Sabine Lappe

Testing hypotheses about compound stress assignment in English: a corpus-based investigation

Abstract This paper tests three factors that have been held to be responsible for the variable stress behavior of noun-noun constructs in English: argument structure, semantics, and analogy. In a large-scale investigation of some 4500 compounds extracted from the CELEX lexical database (Baayen et al. 1995), we show that traditional claims about noun-noun stress cannot be upheld. Argument structure plays a role only with synthetic compounds ending in the agentive suffix -er. The semantic categories and relations assumed in the literature to trigger rightward stress do not show the expected effects. As an alternative to the rule-based approaches, the data were modeled computationally and probabilistically using a memory-based analogical algorithm (TiMBL 5.1) and logistic regression, respectively. It turns out that probabilistic models and the analogical algorithm are more successful in predicting stress assignment correctly than any of the rules proposed in the literature. Furthermore, the results of the analogical modeling suggest that the left and right constituent are the most important factor in compound stress assignment. This is in line with recent findings on the semi-regular behavior of compounds in other languages.

Monosyllabicity in prosodic morphology: the case of truncated personal names in English

The present study is concerned with the structure of truncated personal names in English. Truncations have received a lot of attention in recent years. Previously judged a highly irregular and idiosyncratic class of word-formation processes, truncatory processes have now become an important test case for the research program of prosodic morphology (McCarthy & Prince 1986, 1993). The structure of truncated names in various languages has been shown to be not idiosyncratic at all; rather, truncated names are exponents of a word-formation process whose output is to a very large extent determined by the phonology and the prosody of the language. The data provided in (1a-c) show examples of name truncations in German, Spanish and French, for which detailed analyses have been proposed within the framework of prosodic morphology (Fery 1997, Ito & Mester 1997, Wiese 2001 for German, Pineros 1998, 1999, 2000 for Spanish, Scullen 1997, Nelson 1998 for French; cf. also Weeda 1992 for a survey of truncatory processes in various languages, Mester 1990, Benua 1995 for Japanese). (1d) illustrates the English process to be studied here.

English Prosodic Morphology: Prosody

Whereas the discussion in the previous chapter focussed on morphological aspects, the present chapter will review in detail previous approaches to the phonological and, in particular, the prosodic characteristics of truncatory processes in English. Based on the desiderata, both empirical and theoretical, that will be shown to characterise the state of research, I will provide an outline of the central proposal of this book.

Constraint Interaction: Syllable Structure

Traditionally, the structure of truncated forms has mainly been described in terms of the number of syllables they contain. However, the investigation of the structural aspects of truncated forms in chapters 4–6 has shown that the structural characteristics of truncated forms comprise more than the mere number of syllables.

The Structure of Monosyllabic Truncated Names

This chapter will present an in-depth analysis of the structural properties of monosyllabic truncated names and a comparison of these properties with the structural makeup of th.eir bases. Its structure is as follows: We will first be concerned with the structural properties of truncated names. Section 4.2 will describe these properties in terms of basic syllable structure, whereas section 4.3 will focus on phonotactic restrictions on word-final consonant clusters. The second part of the analysis will then deal with the relationship between truncated forms and their bases. Section 4.4 will provide a comparison of truncated forms and their bases in terms of their segmental makeup, investigating in how far truncated forms faithfully copy correspondent segments. Section 4.5 will then investigate the question of which segmental material of the base form survives in the truncated form. A summary of the findings from this chapter will be provided in section 4.6.

Constraint Interaction: Cluster Phonotactics

This chapter will be concerned with restrictions on consonant clusters in truncated forms. The truncatory patterns to be analysed are monosyllabic truncated forms (names and clippings) and suffixed disyllabic forms. It was shown in chapters 4–6 that consonant clusters do not exhibit a uniform behaviour in truncation. By contrast, we find clusters which are categorically banned alongside those which are categorically retained. In between we find clusters which are truely variable (in the sense that chances that the cluster is retained and chances that it is simplified are approximately even) and clusters which exhibit clear tendencies to be either avoided or retained. To complicate matters further, we find that patterns of preservation differ across truncatory patterns for some clusters, whereas for others they do not. The examples in (1) illustrate this situation for some of the pertinent clusters.

Constraint Interaction: Segmental Changes

This chapter will be concerned with phenomena affecting the segmental stability of truncated forms. The empirical analysis in chapters 4–6 has shown that the overwhelming majority of truncatory outputs faithfully copy the segmental material of their bases, so that no segmental changes occur. Forms which have traditionally been used in the literature to substantiate the alleged formal unpredictability of truncatory outputs have been shown to constitute idiosyncratic forms if seen against the vast number of predictable, segmentally stable forms. Examples of such idiosyncracies from name truncation are Bill (< William, changing [w] to [b]) and Bess (< Elisabeth, changing [θ] to [s]).

Constraint Interaction: Word Structure

The inventory of English Prosodic Morphological processes is illustrated in (1). Most of the examples have been repeated from (1) in chapter 1.

The Structure of Y-Hypocoristics

The aim of this chapter is to provide a description of the structural properties of y-hypocoristics. The relevant corpus is the combined list of base-derivative pairs in Phillips (2002) and Wallace (1992; cf. sections 3.2.2, 3.2.3 for details about the corpora). The focus of the analysis will be on the question whether y-hypocoristics differ in structural terms from their monosyllabic counterparts. Recall from the review of previous analyses in chapter 2 (section 2.3.2) that one position holds that y-hypocoristics are exclusively derived from simple truncated forms. According to this analysis, a form like Mikey is derived from the truncated form Mike, which in turn is derived from the base Michael. A second approach, however, which has been put forward for a parallel process in German (Fery 1997, Wiese 2001), holds that it is not necessary to postulate the formation of a simple truncated form as an intermediate step in the derivation. In this approach, a form like Mikey is immediately derived from the base name Michael. The chief source of evidence used in this debate comes from the phonotactics of intervocalic consonant sequences in y-hypocoristics, which is argued to mirror either word-final phonotactics or intervocalic markedness preferences.

English Prosodic Morphology: Morphology

The aim of this chapter is to introduce and define the class of phenomena that will be analysed in this book. We will adopt the perspective of morphological theory, focussing on two major questions: • What can the research discipline of Prosodic Morphology (McCarthy & Prince 1986, 1993a et seq.) contribute to a definition of truncatory patterns in English? • What is the status of truncated words in the English word formation system?