Word length and the location of primary word stress in Dutch, German, and English *.(Report)

Abstract

This study addresses the extent to which the location of primary stress in Dutch, German, and English monomorphemic words is affected by the syllables preceding the three final syllables. We present analyses of the monomorphemic words in the CELEX lexical database, which showed that penultimate primary stress is less frequent in Dutch and English trisyllabic than quadrisyllabic words. In addition, we discuss paper-and-pencil experiments in which native speakers assigned primary stress to pseudowords. These experiments provided evidence that in all three languages penultimate stress is more likely in quadrisyllabic than in trisyllabic words. We explain this length effect with the preferences in these languages for word-initial stress and for alternating patterns of stressed and unstressed syllables. The experimental data also showed important intra- and interspeaker variation, and they thus form a challenging test case for theories of language variation. (1)

1. Introduction

Most analyses of primary word stress in Dutch, German, and English assume that the location of primary stress is independent of the position of secondary stresses to the left of the primary stress. Thus, in all accounts of primary stress that are rule-based or grounded in the Principles and Parameters framework, the parsing of syllables into feet is from right to left, and primary stress is assigned to the right-most foot. Since the word-final syllable is left extrametric under some conditions, at least at the stage where primary stress is assigned, this results in primary stress on the antepenultimate, the penultimate, or the final syllable (e.g., Hayes 1980; van der Hulst 1984; Kager 1989; Venneman 1990). The footing of the syllables to the left of the three-syllable word-final window thus plays no role in primary stress assignment, as is explicitly stated in van der Hulst and Kooij (1992), who claimed that primary stress assignment and secondary stress assignment are separate algorithms ("main stress first approach").

Also in most analyses framed within Optimality Theory, the position of secondary stress does not affect the position of primary stress (e.g., Nouveau 1994; Fery 1998; Hammond 1999; Pater 2000). The analyses adopt constraints stating that every prosodic word ends in a foot and that primary stress is on the head of this foot. These constraints are ranked high, and may only be violated in order to satisfy constraints on the well- formedness of feet (such as, feet are binary, feet are left-headed, heavy syllables cannot be in the dependent positions of feet) and by constraints forbidding stress on the word-final syllable. Importantly, the constraint forcing all syllables to be parsed into feet is low-ranked. As a consequence, the assumed constraint hierarchies force well-formed feet at the right-edge of the word. The preceding syllables are parsed into feet only if the resulting feet are well-formed.

Germanic languages, as many other languages, tend to have wordinitial (secondary) stress (e.g., Halle and Kenstowicz 1991; Booij 1995; Fery 1998; Trommelen and Zonneveld 1999a: 484; Pater 2000; Zonneveld and Nouveau 2004). Speakers transfer this preference for initial stress in existing words to pseudowords (Baker and Smith 1976), and listeners take advantage of it in speech segmentation (Cutler and Norris 1988) and in phonetic judgments (Slootweg and Rietveld 1989). The preference for initial stresses may overrule other constraints. For instance, the Dutch word ,Jeroso,limi 'tani 'friars of the order of Jerusalem' bears word-initial secondary stress, even though this results in a sequence of two unstressed syllables (Zonneveld and Nouveau 2004: 389), and Germanic languages, like many other languages, prefer alternating patterns of stressed and unstressed syllables (e.g., Prince 1983; Selkirk 1984; Kager 1989; van der Hulst and Kooij 1992; Hung 1994; Wiese 1996). The preference for initial stress is not satisfied, however, if it leads to unwell-formed feet or stress clashes. More importantly, it is claimed that it would not affect the position of primary stress. Thus, Pater (2000) and Zonneveld and Nouveau (2004) state that the constraint forcing words to start with a stressed syllable is satisfied only when this can be done at no expense of the constraint Align (Prosodic Word, Right, Head of Prosodic Word, Right), which forces the head of the prosodic word to be at the word-final syllable.

In the present study, we questioned the independence of primary stress of the footing of the preceding syllables. More specifically, we investigated the role of the preference for word-initial stress in combination with the preference for alternating patterns of stressed and unstressed syllables. For this we compared tri- and quadrisyllabic existing words and pseudowords in Dutch, German, and English. If the location of primary stress is determined only on the basis of the three final syllables, we may expect that the number of syllables in the word is irrelevant. Trisyllabic and quadrisyllabic words ending in the same three syllables would have primary stress on the same syllable (counting from the right edge). If, in contrast, the footing of all syllables in the word would affect the location of primary stress, we may expect a difference between trisyllabic and quadrisyllabic words. The preference for stressed initial syllables in combination with the preference for alternating patterns of stressed and unstressed syllables may favor penultimate primary stress in quadrisyllabic words (,[TEXT NOT REPRODUCIBLE IN ASCII.] with' indicating primary and, indicating secondary word stress). In contrast, trisyllabic words with secondary stress on the initial syllable and primary stress on the penultimate syllable would not show a perfect alternation of stressed and unstressed syllables (,[TEXT NOT REPRODUCIBLE IN ASCII.]), but a stress clash. Such words are predicted to favor initial, antepenultimate, primary stress.