Dean Williams' approach to stuttering therapy:

"Forward Moving Speech"

by Robert Quesal

Dean Williams' approach to therapy is referred to as "Forward Moving Speech," also known as the "Iowa Model" of therapy. As far as I know, the basic therapy ideas may have had their roots with Wendell Johnson, and the approach has been modified over the years. The approach is discussed by Williams (1971; 1979) in Travis's Handbook and Gregory's (1979) Controversies About Stuttering Therapy. What is presented in this paper has been further modified, at least to some extent, by me. The therapy is based on the following ideas: First, most of the time the person who stutters speaks fluently, so we need to try to identify those things that he/she does when he/she speaks successfully, and focus more on speaking fluently and less on "not stuttering." Second, when he/she is disfluent, he/she is doing something, somewhere to interfere with the ability to speak fluently. Third, he/she can modify those interfering behaviors in order to speak more effectively. Fourth, ultimately, the client has to become the world's foremost expert on how he/she talks. Finally, the ultimate goal of therapy is not necessarily perfect fluency. Instead, the goal is for the person who stutters to become the most effective speaker that he/she is capable of becoming.

What the client needs to know early in therapy

We begin therapy with what amounts to a short course in speech physiology. The person who stutters needs to understand that speech is a complex process, and that because of that complexity change may take some time. Most individuals, including some speech pathologists, don't have a sufficient appreciation for the complexity of speech. We have to explain to our client that speaking involves respiration, phonation, and articulation, and that we rely on various types of feedback as we talk. Most people think that we open our mouths and the words fall out; people who stutter are no different. An understanding of speech begins to get the person who stutters away from his animistic way of thinking about speech and stuttering. Stuttering doesn't just "happen," stuttering isn't an "it," words don't "get stuck" anyplace in the speech mechanism. An understanding of speech physiology gives the person who stutters the vocabulary that he/she needs to be able to describe what he/she is doing when he/she is talking and/or stuttering. The vocabulary is simplified somewhat by talking in terms of the "Five Parameters of Forward Moving Speech."

The "Five Parameters of Forward Moving Speech"

There is really nothing magical about the five parameters. We could probably add one or two or delete one or two and it wouldn't affect the therapy approach too much. Simply stated, the five parameters are things one must have, or be doing, in order for speech to flow smoothly. An interruption in one or more of the parameters will lead to a disruption in fluency. I'll discuss each one separately.
Most basically, we have to have air flowing in order to speak. Respiration provides the energy or driving source for speech. The client has to be aware of events occurring at the level of the rib cage, and develop awareness of how much air is in the lungs.
. Movement occurs at a number of levels. We typically think of movement of the speech structures, but we also have movement of the rib cage to consider, as well. It is obvious that without movement, we can't speak.
Timing (rate).
Timing of events is important, because if there is not good "coordination" of the various speech structures and levels of the speech chain, the output will be affected. A basic example of this is voice onset time. If release of constriction precedes onset of voicing by too long a time, we have a voiced sound being produced like a voiceless one (e.g., /pob/ for /bob/). We can refer to timing in terms of "rate" because often if we get a client to slow his/her rate, it leads to considerable change in terms of timing of various speech movements. I'll discuss this further later in the paper.
Sound (voicing).
We obviously have to have some type of sound in order to produce speech, since speech is just a series of sounds. We differentiate sound from voicing, however, because not all sounds are voiced. (Some therapy approaches teach "continuous voicing" as a strategy. That never made sense to me because continuous voicing would lead to a robotic voice.)
It's important to differentiate "tensing" from "tension." Tensing refers to muscular effort or pressure, not an uncomfortable emotional state. What we're talking about, then, is something like tensing of the vocal folds. Since many clients come to us thinking that there is a direct relationship between their speech behavior and their emotions, it's important that we make it clear that this parameter does not refer to an emotional state.

The interrelationship between the five parameters.

Even though I just talked about the five parameters separately, and even though some of what we do in therapy involves "isolating" the parameters, it is important for you and your client to realize that they work together. For example, increased subglottal pressure may lead to increased vocal fold tension. Increased muscle tensing may affect movement, etc. Often a change in one parameter, such a rate, may lead to changes in the others.

Increasing client awareness of the five parameters

Once we have introduced the five parameters to the client, we need to help the client to become more aware of those parameters in his/her speech. This is the point at which we "isolate" the parameters, asking the client to focus on just one thing, and in just one area. A good example of this type of exercise is to have the client focus on tongue movements for a short time, perhaps a minute or so. We don't consciously think about our tongues when we talk, but it's relatively easy to become more aware of tongue movements. By having our client focus on tongue movements, we can have him/her begin to develop an awareness of rate of tongue movements, and tensing of the tongue muscles. Awareness of these two parameters can be addressed separately, however. In other words, we don't say to the client "focus on the movement and tensing of your tongue." We say, "focus on the movement of your tongue." Later we would have them focus on the tensing. Similarly, we can have the client focus on breathing during speech, to develop an awareness of airflow and movement of the rib cage. We can have the client focus on speaking rate in general. At this point, we're not worried about "stuttering" or "not stuttering"--we're simply having the client begin to sense the behaviors of speech at levels which he/she has never thought about before.

