Tuesday, April 17, 2012
By David Barker
Author and Publisher of Materials for Japanese Learners of English
Back to Basics Blog for Teachers
Pronunciation is one element of language courses that often gets overlooked. Part of the reason for this is that experienced teachers know how difficult it is to learn the sounds of a foreign language as an adult, especially if that language is nothing like your own. This basically means we accept that Japanese students will always have a Japanese accent, that Koreans will always have a Korean accent, and so on. Incidentally, I always used to think in terms of learners “gaining” the accent of a foreign language, but I remember hearing a friend talking about a Japanese person he knew who had managed to “lose” her Japanese accent. That is an interesting way of looking at it. I wonder which viewpoint is more common among teachers?
Anyway, as well as acknowledging the difficulty of the task of teaching pronunciation, most teachers also realize that even with a heavy accent, the majority of learners will be able to make themselves understood to proficient speakers of English. The combined effect of these two beliefs is that pronunciation often gets relegated to a once-in-a-while exercise with the sole purpose of providing a bit of variety in the course.
There are at least two problems with this way of thinking. The first is that teachers, particularly those of monolingual classes, are often very poor judges of how comprehensible their students actually are to regular speakers of the language. When I lived in New Zealand, I did the examiner training for IELTS (International English Language Testing System). As part of the workshop, we had to watch videos of candidates speaking and assign grades. What soon became clear was that teachers were giving far higher grades to students of nationalities they were familiar with. For example, two teachers who had worked in Korea gave a Korean student a high grade for her speaking, whereas the teachers who had mainly worked with European learners gave her a low one. Their reasoning was, “We can’t really understand what she is saying.”
The second reason why pronunciation deserves more attention in language courses is that a learner’s knowledge of the sounds of a language will directly affect their ability to perceive and recognize those sounds. In other words, having good pronunciation is just as important for listening as it is for speaking. My limited understanding of how recognition systems work is that they compare sensory input with stored representations of a variety of forms. For example, we learn how the word “boy” sounds, and we then create and store a template of it in our brains. When audio signals reach our ears, they are run through the database in order to find matches. The same principle applies to the recognition of words and letters. You recognize “x” as the letter that comes before “z” because the marks on this screen fit the representation of that letter that you already have stored in your brain. Of course, you would probably recognize it if I wrote it as “X” too, and even if I wrote it by hand. The human brain has an incredible tolerance for variation that allows it to recognize shapes in a way that computers cannot. That is the theory behind those weirdly shaped letters you have to input manually on some blogs in order to post a comment. The system works because humans can tolerate greater manipulation of basic forms than computers can.
Even so, there are limits to the tolerance (I am using the word here in its engineering sense) of even the human brain’s recognition systems, and these become stricter when representations of objects or phenomena resemble each other. For example, in many cases, it is impossible for us to distinguish between “1,” “l,” and “I” when written in isolation because they look so similar. When that happens, the knowledge of language and context that I described in my previous entry kicks in and allows us to make inferences that go beyond the information that is being provided by the senses.
When a language student learns a new word, they create a template for it and store that template in their database. It is quite possible that when they reproduce the word from its template, the audio signal that results will be within the limits of tolerance of proficient speakers of the language, so the learner will be able to make him or herself understood. A problem arises, however, when the focus switches to listening. Because the template the learner has created does not really match the signal produced by proficient speakers, and because the learner’s recognition system will naturally have a more limited tolerance owing to their lower mastery of the language, there is a very good chance that they will not recognize what they are hearing. It’s a bit like going to meet someone that you have never met at an airport armed only with a photograph that was taken twenty years ago. If the person doesn’t actually look like the photograph, there is a good chance that they will walk right past you without you recognizing them at all.
Like all language teachers, I constantly struggle to make myself understood to my students. I have often noticed that the reason my students cannot understand what I am saying is that they have learned an incorrect pronunciation of a particular word. The following is a typical example of a conversation in one of my classes:
Me: Can you close the curtain?
Me: The CURTAIN.
Me: (gesturing) The curtain!!
Student: Ah, kah-ten!!
It is almost as if they are correcting my pronunciation to match their internal representation of the word. Every teacher in Japan knows that we can easily make ourselves understood by simply saying a word the way our students say it, and I suspect the same is true of any teacher with experience of teaching a particular language group.
My point is that learners need to learn words as accurately as possible so that the template they create reflects the audio signal that is produced when proficient speakers of the language pronounce that word. If a learner creates a template that is significantly different, it might be close enough for their recreation of it to be understood by proficient speakers, but it may not be close enough for them to recognize the word when they hear it.
As teachers, I think we need to start realizing that pronunciation is just as much a listening skill as it is a speaking one, and we need to start giving it greater prominence in our courses.