Language and Speech

The experiments reported in the present paper were set up in order to restrict some of the syntactic and semantic criteria involved in the task of tonic identification. Experiment 2 examines tonic identification in a particular type of contrastive sentence and in cleft sentences. Experiments 3-5 examine the effect of the discourse on tonic identification. The results indicate that when the information focus is a contrasted element that contradicts a previous element, the information focus is identified consistently as tonic. When the information focus is a "new" element realized as the leftmost lexical item in a sentence, it is normally identified as the tonic, but if judges are asked to identify tonics in utterances that do not have a clear "given/new" structure, then several tonics are identified. The results indicate, therefore, that the tone group in its unmarked form will contain several peaks of prominence.

The study of dirty-word descriptions extends our knowledge of semantic, comprehension, and contextual language processes. Two studies were conducted to examine how we interpret and react to dirty-word descriptions. Subjects were asked to judge how much they would like a fictitious person described with dirty and non-dirty adjective pairs. Liking was significantly influenced by (a) semantic interpretation (connotation v. denotation), (b) the intrinsicalness of the adjective to the person described (prenominal adjective order), and (c) the contextual relation between the speaker and the listener (friend v. enemy).

We have investigated the movements of the epiglottis in speech by speakers of Hebrew. The epiglottis functions as an articulator in the production of pharyngeal consonants and in the vowel /a/. It is also involved in whisper. In pharyngeals, the epiglottis articulates against the posterior pharyngeal wall; the constriction varies from a full closure (pharyngeal stop) in some examples of /?/ in slow careful speech, through a narrow opening for the fricatives /h/ and other examples of /?/ in connected speech, to a fairly open glide in yet other examples of /?/ in rapid or casual style. In the vowel /a/, the opening between the pharynx and the epiglottis is of the same general shape as for the pharyngeal consonants, but larger. In the case of both consonants and /a/, however, the size of the opening is never large enough to pull the lateral edges of the epiglottis away from the posterior pharyngeal wall. That is, air passes only through a passage bounded by the epiglottis and the pharynx, and never escapes along the tongue root lateral to the epiglottis. The epiglottis folds toward the pharyngeal wall independently of the tongue root in consonants; this independence is seen in some cases of /a/ and not others; in whisper, the epiglottis is generally more retracted than in normal speech but otherwise moves as usual. Our conclusions are based on 100 minutes of videotape made using a fiberscope positioned in the upper pharynx, spectrograms, and dissections. We conclude that the epiglottis is an active, independent, and acoustically significant articulator.

A source of regularity in sentence construction is the recurrent use of certain fixed syntactic formats in formulating explanations, descriptions, etc. It is demonstrated that subjects exploit these regularities in sentence perception. They assimilate some of the features of the sentence to formulation frames, rather than determining its actual morpho-syntactic structure, and do not normally check the conformity of the sentence to the presumed format. This is shown by an experiment on perceiving "perverse" sentences, which deviate from the normal formats so that normal relationships do not obtain. It is further shown that expectations raised by the context influence the process of sentence analysis and interpretation.

This paper reports a preliminary investigation into the pitch correlates of politeness formulae produced by Japanese and English informants of both sexes. Significant pitch differences in the expression of politeness occur between the two language communities: The Japanese female subjects adopt an extremely high pitch clearly separating themselves acoustically from Japanese males in the same circumstances, while English male pitch is considerably less differentiated from English female pitch. These pitch differences arise, it is suggested, because of a contrasting sociosemiotic function assigned to pitch in the two speech communities whereby Japanese high pitch is typically reserved for the enactment of female roles while English high pitch is adopted by both sexes to express politeness.

A review of relevant literature indicates a gap in the available data on the normal acquisition of the animate third person personal pronouns of English. In this study a comprehension task (picture verification) employing this class of words was run on 100 children from three to seven years of age. The results show that comprehension of these forms beyond chance level first appears at age five with continuing improvement through ages six and seven. Separate analyses of the comprehension of the semantic components of the forms — i.e., Gender, Number, and Case - suggest differential acquisitional curves for these aspects, with mastery of the Gender distinction appearing to lead the other two.

Strong evidence was obtained that coding operations are variable and efficient during visual word recognition. Subjects were asked to make lexical judgments about words and nonwords when phonological recoding was either beneficial or detrimental to making their decisions. It was found that subjects utilized a phonological strategy where it was helpful and abandoned that strategy where it was a hindrance. The results are interpreted as providing support for a dual retrieval model of word recognition.