like the operation to be performed? NO! Things are, however, different in the classroom. Education is an industry that is based on interaction. To a certain extent learner expectations is a valid argument, because the teacher needs to get the students to work with them. They need to get the students motivated and the students will only feel motivated if they are doing what they want or feel they are making some sort of progress in their learning (if not in their language proficiency) based on their own beliefs. In the opposite vein, the teacher is supposed to be the expert and as such can and really should make some of the decisions, even unpopular ones, in class. In short, expectations can and should be changed based on the implementation of sound theory and excellence in teaching. Teachers have the responsibility and the power to shape students and parents beliefs about language and the language learning process. Students (and their evil parents) will latch onto anything that they think might work. If you have decided to teach against the grain then you also need to be prepared to convince the students and maybe their misinformed, but well-meaning parents, that what you are doing has some validity. History is nothing more than a series of ever-changing expectations based on what has already been done. Actively engage the history of language teaching by changing expectations otherwise you will be caught in the historically vicious cycle of ineffective teaching.

3. Of the seven arguments Thornbury proposed in the book, which one do you personally think is the strongest?

For me the advance organizer argument and the discrete item argument can be seen as both going together and also as being the most effective. It should be clear from my comments above and from my selection of these two that I see focused grammar teaching as being most effective not in relation to production per say, although that is the end goal, but in reception. I have two major fears in relation to grammar focused instruction that endeavors to produce output, the first of these is that the student`s output will not be the same as that of a native speaker because it will be too tightly constrained. The second and more damning concern is that I believe grammar focused production actually coerces students into using different cognitive structures and mechanisms for their second language than they do for their first language. If we teach things like rules and sentence grids and the like and these are not things that native speakers use to produce language then the non-native speakers (our learners) are not using their brains in the same way as native speakers. This can`t be a good thing!

4. Are there any additional reasons that you can think of to support grammar-focused instruction? (Think of our Korean situation.)

This really depends on what it meant by `grammar-focused`. If it means only the sentence structure, then ̋Nő, but as we discussed last week, grammar functions on a whole variety of levels and in modern definitions encompasses most aspects of language including vocabulary, pronunciation, and fluency. Looking at grammar in this way, then we need to conclude that a study of grammar not separated at all from meaning is not only helpful, but also necessary. Again, the whole point of formal instruction is to systematize learning, thus making it more efficient. Approaching grammar from the right prospective can do that.

5. What are some of the major arguments against grammar-focused instruction?

All of these arguments focus on the idea that grammar instruction brings very little. While they may not question the role of grammar in language (the acquisition argument, the natural order argument) they all question the efficacy of spending invaluable class time (especially in a foreign language learning setting) on focused grammar instruction.

The knowledge-how argument

            The simple idea here is that knowledge of grammar is vacuous. Knowing grammar rules will not make someone a good user, especially speaker of a language. People become good users of a language by using the language in ways and situations that increase step-wise in perceived difficulty. In reality, however, learning language is infinitely more difficult than learning how to do other things such as hit a baseball or ride a bike. It might be that we need some way of organizing and recognizing elements in language due to its sheer complexity and variability.

The communication argument

            This argument is similar to the one above in that it revolves around the idea of the purpose of language being its use for communication. The argument goes that an over emphasis on grammar has a tendency to limit communication in most learners. They become too focused on their accuracy and not on the communicative purpose of the situation. Grammar reinforces too much `learning` and not enough `acquisition` (These terms are being used in a Krashian sense). Grammar also has no affect on real use and thus is a terrible waste of time. The value of classroom learning should be that it makes the process simpler and more efficient, but decontextualized learning of grammar rules is not at all efficient if the goal is communicative competence.

The acquisition argument

            This argument revolves around the belief that first language acquisition (FirLA) and second language acquisition (SLA) follow the same basic set of properties and principles. Since babies acquire their first language without too much specific or rather modular attention to grammar, then why can`t second language learners do the same? So simple. Well, the problem is that FirLA and SLA are not exactly the same process. While we need to necessarily assume that the L1 and the L2 require the same cognitive components and procedures for use this may not apply to acquisition (remember this is just a possibility. There is no conclusive body of evidence to support this view yet). From the affective side, for example, they would seem to be light-years apart.

