Unfortunately, it is a too common scenario. One that I have personally experienced, witnessed, and had described to me by friends, colleagues, and clients. A person undergoes stuttering therapy--maybe an intensive clinic, maybe weekly sessions over several months--and makes dramatic changes in his speech. Following treatment, he feels on top of the world, brimming with hard-earned confidence, fluency, and freedom.

And then, somewhere down the line, he comes crashing back down to Earth.

What became so routine within the confines of the intensive experience or over the course of the weekly sessions comes undone. Sometimes it happens suddenly. A monstrous block that comes out of nowhere during a phone call, or an embarrassing comment from a stranger can be all it takes to quash the newfound fluency. More often, however, it occurs gradually: sort of a slow erosion of skills that seemed so sturdy when they were first learned.

Why is this so common with persons who stutter (PWS)? What is it about changing the way one speaks that it can be accomplished so readily, but often fails to endure for long? That is what this paper will explore.

I will start by stating that it is not my intent to critique stuttering treatments. I will also not be theorizing on the construct of relapse, although it certainly is an interesting enough topic in itself. I will simply note that relapse rates following stuttering intervention vary widely depending on the source and the definition of relapse that is used. I have seen numbers as low as 30% (Craig, Feyer, & Andrews, 1987) and as high as 84% (National Stuttering Association, 2009). Some researchers go so far as to pessimistically speak of an "inexorable propensity for relapse" (Dayalu & Kalinowski, 2001, p. 405) among PWS. While I hope that reasonably-minded people are not quite so fatalistic, certainly relapse is an acknowledged problem that continues to plague even the best stuttering treatments. Lastly, I will be referring mainly to interventions that heavily or entirely focus on speech restructuring techniques (i.e., fluency shaping), since the outcome of those approaches is a new manner of speaking that can potentially enable one to pass for a normally fluent speaker.

From the perspective of PWS, there are two common explanations given for relapse:

• Constantly using speech restructuring techniques is difficult and exhausting
• The unnaturalness of one's speech following such therapy is unacceptable

From the perspective of some clinicians, I might add a third reason commonly given for why so many clients relapse:

• Lack of follow-up practice

These explanations for relapse are not without merit. In fact, I believe they apply to most cases, at least in part. However, for a significant portion of PWS, they may be too superficial. Perhaps they fail to account for the internal struggle that occurs when a person attempts to change thoughts and behaviors that have prevailed for so long. The rest of this paper will deal with other reasons why someone might be reluctant or unable to continue speaking fluently once they have learned how to do so.

We all have various "identities" that, together, coalesce to form what we collectively call our "self." These identities can be tied to characteristics over which we have no control, such as race, gender, nationality, family position, and so on. They can also stem from our own interests, talents, and achievements, such as "Baseball Player," "Mother," "Dentist," or "Vegan." For PWS, the identity of "Stutterer" will also be there, somewhere, playing a larger or smaller role from one person to the next. For some people, the stuttering identity is negligible, forming only a fraction of the self, on par with other minor attributes like hair color or musical ability. For others, stuttering plays a larger role in how they define themselves. For them, it is a key component of their composite identity, equally as important as other major characteristics such as gender or profession. There are also some for whom the stuttering identity is dominant over all others. It becomes the lens through which they view and engage with the world. They are a "Stutterer" first and foremost; everything else is secondary.

The idea that several identities form one's self parallels a theory of personality called Personal Construct Psychology (PCP) (Kelly, 1955). The general premise of PCP is that "a person always chooses in that direction which he anticipates will increase the total meaning and significance of his life." (Hinkle, 1965, p. 21). Basically, the theory says that we learn to construe the world in certain ways and then we behave in ways that are in line with our constructs; this helps us anticipate future events, making life more predictable and less uncertain. Some theorists have sought to apply this psychological system to stuttering (e.g., Fransella, 1972). The supposition is that, if a person has stuttered since early childhood (as is the case for most people with developmental stuttering), that person will never have known what it means to be a normally fluent speaker. For that individual, stuttering is more meaningful than fluency. Or, to put it another way, the identity of "Stutterer" carries more meaning and significance than the identity of "Normally Fluent Speaker." It is not hard to see how this perspective would inform a clinician's approach to therapy--as the client worked on decreasing stuttering and increasing fluency, he or she would also need to learn to construe what it means to be a normally fluent speaker.

Many PWS have the same reaction when they first learn tools to control their stuttering: it is remarkably easy to do. Modifying motor aspects of speech such as rate, voice, breathing, and articulation is merely a matter of learning slightly different ways to use one's speech mechanism. Maybe a little slower, maybe a little more movement, maybe a pause here or a gentle onset there, but the actual techniques themselves are not that difficult to learn in a short amount of time. The assumption is, once those skills have been taught, the extent to which clients choose to use them in everyday life is their choice. The message, implicitly or explicitly stated, is this: "You've been taught the skills, now it's your choice whether or not you use them."