The next step is to have the client begin to vary the five parameters. Initially, this can be done by letting the client choose the parameter he/she wants to vary, and how he/she wants to vary it. We may begin with something as simple as saying "try to talk differently." More likely, you would direct the client to one of the parameters, and have him/her vary it, such as having the client speak at a "different" rate. At this point, how the rate is varied is up to the client--all we care about is that the rate is different from the client's habitual rate. Other tasks might include having the client speak with a clenched jaw, having the client speak with a breathy voice, having the client exaggerate movement, etc. The point of this is threefold: First, it helps the client to practice developing awareness of the things that are involved in talking; second, it helps the client to realize that there is a fairly wide range of things we can do with our speaking mechanisms; third, it helps the client to realize that he/she has a certain amount of control over the speech mechanism. In other words, it helps to show that things don't just "happen," but that the client can exert some control over the mechanism.

Once the client demonstrates the ability to change behaviors on his/her own, the next step is to begin to do it under your direction. As an example, we would ask the client to vary speaking rate based upon a visual signal from us. For example, hand up--talk faster, hand down-- talk slower. We could also ask the client to increase/decrease tensing in some area, such as the rib cage or jaw, following the same signal. The client should make the changes gradually, such as moving from a fast rate to a slow rate and back again. This illustrates to the client that it is possible to make gradual changes in behavior, and to make adjustments as one is talking. The notion here is that as the person who stutters realizes that he/she has a certain amount of "control" over what he/she does as a speaker, he/she will start thinking about speech less in terms of a dichotomy -- stutter/not stutter -- and more in terms of a continuous behavior that can be changed at various times.

Even though I have moved through the above "steps" sort of quickly, let me reemphasize that a considerable amount of time should be spent getting the client to explore the five parameters and the changes in the five parameters. You need to remind your client that in order to be able to modify speaking behaviors, he/she needs to know what goes into talking, and be aware of the way he/she talks.

Contrasting fluent and disfluent speech

Once the client does have a good understanding of what he/she does as a speaker, we get to the point where we can have the client begin to contrast fluent and disfluent speech. As I mentioned earlier, when the client is disfluent, he/she is doing something, somewhere to interfere with the ongoing flow of speech. By describing interfering behaviors in terms of speech physiology and the five parameters, it gives the client specific things to try to change, and it makes stuttering something that can be discussed and thought about in concrete terms--we get away from using language like "it," "the word got stuck," "it just happened," "I stuttered because I was nervous," etc.

Initially, we would prepare the client for this part of therapy by explaining that for a certain period of time, we would be stopping him/her when a disfluency was noted. For example, we would say, "For the next 10 minutes, I want you to talk to me and focus on what you are doing. When you have trouble, I'm going to stop you and ask you to tell me what you did to interfere with speech." The language we use here is directed toward what the client is doing, not what is happening. We have to be careful to be sure that our language is descriptive, and that the client uses descriptive language, too. Therefore, we don't say, "tell me what happened;" we say, "tell me what you did." The reason for this is pretty basic, but pretty important: If the client is thinking in terms of stuttering happening to him/her, and if you talk that way, it suggests that stuttering is something that the person who stutters can't do anything about. If that was truly the case, there would be no point in doing therapy. (Pardon the digression.)

We have the client contrast fluent and disfluent speech. We stop the client when he/she is disfluent, and ask the client to tell us what he/she did, and to describe it in terms of the five parameters. At this point, we could say that therapy becomes quite individualized. The client is beginning to think about and describe the things that are specific to his/her stuttering. We can observe what the client does, and perhaps offer some suggestions as to what the client is doing, but only the client knows what is going on. When going through this "contrasting" stage of therapy, it is likely that the client will identify a relatively few things that interfere with the forward flow of speech--these will become the "targets" for therapy. These may include things such as too fast a speaking rate, too much laryngeal tension, or improper breathing. However, it is up to the client, under your guidance, to identify those things specific to his/her speech that affect fluency. One caution at this point: Sometimes, the person who stutters will simply replace a word or phrase that is not too descriptive (e.g., "the word got stuck"), with what seems like a descriptive word or phrase, but that is really just as meaningless. For example, the person who stutters will begin to use "tensing" as a descriptor. You ask, "What did you do there?" and the client responds, "I tensed." We as clinicians like that, because the language is more descriptive. However, if we listen to our client over a period of time, we find that what "tensed" means is "stuttered." When the client is fluent, he/she is not tensing; when he/she is disfluent, he/she is tensing. All that has happened is that "tensed" and "not tensed" have replaced "stuttered" and "not stuttered." A way to guard against this is to make sure that the client is using descriptive language rather than just labels. In other words, "I tensed my vocal folds" is better than "I tensed." Further, to say, "I tensed my vocal folds and could also feel some pressure in my rib cage" is much better. Your client won't reach this level of description without some help from you, so this part of therapy will also take some time. What the client needs is a good understanding of the way his/her disfluent speech is different from his/her fluent speech.