The natural order argument

            This argument should be familiar to you who took the SMU-TESOL SLA class and is a more specific rendition of the argument directly above. It claims that all second language learners (like first language learners) acquire morphemes in the same order regardless of the order they are taught in. According to this argument, grammar teaching doesn`t affect the students` output, so teachers shouldn`t bother. Providing a minimal amount of meaningful input on a continual basis would seem to be the most important role of the second language teacher. In this way the natural innate mechanisms of the learners can be activated. The problem here is that the natural order studies have been shown to be full of holes and we are not sure how much access second language learners have to their innate mechanisms or even that such mechanisms exist in the first place. The fact that so much of this argument is based on mere conjecture makes it pretty weak in the face or various evidence. Additionally, English is a morphologically impoverished language in general so morphological plays a rather minute role in overall acquisition.

The lexical chunks argument

            This argument draws attention to the relatively new idea that language learning is essentially vocabulary leaning. By vocabulary here, I mean lexical items and, as mentioned last week, lexical items can be morphemes, words or whole phrases or sentences. Since lexical items are primary, grammar must then be secondary. In effect grammar is coded in/on the lexical chunks that people use and internalize. Focusing people`s attention on the structure of such chunks actually detracts from their internalization as complete chunks and this, in turn, makes them virtually impossible to use in the way a native speaker might. Good point!

The learner expectation argument

This is the opposite side of the learner expectation argument where in, since the advent of the communicative approach some learners only want to practice language in a setting as contextualized as we can make it in the foreign language classroom. This works the same as above - don`t let learner expectations pin you to one type of teaching. Work with the students expectations but you can also shift them. You SHOULD do so.

6. Can you think of any other arguments against grammar-focused instruction?

A lexical/psycho-linguistic argument

According to this argument, all language-specific grammar is encoded in the lexical entries of lexical items. Thus, a native-speaker-like control of a language necessarily entails a native-speaker-like system of lexis. So, for example, when a native English speaker starts to put an utterance together, as they fire the necessary lexical items, the firing of those items tells them what can go next because all possibilities are listed within the lexical entries (Lexical connections). It is a simple and very powerful idea. This is a kind of updated model of Phrase Structure Grammar

            Grammar-focused instruction without much real practice will result in grammatical information not being housed in the mental lexicon or in procedural memory but rather in the general declarative memory (with links of course to language). This results in the lack of productive capability of students taught in this approach not to mention that the basic cognitive architecture of their L2 differs from that of the L1 or the L1 speakers of the target language.

7. Looking at how the most popular methods over the past hundred years have dealt with grammar, what would you say is the general trend?

In general, in looking at most of the teaching methods over the last hundred years the role of grammar has been reduced. The greatest single reason for this is the research that has been done in FirLA from the 1960s onward. Many of the newer theories in SLA (error analysis, Interlanguage, the natural approach) are a direct result of FirLA research. The logic runs as follows: Direct grammar instruction has little to no effect on the first language acquisition process and if the second language acquisition process is the same or quite similar as first language acquisition then it should be clear that grammar focused instruction is equally unnecessary in second language learning. This basic innateness hypothesis coupled with the idea of communicative competence dealt a strong blow to grammar focused teaching starting in the 1970s , but the failure of both to pan out completely have lead to a recent resurgence of grammar as the main focus of language instruction.

8. What is the difference between focusing on form and focusing on accuracy?

Upon first reflection, these two might appear to be the same. Both of them try to get the learner to focus on their errors and/or mistakes, but the reason and how they do this is different. Focusing on accuracy is really doing so for accuracy`s sake. The reason for doing so is just to create error-free language. Focus on form, on the other hand, is focusing on mistakes in context (with consideration of meaning) so that the learner can learn from their mistakes and use them to build a more sound linguistic system. Focus on form is supposed to enhance awareness and foster consciousness-raising in context. That is the chief difference.

9. How does the E-factor work?

The E-factor revolves around balancing the two great concerns of efficiency and efficacy. This means that the teacher gives to give presentations, pick materials, and design practice that is effective, but is also simultaneously efficient. This sounds easy but is quite hard, and like everything else requires a good knowledge of the students.

10. How does the A-factor work?

The A-factor relates to appropriacy. This means that we as teachers have to gauge our teaching based on what we know about the students and the class in general, such as goals. This relates to age, level, purposes for studying, attitudes, the time of the class, the duration of the class period, their previous studying history, and many other factors including linguistic factors related to naturalistic or authentic usage patterns.