Carlos Gussenhoven

Laboratory Phonology 7

This collection of recent papers in Laboratory Phonology approaches phonological theory from several different empirical directions. Psycholinguistic research into the perception and production of speech has produced results that challenge current conceptions about phonological structure. Field work studies provide fresh insights into the structure of phonological features, and the phonology-phonetics interface is investigated in phonetic research involving both segments and prosody, while the role of underspecification is put to the test in automatic speech recognition.

Types of Focus in English

Speakers conduct conversations so as to establish a common understanding with their hearers about some aspect of the world. In a discourse model, the speaker keeps track of the development of this common understanding, and labels his linguistic expressions for the way the information they convey relates to the information in the discourse model as developed at that point. In English, pitch accents are used for this purpose. Broadly, their location indicates the size of the ‘focus constituent’, which contains the constituent(s) whose information status is being signalled, while their distribution within the focus constituent expresses the type (or meaning) of the focus. Different focus meanings are distinguished depending on whether the information represents new information or concerns a correction of existing information, on whether the information reflects a change in the world or a change in the hearer’s knowledge about the world, and on whether new knowledge about the world is immediately or only potentially relevant to the hearer. These meanings form a natural class with those proposed earlier for melodic aspects of English intonation.

The English Rhythm Rule as an accent deletion rule

In order to account for the accentual and rhythmical structure of English, a binary-branching prosodic constituent structur is assumed, in which minimally the syllable and the foot must be headed. Feet are potentially marked as accented. This representation makes it possible to describe the prominence patterns of word groups as resulting from three accent deletion rules, the Compound Rule, the Initial Accent Deletion Rule and the Rhythm Rule. It was shown that the structural change effected by Initial Accent Deletion cannot be expressed in theories which represent stress as a relative concept. Moreover, this rule, which like the Compound rule is a lexical rule, provided evidence for the existence of a stratum in the lexical phonology of English in which compounding and so-called ClassII derivation take place. The Rhythm Rule is a postlexical rule, which was shown to apply to the output of the other two rules. Without the aid of any conditions or constraints, it accounted effortlessly for the stress-shift data presented in the recent literature. It could moreover be shown that apparent cases of stress shift in unaccented speech (in which the Rhythm Rule does not apply) should not in fact be viewed as the output of any stress-shift rule at all, but should be explained as the effect of preboundary lengthening as applying to the different constituents in the prosodic hierarchy. It was argued that an analysis of sentence accentuation whereby focused constituents have to be assigned accents can run into problems that do not exist in a ‘deaccenting’ analysis, in which nonfocused constituents are deprived of their accents. Finally, it was argued that English, unlike Dutch, lacks phonological rules that refer to primary word stress, and that, at best, primary stress may reveal itself in low-level timing distinctions.

Emphasis and Tonal Implementation in Standard Chinese

Abstract Despite the greatly improved understanding of tonal articulation in Standard Chinese, no consensus has been reached on the most appropriate model of tonal implementation [Xu, Y., & Wang, Q. (2001). Pitch targets and their realization: Evidence from Mandarin Chinese. Speech Communication, 33, 319–337; Kochanski, G., & Shih, C. (2003). Prosody modeling with soft templates. Speech Communication, 39(3/4), 311–352]. To shed new light on the issue, all four lexical tones, embedded in sentences with different preceding and following tonal contexts, were elicited under corrective focus, with two degrees of emphasis (Emphasis and MoreEmphasis), in addition to a NoEmphasis base-line condition, so as to bring systematic variation in duration and F0 to bear on the issue of tonal realization in different pragmatic contexts. Results showed comparable increases in syllable duration from the NoEmphasis condition to the Emphasis condition and from the latter to the MoreEmphasis condition. F0 range expansion, however, was non-gradual: while there was a substantial increase in the F0 range from the NoEmphasis to the Emphasis condition, the expansion from the Emphasis to the MoreEmphasis condition was marginal. Analyses of the F0 patterns revealed that under emphasis, lexical tones were realized with magnified F0 contours which were adapted to both the neighbouring tones and the durational increase of the tone-bearing syllables, and therefore maximally distinguishable from each other. Implications of these findings on models of tone and focus realization are discussed.

Discreteness and Gradience in Intonational Contrasts

The intonation systems of many languages apparently allow gradient linguistic categories by the side of discrete or categorial contrasts. It is argued that these gradient meanings arise when the intonational phonology leaves a certain amount of unused phonetic space, which may be exploited by speakers to convey “meanings” which are naturally associated with nonlinguistic pitch variation. Because these meanings are often similar to the linguistic meanings attached to the phonological intonational categories, the issue of gradience versus discreteness is more difficult inintonation than in other linguistic domains. Three experimental techniques that can be used to investigate the gradient or discrete nature of intonational differences, the “semantic task,” the “imitation task,” and the “pitch range task,” are discussed and evaluated. In addition, it is pointed out, following an earlier discussion by Newport, that categorical perception is a sufficient, but not a necessary, property of phonological discreteness.

