Understanding Fluency Training
Fluency is the automatic process of producing speech. We can ask, "Where do the words come from?", and "How can we train this part of us to be more fluent?" Our goal is to have smooth correct speech in a new language. What should we be doing and what should we be thinking about? This article will provide an understanding of fluency, and a set of ideas about how to structure fluency training.
Traditional methods of language training often ignore verbal fluency. But what is fluency? You are fluent in your native language. You can speak without planning which words to use to compose the sentence you want to produce. You have an idea in your mind. Something inside you produces the sentence. Often you do not know exactly what you will say until you hear yourself say it.
When one wishes to develop verbal fluency in a language, it is necessary to physically train the nerves and muscles of the human vocal apparatus. We can think of the training process as being identical to the methods used to train people in the martial arts. Students assemble in a gymnasium and follow the teacher in a series of movements. The teacher critiques their efforts, helping them to achieve perfect form. They practice each movement many times; perhaps thousands of times. Each time they practice, the teacher coaches them in the small changes they need to make to achieve perfection. Then when the skill is required, it functions automatically, without consious thought.
In the case of a fighting skill, tae kwondo or wu shu, the trained behavior is triggered by an external event; perhaps the challenge of danger. The defensive motions are produced as surely and reliably as the pulling of the trigger of a pistol fires a bullet from its muzzle. Wu shu is a physical skill, not a mental skill. Speech fluency is a physical skill, not a mental skill. You would not expect to be able to lean to play basketball well only from a book. Practice, much practice, is necessary. In the same way, grammatical speech is a part of fluency. Learning grammar from a book can give teacher and student a common way to talk about speech, but it is only speaking practice that brings the student to fluent and grammatical speech. The student will not learn to speak by reading. The student will not learn to speak by writing. The student will not learn to speak by listening. The student learns to speak only by speaking!
Speech is only slightly different from wu shu. Speech behavior is triggered by an intention in the speaker's brain. Many people, if they think about it, have had the experience that they do not plan what they will say in any detail when speaking in their native language. They do not use book-learned grammar to assemble sentences. Often they do not know what they are saying until they hear themselves saying it. It is a little bit like listening to someone else speak. The speaker is aware of his intention related to the spoken sentence. It may be a feeling, a remembered sound, or a mental picture. But he is strongly aware that he did not plan the sentence!
Human speech can be modeled by a finite state machine in which each state is a word or a phrase. The model predicts the likelihood of the next state when chosen from all possible states. An example is the phrase fragment, "How are . . . (fill in the blank)?" You can easily supply the next word, "you". But the phrase fragment could also be completed with, "sausages made", or "we going to get home". Your brain chose the most likely completion for the phrase. Mathematicians call this sort of model a 'conditional probability distribution". Computers are good at processing such a model, but our brains are better at it.
It is strange and wonderful to realize that there is a part of our brain that is creating intelligent noises without our full and detailed participation. We are speaking on automatic pilot. We can do so because we have a deeply embedded pattern for each kind of sentence we might utter. It is almost as if we have said everything before and we must simply access a recording of what we wish to say.
If we can have an understanding of the mechanism of fluent speech, we can justify the efforts required to gain mastery of it. You might think of this discussion as a pedagogical theory. But it is more. It is a fascinating exploration of a startling idea.
Philosophers and cognitive scientists have taken the term "zombie" to describe some phenomena of consciousness we are concerned with here. A zombie is a person who appears to be just like anyone else, but he has no conscious awareness. The idea comes from the Haitian voodoo tradition. In that culture there was a belief that the dead could be reanimated and used as slaves. But we are specifically talking about philosophical zombies here, rather than the zombies found in Hollywood movies. We can think of a zombie as a person who is dead inside. We will ignore the argument that one cannot have a functioning person who has no consciousness. That may or may not be true. We are engaging here in a thought experiment, and it is a condition of the experiment that we ignore this problem.
We don't understand consciousness very well. We don't know what it is, or where it comes from. We experience consciousness, so we believe it exists. Part of the problem is that the word "consciousness" is just a noise. It is not a real thing like a rock or a cat. That makes it hard to agree on what the noise "consciousness" denotes.
Inherent in this definitional problem are two issues. The first is subjectivity. If I see a green colored object and you see the same object, we are never sure we are seeing the same color. Perhaps your green is my yellow!
The second problem is that the concept of consciousness is amorphic. It cannot be clearly defined. We can both be using the word "consciousness" but mean different things. We are comfortable with this ambiguity. We deal with it every day. But this ambiguity poses problems for us in finding an accurate agreed-upon definition of consciousness, or anything else for that matter.
For example, we all use the word "love", but we do not agree on what it means. Is it a feeling? Is it a behavior? What is the felt quality (qualia) of love that we experience? Is it the same for me as it is for you? Dictionaries give us the illusion that we have shared definitions. But in real life we have tacitly agreed to not agree. Is it part of our culture? Or does it run deeper than that? Perhaps it is just part of being human.
We understand that we cannot disambiguate the idea of consciousness. At least, not here in this thought experiment. In fact, we may never do so, as the elements of consciousness that we experience are not subject to reproduction and validation. We call the subjective feelings that accompany experience "qualia". Qualia cannot be seen or felt or measured from the outside. This means they are not open to normal means of scientific inquiry. This is what makes consciousness a dark mystery.