Changing ineffective speaking behaviors into more effective ones Once you are comfortable with the client's knowledge and understanding of the way he/she talks, you move on to changing ineffective speaking behaviors into more effective ones. There is a "bridge" here between the previous step in therapy and this step, and that is the ability to identify differences between fluent and disfluent speech. At this point, however, the client has to begin changing interfering behaviors. Initially, when the client is disfluent, you would ask him/her to "do something different" when in a disfluency. It's up to the client to make the change, but he/she has to do something. Very few people who stutter can't make some adjustment, even if it's to stop talking, when in a block. The notion here is to show the person who stutters that his disfluencies do not control him/her, but that he/she has some degree of "control" over what he/she does during a disfluency. You will need to "shape" these changes with the client, and in some cases will need to have the client "recreate" a disfluency in order to practice changing from ineffective to more effective speech. This step of therapy can prove difficult at times, because what the person who stutters wants most to do is "get out of" a block, and you are asking him/her to take the time to analyze the block and to work through it. Ultimately, the client should come to realize that he/she is not "helpless" in the face of a block, and that there are constructive, positive things that he/she can do to work through a disfluency.

Changing on the fly

As the client's awareness of his/her speech behavior increases, the client should learn that it is possible to change behavior before an overt instance of stuttering. For example, the client whose disfluencies often consist of increased laryngeal tensing can learn to speak with those muscles more relaxed most of the time, or perhaps all of the time. The client doesn't have to modify behavior just when a stutter is "coming on" or "happening." Instead, the client can try to use a more effective speaking pattern all the time.

I think that it is important for the client to realize that he/she will not be fluent all the time, nor will he/she always be able to change from ineffective to effective speaking behaviors. This will be particularly true in situations where emotional arousal is high. You have to stress to the client that this is the way he/she has been talking all his/her life, and that the changes in therapy will take time to become ingrained. However, with this approach to therapy, the client has developed positive behaviors to use in place of "tricks" that he/she may have relied on before. You need to help the client to realize the changes that he/she is making over the course of therapy, and to stress the important role of the client in terms of identifying, describing, and changing behavior. We can't "make" our clients fluent--we simply provide them with the knowledge and skills necessary to help them to be more fluent. However, it is up to them to utilize the new information.

Modifications for younger clients

With younger clients, you have to informally assess the level of complexity that they can understand, and modify your language and terminology accordingly. For example, we wouldn't talk about the larynx, we'd talk about the "voice box." Airflow might become simply "air," voicing might become a "buzzing sound," and tensing might become "muscles pushing." These are just suggestions, and you will have to work with your client to reach a mutually satisfactory vocabulary. As Williams (1986) points out, kids generally are a lot more "tuned in" than we give them credit for. I would say that our primary responsibility early in therapy is to make sure that what we are saying to the client makes sense. If the client is not able to think about and talk about the way they talk, a different approach to therapy is probably called for.


Certain clients, who have the ability and motivation to think about and describe their speech, will benefit from an approach which actively involves them in the process of change. The "forward moving speech" approach is designed to do just that. If you and your client spend time in therapy learning and talking about things specific to your client's speech, the client can come to develop a better understanding of the things he/she does as a speaker, and can use that knowledge to become more effective speaker. In addition, this knowledge can be used by the client to help in long-term carryover of fluency skills.


Williams, D.E. (ca. 1986). Talking with children who stutter. In: Counseling Stutterers. Memphis, TN: Speech Foundation of America.

Williams, D.E. (1979). A perspective on approaches to stuttering therapy. In: Gregory, H.H., Controversies about stuttering therapy. Baltimore: University Park Press.

Williams, D.E. (1971). Stuttering therapy for children. In: Travis, L.E., Handbook of speech pathology and audiology. New York: Appleton-Century-Crofts.

Questions and comments about this paper can be sent to

Readers are free to cite any and all of this paper, but I request that you cite that the information came from me. Thanks.

Robert W. Quesal, Ph.D.
Back to Stuttering Home Page