Stephen van Vlack

Sookmyung Women`s University

Graduate School of TESOL

Approaches to English Grammar

Fall 2006

Week 10; Answers

Thornbury (1999) Chapter 7: How to deal with grammar errors, pp. 113-127.

1. What are some of the major areas that errors can occur in?

Errors can and do occur in all areas of language, from syntax and morphology to lexis and pronunciation as well as other areas. Thornbury identifies four specific areas in which he claims errors occur. They are lexis, grammar, discourse, and pronunciation (for spoken language). For us in discourse it should be clear that all of these relate to a larger concept of grammar. Lexicalists, of course, claim that lexical errors are by far the most disturbing of the four, for if a different word or expression is used it often confuses the listener much more than just mixing up an article or a mispronounced phoneme. Of course it is all a matter of degree. And when we are focusing on grammar as a communicative endeavor then it should be understood that there are major plateaus of achievement. The first of these is simple intelligibility. It should certainly be noted that to be intelligible a speaker must not commit too many errors or too many errors of the damning kind. This brings to the forefront the idea of errors as an indicator of our students level. Major production focused proficiency exams (like the MATE) all use error analysis as one way of trying to determine the test taker's level. In the beginning of discourse as well we distinguish different levels which could be used to describe the learners proficiency in a specific grammatical point. Again different levels are distinguished through a large extent by the learner's instances of error vs. instances which occurred error-free or with an acceptable degree of comprehensibility (what they can do with language with or without errors). From this point of view it should be clear that errors are good.

Errors are good for a few basic reasons. The first reason errors are good is that errors tell us that we are allowing our students to produce enough, well at least some, language in the class. Here we need to make a distinction between errors produced in real-time as in a more natural type of speaking or writing exercise and ones produced out of real-time, as in a simple form feeling exercise where students have time to choose specific forms. Being able to accurately fill out a form out of real time contraints does not mean that students can actually produce the same forms accurately in real-time. If we want to get our students to speak and write, that is produce language, with some degree of accuracy then we need to give them practice in real-time and we need to evaluate them in real-time. So if our students are producing errors in the class it means that we are enabling them to practice and hopefully more in real-time. Without these errors we have only a vague idea how well our students are actually doing. Also, based on the idea of a self-monitoring system, it is extremely important for students to be given practice in real-time so that can produce errors and so they can learn to hear their own errors, with the whole of that this self-monitoring system will get them to start to fix their errors. Self correction is an extremely welcome and also extremely necessary aspect of linguistic development. When students do not self-correct in the face of a dire need for such correction then it can be anticipated that the student's language system will go still and stop developing if it has not already done so. Therefore, practice and errors need to be seen as being closely co-related.

We can make a basic distinction between form based errors in areas such as were listed above and usage based errors. When we teach, and especially when we teach grammar, it is the clear form errors that we generally pay much closer attention to. Form errors are, in practice, easier to recognize because they are not opaque. We teach forms in sets of rules and we tend to teach them separate from usage, thus our attention is drawn to them. For both teachers and learners, it is the usage errors which are generally harder to pin down.

The four way distinction Thornbury (1999) gives us (lexical, grammatical, discourse, pronunciation) is a in a way too simplistic as all of them can be both form or usage based errors. The basic distinction between the two is that form errors relate to things that native speakers could not say (forms native speakers cannot use) while usage errors relate to things native speakers would not say (forms they generally would not use). These forms can relate to concerns that seemingly go beyond mere language forms and can relate to cognition or way the world is perceived as in the following usage-based error.

(1) A large foot traveling at a tremendous velocity managed to make impact on the poor armadillo`s rump.

(2) A snake lay under him as Drake threw his body through the air over it.

There are no form errors in (1) or (2) but they are decidedly strange. These are extreme examples of usage-based errors. Additionally, there are usage errors based on lack of knowledge or skill which often lead to wordiness as in the examples below.

(3) It was strange to see the porcupine wearing those things humans usually wear to cover up the bottom half of their bodies.

(4) The car went down the street at 120 mph.

