Abstract
This chapter is principally a synopsis of the research into language factors in stuttering that was published subsequent to the S.F. Brown series. A few of these studies were reported within a short time of Brown’s last article (Brown, 1945), but the majority did not begin to appear until almost two decades later.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Footnotes
This study by Van Riper and Hull is hardly ever mentioned in the extensive literature on stuttering “adaptation” although actually it was in this study that the “adaptation effect” was first noted. The publication by Johnson and Knott (1937) is acknowledged, erroneously, as reporting the first study dealing with that phenomenon. The adaptation literature, following Johnson’s bias, has carried the unwarranted assumption that adaptation reflects learning (see Wingate, 1986a, 1986b). Van Riper’s account of this phenomenon, that he discovered, was quite different.
It is of some interest that this article appeared in the Journal of Speech Disorders in the pages immediately sequential to the final article in the Brown series (Brown, 1945).
Whether length is measured in letters, phones, or syllables.
The data were obtained in conjunction with another study that required the subjects to talk continuously, except for rest periods, 10 hours a day for 5 consecutive days. The speaking was essentially soliloquy, although the subjects were most likely aware that the experimenter, and perhaps others, were probably listening. These unusual circumstances do not seem to materially qualify the fact that the samples were self-formulated speech. The total corpus consisted of 248,806 words, of which 17,143 were stuttered.
The frequency distribution of initial phones in his random sample of the total corpus (Hejna, 1955, table 29) correlates.90 with a composite ranking of word-initial phones based on several extensive studies of normal speech (see Chapter 5, Table 5.7).
Word frequency values are based on large samples of written English.
Although “information load” undoubtedly had appeal from the standpoint that it seemed to offer some quantification of “meaning,” it seems likely that a substantial dimension of its appeal lay in it being semantically captivating.
Even though Schlesinger et al. (1965) related sentence position, grammatical class, and word length to predictability in their introduction
To justify her contention that consonants are unpredictable, Taylor cited Sumby and Pollack (1954) as having reported that “the uncertainty of prediction is about 1.5 bits greater for the initial letter than for the mean of all the remaining letters in monosyllabic words.” (The italics are mine.) Stutters relate to phones rather than letters. At the same time, similar findings are reported for phones (Bruner & O’Dowd, 1958; Miller, 1963). However, this emphasis on the initial phone misses the point that the stutterer is not having trouble with (or “on”) the initial phone. (See Chapter 1, pp. 11 ff.)
But no effect for “difficult” versus “easy” consonants, which were also incorporated into the design.
Possibly the information load concept had already biased the reading of Brown’s account.
One should take note here of a crucial difference between word meaningfulness and utterance meaningfulness.
This ratio is closer to the usual overall distribution of content and function words. See Chapter 5.
A good description of Kannada is presented in Schiffman (1983). See also McCormack (1966).
This percentage ratio of content words to function words is comparable to their usual overall distribution in the language. See pertinent data and discussion in Chapter 5.
It is appropriate to note that in Brown’s own data a careful separation of the grammatical classes in terms of actual percentage-of-stuttering values would place adverbs with function words. See Table 3.3.
Note that the grammatical factor is equated with content words, which are presumed here to be the words “crucial for the meaning” and therefore, presumably, are evaluated as important, difficult, requiring special effort, likely to be stuttered, etc. This account is a restatement of the Johnson conjecture, implicit in Brown’s work.
It is of considerable interest that four other children “were eliminated as subjects because their recorded speech was unintelligible or did not contain an arbitrary minimum of five nonfluent words.” In itself this statement makes a cogent commentary on the claim that all children are disfluent, and supposedly in about the same proportion.
Once again word repetitions were counted as instances of stuttering, a distorting and misleading practice.
”Cycles” does not necessarily imply a regularity of recurrence.
Said to resemble Hockett’s (1955) macrosegment, “a unit commonly considered to exist between two terminal juncture points” (p. 854).
In addition to these two reservations, it is also curious that: (1) there is a preponderance of verbalization in the postverbal phrases, (2) more content words occur in these phrases, and (3) there is a higher correlation between function words and “stuttering” in all three units.
Possibly the difference may have been minimized by some amount of familiarity and practice (i.e., adaptation). In this case randomization may have operated to confound rather than objectify.
As found also in his other publications reviewed here, Bloodstein is not alone in this claim. There are many others who subscribe to this position, notably E. M. Silverman, F. H. Silverman, and D. E. Williams. However, Bloodstein has made the greatest issue of the purported difference in childhood stuttering.
See the relevant tables in Chapter 2 (Tables 2.4 through 2.10). Why they occur so much more frequently in the speech of young stutterers is a question that remains to be answered.
Because of the issue being drawn here, it seemed appropriate to report this information regarding position at this point rather than deferring it to the section on Position.
Italics mine. This statement indicates that word and phrase repetitions are also “types of stutterings.”
A constraint imposed by constituent structure analysis which, possibly defensible for purposes of some kinds of study, creates an artificiality and clear bias relative to the study of spontaneous speech.
Italics mine.
The salient features of Kannada are described earlier in this chapter. 31The 1955 publication is an expanded version of the 1942 report. 32Forty two of the 46 children evidenced “brief repetitions;” the other four youngsters evidenced prolongations at these loci. 33Cited by Rieber and Wollock (1977, p. 7).
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1988 Marcel E. Wingate
About this chapter
Cite this chapter
Wingate, M.E. (1988). Corroboration, Extension, Complication. In: The Structure of Stuttering. Springer, New York, NY. https://doi.org/10.1007/978-1-4615-9664-6_4
Download citation
DOI: https://doi.org/10.1007/978-1-4615-9664-6_4
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4615-9666-0
Online ISBN: 978-1-4615-9664-6
eBook Packages: Springer Book Archive