Translation Plateau

Having met many kinds of students over the years, I find myself in a constant state of sorting out: what is a personality trait? What is cultural inheritance? What is natural to all language learners? After a number of years many broad patterns appear. The cultural inheritance is interesting and broadly affects many things in language learning. But more broadly, we are all given a mind that seeks to be efficient, and time and desire (presumably) to learn a language: what accounts for differences in ability and progress? What advice can we give all language learners based on what we know? In other words, I presume that there are generalizations one can make about language learners regardless of their cultural or national background, and I hope to find them, and in this case, analyze them more deeply.

I am lucky in that in the situation I am in, almost all are sincere; most are not distracted by having to make a living, or survive a war while they are learning. And the language I am teaching is English; while it is different from others in key respects, I'm also sure that in most respects it is similar to all languages, and, as I often assure my students, if I were learning another language, I would be facing the same problems they are today. So that, while my observations are based on the ESL classroom, it's a running question: how would this be different in another language classroom? In some cases, probably quite different, as we will see.

The subject of this tome is a problem that occurs with intermediate language learners, and doesn't really apply to lower level or higher level learners. It is specific to those who have a wide base of language skills but are not quite ready to jump into full fluency, in fact, not even close. They arrive at what I call the translation plateau, (which I have defined below) at which point their test scores, and seemingly their ability, levels off indefinitely, but sometimes only for a short while. It could be that not all language learners reach this plateau, but enough do that I've begun to study it.

To students in the old TOEFL scoring system, the level at which this plateau would be found is considered to be about 440-470. Students who take the TOEFL regularly watch for their progress and are justifiably disappointed when hard work doesn't pay off in visible score jumps. When a student's skill level levels off, it appears to them that their skills are going both up and down, because over a period of time, their test scores actually rise and fall, given that at that level, there is a lot of luck involved in any given TOEFL test, and on any given day their luck could be either good or bad. So one could say that the first sign that a student has arrived at this plateau might be the complaint that their score had actually gone down. Upon more careful observation, however, I have noticed other patterns or trends that seem to accompany this particular problem. Stated briefly, they are that the student is an inefficient reader, a poor retainer of vocabulary, and a poor master of word-forms (the scourge of the grammar section). These all boil down to a single root problem: that they are still translating every word, or too many of the words, as they read. Thus the translation plateau.

Roughly defined, the translation plateau is this: When an intermediate student has the skills to read sentences without translating every word to the native language, and yet persists in translating every word, the process he/she is forced into using is unnecessarily bulky and inefficient, and results in several accompanying problems which collectively stunt that student¹s progress until he/she changes his/her parsing system to a more efficient one. This change is difficult, and may take months or years, or may in fact never take place; nevertheless at the time the student is in this transitory period, he/she can be said to be at the translation plateau.

A closer look at the problem revealed several things to me. One was that, whatever system they were using to master vocabulary, it wasn't really working. They were unable to really use enough of it at any one time to stay in English while reading a sentence. I began to tackle the problem from the angle of their familiarity with words.

The problem hit home to me one day in a TOEFL class (yes, they still have such things, though this story occurred in the mid-1990's). I had a fairly high-level bunch, who understood what I said, but they were stuck on a particular question that hinged on the word brisk, which in turn appeared in a sentence such as, "It's a nice day to take a brisk walk." I was about to chastise them for not knowing or being able to guess the meaning of the key word, when it occurred to me that what had really gotten them (and this was true of the vast majority of a class of about thirteen) was that, having translated the word walk, they were unable to grasp that brisk had to be an adjective.

In other words, staying at the surface, in English, with the word walk, one should clearly see that it was a noun, due to the article at the beginning of the noun phrase. Otherwise one would have to interpret take as causative, or rearrange grammatical understanding of English, which many of them were apparently willing to do. For, rather than stay at the surface, most had instantly translated walk into their native languages (where it was no doubt a verb and only a verb) and been stumped. For the end result of translating every word in virtually every translating situation is a grammatical jumble that is very familiar in the art of translating but is very difficult to work with, on a practical level, for a student who is wading through, say, a college textbook or a TOEFL exam.

Why, I asked myself, had so many high-level students fallen into that trap?

A better question could be framed as, was ETS (makers of TOEFL) aware of the trap they had set? And of course, they were, as they well know which of the millions of English words can be both verb and noun, and willingly exploit that list.

But back to our victims. Why had so many people translated what appeared to be a relatively easy sentence? And what accounted for the fact that, having fallen into the trap, they were unable to get back out?

In order to answer the above questions, I should mention that I work under a number of assumptions, and these may or may not be provable or even true. I willingly and readily challenge all readers to contribute in some way to proving or disproving these theories. They are based very practically on watching students over a period of about twenty years; as I said above, most were ESL students who were not for the most part distracted and thus were able to contribute most of their sincere energy to language learning.

Here are a few of my assumptions:

By far the most crucial aspect of a student's progress is the way they tackle a sentence.
The most crucial determinant of how they manage to understand the sentence is their relative familiarity with the vocabulary in it.
Learning a second or third language is very different from learning a first, because at some level all learners are going to use their first as a guide in understanding their second; they will also have habits that will influence their development.