These are things that are hard to identify exactly and are even harder to correct, but we need to do so because these are just as bad as form based errors and sometimes worse as far as comprehensibility goes. As mentioned both above and throughout this course maybe the best way of dealing with usage based errors is through continued and increased practice. This practice not only in producing the language but also important practice in perceiving the language. Perception is the mother of invention. You can't write until you read, you can't speak until you can listen. There's no point in any language user's life when they can afford to stop listening and reading. This is how people are exposed to usage. Usage is not something that is easy for the language teacher to teach. It takes time and effort and a huge amount of exposure (input). As far as the sequence goes, it would seem advisable to work on forms first before usages, but this does not mean that we try to teach every single form before we introduce the necessity of usage. It simply means that form should be one step ahead of usage in the never ending to developing linguistic skill. Ibn reality it will be impossible to distinguish these two if the students are creating real language and we probably should not do so.

2. What are some of the major causes of errors?

Thornbury designates two major sources of errors: transfer and developmental errors and I think you are all well versed with both of these. It should be obvious that when Thornbury mentions transfer as a source of error he is indicating negative transfer or interference. The concepts underlying developmental errors are a little more tricky. Depending on the age of your learner and developmental error could arise due to incomplete development in the target linguistic system or it could arise due to incomplete development in cognition. Thornbury wisely sidesteps the issue of cognition and its relation to error, but it is something that we need to think about. As I mentioned above in relation to usage errors, many of these come from interfering cognitive systems and not just interfering linguistic repertoires. Unfortunately this is not a question which we have time to tackle here, but it will be picked up a different course later.

To the standard transfer and developmental fold we can add processing errors, but it turns out that these are usually mistakes. They are mistakes because they are not a result of any specific linguistic problem but rather an overall weakness in the way in which the foreign-language linguistic system is wired in the brain. Because of its tenuousness a developing foreign-language is much more fragile and is much more easily affected by both outside and internal forces. What's interesting is that though these are mistakes they can be just as damning as errors in destroying the communicative effect. We therefore cannot simply ignore them because they are mistakes and not errors. As teachers would need to find a way to make our students foreign-language system less vulnerable to processing effects. Again, the answer is very simple: practice.

3. What are some of the different attitudes toward errors?

Most teachers tend to fall into a continuum in relation to their attitudes towards errors. We can express this continuum by stating the two polar views with a middle of the road view in between. At the one end of the continuum we have people who simply want to destroy all errors because they believe that they will fossilize and students will then develop also to difficulties in the future. At the other end of the continuum we have those people who feel that errors are natural and they are good and hopefully they will simply disappear so leave them be. In the middle of course we have those people who believe that some errors are particularly damning and they should be addressed. These three views are represented simply below.

Kill `em all.

Kill the evil ones.

Let them be.

In reality it would seem that none of these views are able to be sustained in the classroom and this tells us that while it is important teachers have certain principles in their teaching it is actually very hard to keep them and generally because of the nature of language is actually pretty damn stupid to try to keep them hell or high water. The person who wants to kill all errors is going to spend all the time correcting one student. The only way to do this is to give your students only accuracy focused practice and never ever allow them to develop fluency, because as soon as they make a mistake then you will have to stop from and fix the mistake. This simply doesn't work. On the opposite side of the continuum we have people who would claim to leave errors alone, but I don't think that is possible. Anyone teaching and more adult learners knows that they want their errors corrected, not all but some. Also, in reality even though we think we are not responding to errors and probably are in more subtle, covert ways. Again we are not responding to all but we have to and will respond to some. The problem is that our responses are often unsystematic because they are unplanned and as a result students really have very little idea of what they're doing right and what they are doing wrong. And it seems that now we have also drifted into the middle of the road position, where we try to choose carefully which errors to respond to and of course we generally or almost never respond to mistakes. In reality this is a difficult position to hold as well although it seems to be the most common sensible position. During class time when students are talking it is very hard to know whether something is an error or a mistake. You need to have a pretty good profile in your head of each of your students language patterns and proficiency levels. This is hard when you have hundreds of students. In addition, as mentioned above we have things like processing mistakes which are often much worse than errors and usage errors which can be not only more damning but also much harder to deal with than basic form type errors. In the spur the moment the probably really don't know which errors are more damning.

