Abstract In this paper, I present an analysis of the pluralization of haber ‘there is/are’ in Puerto Rican Spanish (e.g., habían fiestas ‘there were parties’) as an ongoing language change from below in which the impersonal argument-structure construction (<AdvPhaberObj>) is being replaced by a personal variant (<AdvPhaberSubj>). Speakers pluralize presentational haber in about 41% of the cases, and linguistic conditioning factors are ‘typical action chain-position of the noun's referent,’ polarity of the clause, verb tense, comprehension-to-production priming, and production-to-production priming. I argue that the effect of these independent variables can be traced back to three cognitive factors: the preference for unmarked coding, statistical preemption, and structural priming. Social distributions can also be modeled in constructionist frameworks, with results for social class, formality, and gender advocating in favor of considering this variation as an ongoing change from below, which allows speakers to position themselves in terms of gender and social class.

Abstract It is common practice in sociophonetics to measure vowel formants at one (monophthongs) or two (diphthongs) time points. This paper compares this traditional target approach with two dynamic approaches for investigating regional patterns of variation: the multiple time point approach, which measures formants at successive time points, and the regression approach, which estimates formant dynamics over time by fitting polynomial regression equations to formant contours. The speech material consisted of monosyllabic words containing all full vowels of Dutch, except for /y/. These words were read out by 160 speakers of Standard Dutch, who were distributed over four regions in the Netherlands and four regions in Flanders, Belgium. The results show that dynamic approaches outperform the target approach in uncovering regional vowel differences, which suggests that sociophonetic vowel studies that apply the target approach run the risk of overlooking important sociolinguistic patterns.

An abstract is not available for this content so a preview has been provided. As you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

Abstract Several western Swiss German dialects roughly grouped around the nation's capital Bern show /l/ > [u] vocalization in various contexts. The spatial boundaries of /l/-vocalization in Swiss German are suspected to have been expanding since being described in theLinguistic Atlas of German-Speaking Switzerlandin the middle of the 20th century. The present study assesses the overall expansion of /l/-vocalization by means of a rapid anonymous survey in 20 urban regional centers situated just beyond the traditional boundaries of /l/-vocalization highlighted by theAtlas. Results show that the expansion of /l/-vocalization mainly progresses in southeasterly, southerly, and westerly directions, but with much less success to the north and northwest, where the equally influential dialectal areas of Basel and Zürich seem to exert opposing influences. Further analysis of the data indicates that somewhat differing constraint hierarchies are at work in the different places to which vocalization has diffused.

An abstract is not available for this content so a preview has been provided. As you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

Abstract In many rural English-speaking communities, linguistic processes such as paradigm leveling come into direct conflict with social processes of standardization. In the US region of Appalachia, an analysis of past be leveling illustrates the progression of the sociolinguistic clashes between these forces. A quantitative sociolinguistic examination of leveled was (e.g., We was there) for 67 native Appalachian speakers was conducted to assess the status of past be in light of economic and educational improvements over the 20th century. The results indicate that leveled was declined sharply, with the youngest speakers demonstrating more standardized patterns. Although the overall rate of was leveling declined across apparent time, the rate of was contraction (e.g., We's there last night) increased, offering native Appalachians a reduced variant to contest the social push toward a fully standardized system.

Abstract Four vowel variables are investigated in the speech of 50 rural, nonmigrant children and adolescents aged 6 to 18 years in the Syrian village Oyoun Al-Wadi. Contrary to previous studies, the children initially acquire the urban and later the rural forms of these vowels. This process of acquisition shows the following. First, children show reversal in the acquisition of forms, and thus the rules associated with them; instead of acquiring the local forms first, they acquire the supralocal forms. Second, children can acquire the rules of the second dialect after the age of 8. Third, gender emerges as significant as boys and girls exhibit different linguistic behavior. Fourth, preadolescents emerge as the age group that is most active in building a social identity. Fifth, social-psychological factors motivate the shift from urban to rural forms. These include local identity, gendered social meanings of the variants, and the growth of the children's sociolinguistic competence throughout their preadolescence and adolescence.

An abstract is not available for this content so a preview has been provided. As you have access to this content, a full PDF is available via the ‘Save PDF’ action button.

Abstract The possibility of referential null subjects in Old English has been the subject of conflicting assertions. Hulk and van Kemenade (1995:245) stated that “the phenomenon of referentialpro-drop does not exist in Old English,” but van Gelderen (2000:137) claimed that “Old English has pro-drop.” This paper presents a systematic quantitative investigation of referential null subjects in Old English, drawing on the York-Toronto-Helsinki Parsed Corpus of Old English Prose (YCOE; Taylor, Warner, Pintzuk, & Beths, 2003) and the York-Helsinki Parsed Corpus of Old English Poetry (YCOEP; Pintzuk & Plug, 2001). The results indicate substantial variation between texts. In those texts that systematically exhibit null subjects, these are much rarer in subordinate clauses, with first- and second-person null subjects also being rare. I argue that the theory of identification of null subjects by rich verbal agreement is not sufficient to explain the Old English phenomenon, and instead I develop an account based on Holmberg's (2010) analysis of partial null subject languages.

Abstract An important account of linear ordering in syntax is John A. Hawkins' (2004) theory of cognitive efficiency and the principles of domain minimization formulated therein. In its latest formulation, the theory postulates syntactic and semantic minimization principles. With regard to the relative strength of these principles, prior research into the dynamics of these constraints has come to differing conclusions. Using the relative ordering of prepositional phrases (PPs) in English as a test phenomenon, the present study contributes to the further development of a theory of syntactic serialization through the multifactorial analysis of naturalistic data from a corpus of present-day British English. We find that lexical-semantic dependency constitutes the strongest constraint on serialization followed by the weight-related, syntactic one. More specifically, our results show that although syntactic minimization has much greater data coverage – it applies to a much larger proportion of the data – the lexical-semantic factor has a much greater effect size, thus is more seldomly violated. In addition to assessing the relative importance of the two minimization principles, we also investigate the effects of other potential codeterminants of PP order, namely themanner > place > timegeneralization and pragmatic information status. Our results suggest that these play statistically significant but tangential roles in PP ordering.