As we look at the students in the story, and at many other similar students in similar positions, we start with the way they handle a group of words on the page. The student may translate one word, several words, or all words; he/she may be unable to translate one, several or all; he/she may even be reasonably familiar with most or all of the words and translate anyway. In our given sentence, they may have translated walk because they didn't understand brisk, or they may have translated all words routinely for other reasons. The variables we can identify are as follows:

First, there is a familiarity hierarchy which can explain the degree to which any given student is familiar with any given word. Roughly defined, the familiarity hierarchy states that some kinds of familiarity are of a higher order than others; for our students to have recognized the multiple grammatical usefulness of the word walk, for example, they would have had to be operating at the upper ends of this hierarchy. A learner may recognize a word on sight; may know the sound, and then, may know the meaning; these would comprise a base of familiarity around which most other knowledge would be attached. In other words, it wouldn't do much good for a learner to know more complex things about a word, if in fact he/she had serious gaps at the bottom of this hierarchy.

One has to fit the learner's translation of the word somewhere in this picture, and I do, but obviously what we would hope for would be that the learner, in the mind's endless and ruthless quest for efficiency, would find this translation ultimately useless and cast it away.

As an aside, I should mention that my approach will have to account for different kinds of learners: some learners are extremely efficient readers, with almost no listening ability; others are extremely good orally, with no ability to recognize words on a page; the vast majority are between those extremes. These different extreme kinds of learners will be considered later, if possible. At this moment I would just like to assume that our typical victim is better at one than the other; i.e. he/she reads and makes it sound good to understand it, or hears, and tries to visualize it to understand it. Being a typical learner, he/she has skills and gaps in both areas, but may be able to use each to help the other.

Given the familiarity hierarchy, then, what happens? The learner has some knowledge or familiarity with each word he/she encounters; and if this is not enough, translation may be necessary or may be the only option; or, worse yet, the dictionary may be the only option. Let's assume that the reader has plenty of familiarity with each word in the sentence. and translates anyway. One would then infer several psychological variables, and perhaps approach the problem by asking the learner, or trying to find out, which is most appropriate.

First, one could say that the learner may have a Lack of Confidence; is afraid that, staying in English, he/she will get the wrong meaning, whereas, coming back to his/her native language (NL) he/she will be more comfortable, more confident, even given the incredible grammatical problems engendered in the switch. This variable we could call Self-Confidence.

Then, there could be another, which I've identified as Cultural Stubbornness; this one can be best described by saying that the student has plenty of confidence, but would rather not stay in English for entirely personal reasons, much as you might want to keep your shoes on in someone else's house, even knowing full well that they would really prefer, for cultural reasons, that you take them off. This student has seen the new culture but is deliberately choosing not to adapt to it, fully aware of the negative consequences. The learner, however, may not be aware of his/her reasons. I should mention here that these factors are not discreet; in other words, who am I to say that these two can't co-exist in one learner, or be part and parcel of the same larger variable? I can't. Thus I'll continue, and we'll try to sort them out later.

Our learner could have what I'll call Fear of the Abyss, which briefly stated, is the fear that having jumped over the bridge to thinking entirely in English, or whatever new language, one will never be able to jump back, never return to the sense of normalcy and security enjoyed at the moment. This fear, fear of losing a sense of identity, would obviously be very threatening to many, and present a hurdle one could be unable to master. It also presents a cultural question. Would English, representing the culture that it does, not be more threatening to people than, say, French would be to an native English speaker? In other words, this factor would presumably vary according to the learner's perception of the culture that is represented by the language he/she is entering.

Our learner could suffer from Unwillingness to take risks. This has been identified by many as a prerequisite to good language learning, and is certainly at play here. Briefly stated, there is a very real risk that, by staying in English, one will lose meaning or misinterpret in some way that would not occur if one were to continue translating. This is understandable.

Then, there is Force of habit. An argument can be made that, well, sometimes we know what would be best, but continue doing what we've always done, just because we've always done that. And to break a habit is obviously harder than to just learn a new skill out of the blue. So what we, as intermediate language teachers, are faced with here, is undoing the habits that have been developed for these learners. For example, as beginners they no doubt had to make a vocabulary list that had second-language/first-language correspondence, and had to set about memorizing these correspondences: how else could they get a handle on anything that was happening in their new language? If this is no longer an efficient or advisable system for them, then we'll have to help them undo this habit, which obviously will be harder than simply teaching them a new trick.

Finally, we should consider sentence-specific triggers, which is another way of saying that my class could very easily have been reading without translating, until they came upon this particular sentence, and then, as a class, all proceeded to translate every word, simply because there was one that they didn't have a handle on. I would buy this idea more easily if I hadn't seen so many hundreds of students with the plateauing TOEFL score complaint. I have made some other assumptions about learner behavior below. First, I should ask: Could there be more? I would love to hear about them.