One way of mitigating the possible damage caused by erratic and often incorrect error correction is to simply focusing more on positive feedback. Instead of always drawn to students attention to their mistakes which doesn't really helpful since we can't possibly told him about all their mistakes, much more helpful would seem to be telling them what they did correctly. Interestingly Thornbury never mentions anything about positive feedback. It is the only way to really guide students through productive based practice.

4. How can teachers respond to errors?

There are many ways that teachers can respond errors and I think that most of you have a generally well-developed repertoire of responses. I therefore see no need to go through all these different repertoires, but I think the most important thing for us to realize is that we need to use different types of responses for different students and different situations. Again we can make a simple continual based on the concept of directness. Some responses should be more direct, and some should be less direct. Should be noted that the same directness relates as well to positive feedback. Sometimes we want to use individual students as models for other students by focusing on what they have done correctly, sometimes we really don't want to do that. It should be clear that the classroom is a highly variable situation despite the fact that has been specifically created to limit variability in learning. We therefore need to employ our responses in a highly variable yet principled manner.

Simple affective feedback

Recasts

Summarize errors to the class

etc.

Lewis (1997) Chapter 6: Exercises in the lexical approach, pp. 86-107.

5. For Lewis, what is the difference between activities and exercises?

Activities, according to Lewis (1997) are to be done by groups of students while exercises are just for one student working alone. Activities require a more complex array of interactions. Students need to interact with each other as well as the teacher and even with the materials itself in a longer and more complex process. In an exercise the interaction is more or less a one-way interaction between the student and the language materials in the exercise. In this way we can see that while activities are more real world and holistic, exercises allow us to get our students to focus deeply on specific target elements. Therefore the two of them can and need to be used to enhance each other.

6. Lewis says that to learn learners must first recognize and then deduce answers to exercises. How can we make this happen?

The whole idea here is that we want to use exercises to get students to learn and they need to do this through deep processing. In order to get students to process deeply we need to create exercises that are thought-provoking yet doable. As a means to this end Lewis (1997) gives us a simple formula for creating exercises. First we need to make sure that there are some questions that students can actually do. So, part of it should be known. There should also be a limited number of unknown answers. The number should be limited so the students can give use through educated guessing and not simply wild guessing of what the answers should be. If there are too many questions in the first place and too many unknown questions in relation to known questions then the guessing will need to be more wild and less educated or principled. This is as far as Lewis (1997) goes, but to this we can add the simple idea that there should be some patterns in the exercise itself. If there are patterns it would then seem logical to base some of these patterns on the forms that we are pretty sure the students can answer. To put this another way, we use the known questions/answers as prototypes for extending patterns which the students can hopefully be able to guess about and in doing so learn.

7. Of the six exercise types which ones do you currently do and which ones would you like to do more of?

Lewis (1997) gives us six types of exercises we can do with our students to develop their lexical knowledge. They are:

Identifying chunks

Matching

Completing

Categorising

Sequencing

Deleting

Of the six I think four are probably quite familiar to you, and I'm sure you know how to use these also in relation to teaching lexis now. It is the two that I think maybe you don't use so often which I really want to push you to use. These are of course identifying and categorizing. Identifying is an essential first step in developing any kind of knowledge. If we don't identify something we'd probably won't know what it is or that it is even there. Identifying is the key to attention and noticing. Once more, we need to identify things in context so that we can make some sort of value judgment on its importance. Things that are deemed more important will have more attention focused on them. This is a good thing at least for a while. Remember that attention is difficult thing and as new, exciting elements are identified attention will then temporarily focus more on them until they become more known and less new. Again identifying is essential for the beginning stages of learning anything and particularly lexis.

Another really important skill in this list is categorizing. The ability to categorize can not be undervalued, for categorization is the basis of our entire intelligence. Once more, it is through categorization that we are able to systematize and organize all the different input which we have coming in. It is through categorization that we build systems in the mind. Thinking specifically about the mental lexicon we can say that the lexicon is a specialized type of memory. It is memory of which focuses solely on linguistic data. We also know that the lexicon is a complex neural network which organizes information in many, many different ways. We therefore have categories stacked on top of categories in a multidimensional, and not simply one or two dimensional neural network. We must get our students to categorize as much as possible. It is the only way they will ever be able to use any of this knowledge effectively.

8. Of the exercises Lewis supplies which one is your favorite, which is your least favorite?

I think this is something you need to figure out for yourself.