The phonology of tone and intonation in the Dutch dialect of Venlo

The Dutch dialect of Venlo has a lexical tone opposition comparable to the distinction between Accent I and Accent II in Scandinavian. The two word tone patterns are realised in a variety of different ways, depending on the intonation contour, on whether the word has a focus tone, and on whether it occurs finally or nonfinally in the intonational phrase (IP). Twelve such contexts are identified, and an autosegmental-metrical analysis is presented of the contours for the word tones in each of these. The analysis is instructive because of its clear illustration of the distinction between the phonological underlying representation and the phonological surface representation, as well as of the distinction between the latter representation and the phonetic realisation. In addition, because of the complexity of its tonal phonology, the dialect is of considerable typological interest for the study of word prosody and intonation.

Language-specificity in the perception of paralinguistic intonational meaning.

This study examines the perception of paralinguistic intonational meanings deriving from Ohalas Frequency Code (Experiment 1) and Gussenhovens Effort Code (Experiment 2) in British English and Dutch. Native speakers of British English and Dutch listened to a number of stimuli in their native language and judged each stimulus on four semantic scales deriving from these two codes: SELF-CONFIDENT versus NOT SELF-CONFIDENT, FRIENDLY versus NOT FRIENDLY (Frequency Code); SURPRISED versus NOT SURPRISED, and EMPHATIC versus NOT EMPHATIC (Effort Code). The stimuli, which were lexically equivalent across the two languages, differed in pitch contour, pitch register and pitch span in Experiment 1, and in pitch register, peak height, peak alignment and end pitch in Experiment 2. Contrary to the traditional view that the paralinguistic usage of intonation is similar across languages, it was found that British English and Dutch listeners differed considerably in the perception of “confident,” “friendly,” “emphatic,” and “surprised.” The present findings support a theory of paralinguistic meaning based on the universality of biological codes, which however acknowledges a language-specific component in the implementation of these codes.

Word prosodic structure and vowel duration in Dutch

Abstract This paper focuses on the relation between word prosodic structure and vowel duration in Dutch, to the exclusion of phrase-final lengthening and accentual lengthening. Measurements of vowel durations in reiterant speech showed that main stress, secondary stress, and right/left-edge position determine vowel duration. In addition, the experiments made it clear that the durational differences between long and short vowels only surface in syllables with (main or secondary) stress. The observations were summarized in rules which were implemented in a diphone-based Text-To-Speech system (KUN-TTS). The resulting durations showed high correlations with vowel durations measured both in reiterant and lexical words.

The Lexical Tone Contrast of Roermond Dutch in Optimality Theory

The intonational systems of the dialects of Dutch and German in an area covering a large part of the former German Rhineland (the northern half of Rhineland-Palatinate and the southern half of North Rhine-Westphalia), Luxemburg, the northeast of Belgium, and the southeast of the Netherlands resemble those of Norwegian and Swedish: in addition to the tones contributed by the intonation, there is an opposition between two tonal word accents.* The purpose of this chapter is to give an account of the way in which the lexical tone contrast in one of these dialects, that of the city of Roermond in the Dutch province of Limburg, is realised under different intonational conditions.1 The intonation tones with which the lexical tones combine should be divided into tones that mark the focus of the sentence, which appear in the syllable with primary stress in focused words, and tones that signal discoursal meanings, which appear at the boundaries of intonation phrases. Bruce (1977) showed that by making the appropriate comparisons, the lexical tones can be separated from both types of postlexical (intonational) tones: “By comparing F0-contours of words in final and non-final position, in and out of focus and with contrasting word accent out of focus, the individual F0-contributions of terminal juncture [sc. boundary tones, C.G.], sentence accent [sc. focus-marking tone, C.G.] and word accent (accent 1 and accent II) could be isolated.” (Bruce 1977: 37).

Preaccentual Pitch and Speaker Attitude in Dutch.

Native speaker reactions to high-pitched and low-pitched pronunciations of the unstressed syllables before the first pitch accent in an utterance—or high and low “preheads” respectively—show that their pragmatic effect depends on the initial pitch of the following pitch accent. The results were obtained in one specific conversational context, in which the intonation contours appeared on utterances that offered a solution to some problem, and with the help of a limited set of semantic scales. Listeners gave more favorable judgements to high preheads than to low preheads if the following pitch accent began low, but when the following pitch accent began high, they gave more favorable judgements to low preheads than to high preheads. Current theories of intonational structure do not predict this kind of interaction. In these theories, the preaccentual pitch is interpreted as a separate morpheme from the following pitch accent, which suggests that the semantic contribution of the preaccentual pitch is independent of the various pitch accents with which it combines. However, these theories are not ruled out by the results, which can be accommodated if the theories are interpreted appropriately.