For some PWS, I don't think it's quite that simple.

Such an assumption ignores the strong effects that identity, familiarity, and safety play in the process of change. Let's take the examples of three distinct but, to my mind, comparable problems for which someone might seek assistance: substance abuse, domestic violence, and obesity. Nobody enjoys being an alcoholic, staying in an abusive relationship, or walking around as a morbidly overweight person. However, many individuals continue in those roles for years despite numerous attempts to change. They are frequently a puzzle to even themselves, somehow "finding themselves" engaged in self-destructive or undesirable patterns of behavior that cause distress and lower their quality of life. During therapy, they learn new ways of thinking and behaving that, when implemented, allow them to see healthier, alternative options. At that point, it would be quite easy (not to mention convenient) for a therapist to send them out the door, new skills in hand, with a parting "Now that you know how to make better choices, it's your decision whether or not you do so. Good luck!"

Something seems wrong with that equation, as if some numbers were missing. What is missing, I believe, is an acknowledgement that changing a way of life--and by extension changing a core piece of one's identity--requires more than simply learning what to do differently. One also must discover why one has continued to engage in undesirable behaviors, as well as understand the significance and implications of changing those behaviors.

To complete the analogy, a therapist counseling an addict seeking to get sober, a woman trying to leave an abusive relationship, or a man wanting to lose 200 lbs. would not simply tell them that it is their choice to use or not use what they have learned during therapy. While those skills were being taught, the therapist would have helped them explore why they remained stuck in their present states for so long. She would have helped them discover what steps needed to be taken to break their behavior patterns. She also would have assisted them through the process of assigning meaning to their new identities as sober, single, normal weight persons. Addressing the reasons behind old behaviors and exploring the possible consequences of new behaviors would give those new identities significance. It would help make the process of change less daunting. It would also provide a bedrock so that new behaviors will last.

Several years ago I came across an interesting article on the Stuttering Homepage (http://www.stutteringhomepage.com) and filed it away in memory. Recently, I stumbled upon the same article while looking through old boxes at a speech clinic. The article was written by Herb Goldberg, the inventor of the Edinburgh Masker, an auditory feedback device from the 1970's. In it, he speaks of suddenly becoming fluent through use of the device and his trepidation at "entering into a fluent lifestyle." He recalls the jarring trauma of leaving stuttering behind to live in the world as a fluent speaker. Here is an excerpt:

"It is important for a therapy program to effectively deal with a component of the disability that can be overlooked--to help prepare the client to really want to accept the responsibilities of fluency. Therapy programs do not generally give overnight fluency that the client knows will last beyond any doubt. While in the program, a client has the time to 'weasel out' of accepting fluency without being aware that he or she is doing so. A client can unconsciously resist the therapy and 'save face' when the real 'bed rock' reason for resistance could be that the person is fearful of entering into, and assuming the responsibilities of, a fluent lifestyle."

So, let's now return to the hypothetical case from the beginning. This person is just leaving therapy excited and hopeful for his newfound fluency. He has been given the tools to control his stuttering; all he has to do is choose to use them.

Or so it would seem.

• Has he been given opportunities to fully examine what change will mean?
• Has he been helped to explore what he might be giving up should he commit to speaking more fluently? For example, people may not listen to him as intently, he will not be able to blame his stuttering for failures or disappointments, he will have to face the initial awkwardness and uncertainty of living in the world and being treated as a fluent speaker.
• Has he been assisted with "trying on" his new identity and encouraged to discuss any feelings of resistance, ambivalence, or discomfort that may arise?

Craig, A., Feyer, A.M., & Andrews, G. (1987). An overview of behavioural treatment for stuttering. Australian Psychologist, 22, 53-62.

Dayalu, V. N, & Kalinowski, J. (2001). Stuttering therapy results in pseudofluency. International Journal of Language & Communication Disorders, 36(3), 405-408.

Fransella, F. (1972). Personal change and reconstruction: Research on a treatment of stuttering. London: Academic Press.

Goldberg, H. (1986). What would you do if you woke up tomorrow morning perfectly fluent and you knew it would last beyond any doubt? Chicago: Foundation for Fluency, Inc. Retrieved from: http://www.mnsu.edu/comdis/kuster/PWSspeak/goldberg.html.

Hinkle, D. N. (1965). The change of personal constructs from the viewpoint of a theory of implications. (Unpublished doctoral dissertation). Ohio State University, Columbus, OH.

Kelly, G.A. (1955). The psychology of personal constructs, 2 volumes. New York: Norton.

National Stuttering Association (2009). The experience of people who stutter: A survey by the National Stuttering Association. Retrieved from http://www.nsastutter.org/opencms/export/sites/default/nsa/stutteringInformation/pdfs/NSAsurveyMay09.pdf

You can post Questions/comments about the above paper to the author before October 22, 2012.