Benjamin Libet's work toward understanding conscious awareness sheds some light on our problem of the conscious mechanisms involved in the production of speech, or the lack of them. Libet was able to demonstrate that a measured readiness potential in the brain signaled the initiation of an action. This readiness potential occurred about 500 milliseconds before the observed action.
The diagram below shows three events. It is possible to measure, in the brain, the point in time when action is initiated. This is called (Libet) "readiness potential" (RP). This signifies that action has started in the brain, but not yet in the body. About 500 milliseconds later the body starts to move. Careful measurements discovered the time at which the person became aware of the action. The action (RP) started before the person was aware of starting the action! This suggests that it was not the person's "self" that started the action, for what is "self" but awareness.
In regard to speech then, how can I know I decided to say or do something, when the "I" clearly was not involved in the decision to act? When I utter a sentence, the decision to speak, indeed, the entire process of the construction of the sentence, is effected outside of our conscious awareness. The part of us that makes these decisions and carries them out is separate from our conscious awareness. It is as if a zombie is deciding and acting. Our consciousness is only a passenger riding the zombie and observing what takes place.
Our illusion that we have free will is related to the subjective experience that our initiation of an act is simultaneous with the fresh awareness of the initiation of the act. We do not perceive the time delay subjectively. We believe that we, first, decided to act, and then, acted.
This has implications for the training of verbal fluency. If there is no conscious volitional composition of grammatically correct sentences, the teaching of grammar has little or no bearing on the production of correct speech. In fact, if it were so that one assembled sentences deliberately, speech would be painfully slow. Learners of a new language can attest to this.
Cognitive scientists and philosophers call this hidden unaware part of us "the zombie". The zombie is a very useful idea in the understanding of how our brain works and how some learning takes place.
The production of a sentence, a unit of meaning, is more related to the response a martial artist makes to the stimulus of threat. First there is intention or an idea. One is aware of a topic, and an ideational trend within that topic. If the topic is at a low intellectual level, such as gossip or the relating of an event, the mechanism of speech production goes on primarily unaided. At a higher intellectual level one is aware of drawing associations between disparate concepts. This serves to feed the mill of speech.
When a student asks me to train them in verbal fluency, I give them material to memorize. The nature of the material is not critical, save that it consists of correct sentences in a variety of sentence forms using a variety of words and concepts. I find that the lyrics of popular songs serve this purpose well, as enough of them can be found to have a sufficient variety of topics to provide a broad base for the later production of speech. They also have the advantage that they can be sung, thus providing an additional cognitive modality to broaden the learning.
Babies start out learning speech by imitating sounds and exploring the range of sounds they are capable of making. Later they combine the sounds imitatively to form words, and words to form ideas and intentions. It is this ontogenesis of verbal facility that I seek to recapitulate in fluency training.
Students listen to audio recordings of song lyrics. They imitate them. Often the Roman alphabet is used in the student's own language in a way that interferes with correct English pronunciation. Reading at this point is contraindicated. When the student is able to say and sing the words from memory, then meaning can be addressed.
The student is directed to listen to the talking recording, and to repeat it out loud. The lesson consists of the student being coached in pronunciation of the material. This is important, because if listening skills are not refined, pronunciation mistakes will be made. Then the student is given a period of time to memorize the material.
Each lesson consists of three parts. 1/ In the beginning the previous material is recited by the student and corrected by the teacher. 2/ In the second phase of the lesson, the new material is introduced and the student is coached as to the correct pronunciation and prosody (intonation) of the recitation. 3/ In the third phase of the lesson there is discussion of the meaning of the previous materials, and if there is time, free talk.
For the student then, training the zombie in fluent speech is much like the process of physical training in wu shu. Old material is practiced and refined. New techniques are introduced when the student is ready. Finally the teacher helps the student make the connection to meaning.
Training material should consist of audio recordings and written materials. The materials should cover short stories, dialogs and music lyrics. They should all be memorized by the student. Music lyrics should be recited as well as sung, because the production of music occupies a different part of the brain than does speech production. One piece of training material should be attempted at one time, but frequent recitation of recently learned material is important because it aids in the transition from short-term memory to long-term memory.
The teacher is a coach, as in athletic training. He/she must listen and watch the student's recitation, and offer coaching in the finer points of pronunciation and intonation. If speech is slow and halting, the student should be directed to re-study the material. Fluency is demonstrated when speech is smooth, clear and without confusion. Students are often surprised that their recitation proceeds without conscious thought. This is as it should be, because the zombie is at work.
Bibliography
Free Will and Free Won'tMotor activity in the brain precedes our awareness of the intention to move, so how is it that we perceive control? Sukhvinder S. Obhi, Patrick Haggard
The I Illusion
This talk mulls over how/why humans construct these things we call a self or an "I" . -- Deric Bownds
Conscious Mind and Free Will
A Course in Consciousness -- Stanley Sobottka, Emeritus Professor of Physics, University of Virginia, Charlottesville, VA 22904-4714
The Volitional Brain
Towards a Neuroscience of Free Will, Edited by Benjamin Libet, Anthony Freeman and Keith Sutherland
Susan Blackmore on Ben Libet
Mind Over Matter? by Susan Blackmore
