Second Language Learning and Software

I am often asked about learning a language through software. What are the strengths and perhaps shortcomings of learning with the aid of modern technology? Some learners prefer the dynamics of a live classroom where interacting with teacher and classmates is integral to their learning. Some learners prefer the convenience of studying at a time of their choosing and being able to control their own rate of learning. It seems to me that approaching a discussion in terms of individual value and preferences is one way to proceed and the end of such a discussion will sound much like the beginning of it. In other words, attempting to deal with the question in terms of individual preferences doesn’t really address the question I posed.

There are other ways to at least attempt a beginning answer to the question. The first thing to decide is what type of an answer, answers my question. I would want an answer, for example, about weaknesses working with software, not to be remedied by technological advance. If I said, for example, this or that software is not interactive enough, this would not be a possible answer to my original question. I would be answering a question which asks about “today’s” technology. I asked about “modern” technology. The answer I would be looking for has something to do with the inherent nature of language (which may also be true of classroom learning as well).

I notice that most software, by that I mean Rosetta Stone, Pimsleur, and others state something like the following: learning through our method is similar to the way you learned your first language and it will be an enjoyable experience. Mentioning a first language probably conjures up pictures of being fluent to most of us but that is not really an explicit claim that can be made. Perhaps this form of advertising is meant to show that it is easy to learn by such a method since you had no trouble at all learning your first language. Classroom methods of the seventies made the same claim: Total Physical Response, Suggestopedea, Community Language Learning and other of the so-called Designer Methods. What they all had in common was that they all failed to produce fluent speakers. That is to say, none produced fluency to any relatively high level in second language learning. Perhaps that is one reason we, in language teaching, say nowadays we are in a post-method period. We no longer believe someone will discover a cookie cutter method that is a panacea for learning a second language.

One obvious way that these methods are similar to the methods used in software, is that they superimpose one method upon all learners and with languages of extremely differing characteristics. Mass-produced methods are particularly troublesome with the two variables to be accounted for: learners that differ and languages that differ. So this is one reason, I imagine, but not the strongest reason that one method alone cannot produce fluent speakers of a second language. I still would have to find something inherent within the nature of language which makes fluency in a second language a rather different matter from the natural way we appear to become fluent in our first.

It really is a miracle how we learn our first language. We know so much and yet so little about the brain. The two hemispheres of the brain are connected by nerve fibers making it possible for one side to “communicate” with the other. How the “communication” produces language is confounding and there are many different theories. Somehow our language is a natural part of our maturation process rather like flying is to a bird, speech is to us. Flying and speaking emerge before it is necessary; which is one reason we call the process an innate process. At six months, more or less, a child begins to make sounds that start to resemble one-syllable words (e.g. ‘ma’ or ‘da’). The appearance of the phenomena is not the result of a conscious decision. After a year, some children actually begin to imitate sounds: (e.g. ‘mamma’). The appearance of flying or language does not appear to be triggered by an external event, though the environment must be sufficiently rich for the event to occur. There is, of course, mixture with the environment, otherwise there would be no reason for French children to speak French. Direct teaching has relatively little effect. At eighteen months a child has between six and fifty words and there is not a lot that a parent can do about it. There is a regular sequence of milestones, perhaps associated with age (e.g. voice changing in males). By the time a child is three years old, the child will have a thousand words and could be understood by most strangers. Some theorists posit that there is a critical period in which if the process fails to occur, it will not. Some birds don’t fly if they are not out of the nest early enough and they don’t live very long. With humans this hypothesis is less certain as it is unlikely to find teenagers, for example, with no language unless the circumstances involves horrendous interference as in the most famous case study found of this purported phenomena.

Pimsleur software is based on the premise that we learn our new language as we learned our first, by hearing and imitating. I have already suggested we do not entirely learn our first language by imitating. How many children pluralizing the word “foot” go through the process of saying “foots,” then “feets,” and finally “feet”? It is unlikely they are imitating their caregiver. How many children call any round thing “ball” including a quarter or the moon? Imitation theory has been discredited long ago. However, Pimsleur methodology is easy enough. Imitation theory may explain why a Mexican child speaks Spanish but it explains little else about language.

From the first minutes with Pimsleur software, the learner hears native speakers converse. Then the learner repeats phrases and words as they were heard. The learner is actually prompted to say something in a new way before long and this is one aspect of Pimsleur which is interesting. There is a claim that the learner is unconsciously learning the rules of the new language: grammar, word order, verb conjugation etc. I have never been sure how one is supposed to verify claims that depend on the “unconscious” as an explanatory device but I will ignore that point. They could patch it up with another word or words. Presumably one learns to communicate or use language first and then, perhaps, learns the rules of the language. My question is how does this all lead to advanced fluency? Think of English and how you became a fluent speaker. Do you know when to use an infinitive or a gerund? Or do you just use them correctly? How did you find out about irregular verbs? How did you learn as a child to speak differently to your teacher than you did to your neighbor who was your age? How did you find out that some parts of the language are for speaking and some parts of the language are for writing? These are all things you know how to do with your first language and if you could do them with a second or third or fourth language, you could not have done so by learning through software or simply in the classroom. Fluency comes with knowing what to say, who to say it to, how to say it, and when to say it. In other words fluency comes from involvement in a life, in a culture, and not as an outsider learning certain features of a language.

Other software, for example Rosetta Stone, is perhaps the most highly advertised and so they must certainly be successful in business. The learner associates sounds and words with visual imagery, which is an important learning tool. Here too, in fact, Rosetta Stone maintains that this is the way you learned your native language. At first a word matches a picture on the screen. Next you have to select what word goes with what picture. If you are correct, you move on. In time words become phrases and, in turn, they become sentences. It appears as though pattern recognition is important to learning a language in this way. There are ways to work on the classic skills of reading, writing, listening, and speaking within the software. There is a voice recognition feature as well which helps in pronunciation.

Each of the software is surely correct in not using translation to learn the language. I do not see how either addresses fluency issues at any interesting level; however. In a sense they are both foundation builders just as were the so-called Designer Methods. Fluency gives speech the qualities of being natural and normal, including native-like use of pausing, rhythm, intonation, stress, rate of speaking, and use of interjections and interruptions. In second language teaching we are talking about good but not necessarily perfect command of intonation, vocabulary, and grammar. The learner should be able to communicate ideas effectively. The learner should have the ability to produce continuous speech without causing comprehension difficulties or a breakdown in communication.

I do seem to have answered my question without appealing to the way in which a wide variety of languages differ from one another which may and does impact second language effective learning styles. The great strength in technology is that someday effective foundations can be built. Pronunciation certainly has been affected by technology. Some individuals learn best at times of their choosing and by themselves to a certain degree, choosing to go at their own speed and guided by their own lights. The great novelist Joseph Conrad may have profited from software had there been any a century ago. Conrad learned English at nineteen (coming from Poland) and had very little interest in speaking it; thus his friends often commented on trouble understanding him. He did write some of the finest prose in the English language, however.

On the other hand this idea that second language learning is analogous to first language or can be made to be is a little misleading. You became fluent in your first and maybe in your other languages by human contact with speakers of those languages. You learned how to interact with people at very different levels of language in this most human of all actions. Language is defined as a particular system of human communication. The software doesn’t seem to involve much in the way of human communication, so fluency must involve human relationships, cultural relationships, and a way of living that software cannot provide.

–Dr. Paul Schneider, Director of Teacher Education Programs, WAL

Check our more from the WAL Blog at www.wal.org/wordpress!