Language and Speech

This study examined the phonological neighborhood characteristics (frequency, density, and neighborhood frequency) of 138 malapropisms. Malapropisms are whole word substitutions that are phonologically, but not semantically, related. A statistical analysis of a speech error corpus suggests that neighborhood density and word frequency differentially affected the number of malapropisms. Specifically, a greater number of malapropisms were found among high frequency words with dense neighborhoods than with sparse neighborhoods. Exactly the opposite pattern was found among low frequency words. That is, more errors were found among low frequency words with sparse neighborhoods than with dense neighborhoods. More malapropisms resided in low frequency neighborhoods than in high. The average word frequency, average neighborhood density, and average neighborhood frequency of the malapropisms were significantly lower than the same averages computed from randomly sampled control words. Finally, more target words were replaced by error words that had relatively higher frequency than by error words that had relatively lower frequency. The implications of these findings for models of lexical representation and processing are discussed.

Previous research has demonstrated that semantics and pragmatics influence durational modifications in words and segments. The present research investigated specifically how semantics and pragmatics influence preservation of a phonemic contrast. Experiment 1 examined alveolar flapping in American English. Potentially flapped words, for example, writer and rider, were embedded in each of two types of semantic passages: semantically biasing and semantically neutral passages. In addition, these passages were produced in one of two pragmatic conditions: listener-present and listener-absent. The results showed that the phonemic voicing distinction between /t/ and /d/ was preserved in biasing passages and in the listener-present condition. The /t/-/d/ distinction was not preserved in neutral passages or in the listener-absent condition. Experiment 2 examined whether listeners could use the durational differences found to distinguish phonemic voicing in Experiment 1. Our investigation demonstrates that semantics and pragmatics interact with phonological processes in speech production. We speculate that phonological processes may be affected both by speakers' decision processes to adjust their articulation for the benefit of the listener and by speakers' internal structure and interactive activation among linguistic units.

This paper offers a contribution to laboratory phonology, extending experimental methods to the investigation of a cross-linguistic historical problem. The issue in question is the recurrent and unidirectional series of sound changes from /p/ > /f/ > /h/. A large sample of languages was surveyed, which revealed no evidence for direct changes from /p/ > /h/ (contrary to popular claims in the literature). Instead, an intervening /f/ stage appears mandatory. /p/ > /f/ changes are widely assumed to be articulatorily-motivated lenitions. By contrast, /f/ > /h/ changes have received little attention. The survey indicates that this change tends to be motivated in the proximity of /u/. A perceptual experiment was carried out to investigate this pattern, which shows that the change may be initiated due to the acoustic similarity of [fu] and [hu] sequences. Some implications of these findings for phonological theory are then briefly outlined.

The importance of vowel duration for specifying vowel contrasts differs across languages. In English, for example, a number of vowel pairs are acoustically differentiated by both temporal and spectral information, whereas in standard French temporal information plays a much more minor role. Gottfried and Beddor (1988) reported that the effectiveness of vowel duration in perception varies accordingly: For native speakers of English, but not native speakers of standard French, a change in vowel duration affected the perceptual identity of a vowel contrast. We tested the hypothesis that the relative prominence of vowel duration within different dialects of a given language also has perceptual consequences. Vowel duration plays a much more important role in the phonological system of Swiss French than standard French. Given this, we predicted that native speakers of Swiss French, unlike native speakers of standard French, would use temporal information when identifying vowels. Our prediction was confirmed. These findings indicate that just as there are cross-language differences in fundamental aspects of speech perception, so too are there cross-dialect differences, and they support the view that the perceptual mapping between acoustic signal and vowel category is sensitive to global aspects of the listener's phonological system.

This study examined interarticulator timing amongst stutterers and non-stutterers from the perspective of extrinsic and intrinsic timing models of speech motor control. Timing relationships were examined for an oral closure task across stress, rate, and vowel differences. Sequential aspects of extrinsic timing theories were analyzed by peak velocity sequence profiles of upper lip, lower lip, and jaw, and nonsequential aspects by sequence period (the time period covering all three of the articulator peak velocity onsets times irrespective of sequencing order). Intrinsic timing theory was modeled using phase portrait analyses of the same data. Consistently greater phase angle variability was seen amongst the stutterers, and variability was found to be significantly greater at a faster articulatory rate and when altered stress patterns were required. No group differences were found for peak velocity sequencing, but the (sequence independent) sequence period data somewhat mirrored the phase angle findings. This finding calls into question the notion that phase angle analyses reflect purely intrinsic timing phenomena.