This article revisits the etymology of the Dutch third-person plural pronoun hullie similar to hulle, a form of particular interest because of its presence in Afrikaans as well as in American diaspora varieties of Dutch. Although hullie is traditionally analyzed as a contraction of hun (the main third-person plural nonsubject pronoun in standard Dutch) and the noun lie(den) similar to lui 'people', this study argues that its first syllable actually traces back to heur, not hun. The argument is based on a range of historical-linguistic evidence, drawing on both well-known and lesser-known dialect sources. The article furthermore highlights a broader methodological issue: Treating hullie as a reflex of hun has led to incorrect generalizations in the Dutch dialectological literature, artificially inflating the historical distribution of hun at the expense of heur. The findings thus not only resolve a long-standing etymological puzzle, but also underscore the importance of critical etymology for accurate dialect geography and historical-linguistic reconstruction.
Abstract What does it take for a variety to be evaluated as “standard”? Focusing on Flanders, a community with a strict standard language ideology, we examined whether a regionally neutral accent suffices, or whether nonstandard lexis and grammar must also be avoided. In a speaker evaluation experiment, 614 respondents rated four manipulated versions of an official government video, varying in accent (neutral vs. regional) and the presence or absence of tussentaal , a stigmatized nonstandard variety. While unaccented standard Dutch remained the most prestigious variety, a mixed guise combining a neutral accent with tussentaal was not downgraded and was perceived as dynamically prestigious. These findings suggest that a polished accent is a key criterion for standardness, which can offset the negative evaluation of other nonstandard features.*
Abstract This article investigates gender and number agreement on definite articles, adjectives, and pronouns in one Pennsylvania Dutch variety spoken by the Amish of Lancaster County, Pennsylvania. The findings, based on eight speakers, suggest that the nominal system in this variety is undergoing change. In this study, gender and number are inconsistently marked. The historical neuter definite article es has become quite rare. The feminine and plural definite articles are frequently indistinguishable from the masculine. Gender and number marking on adjective inflections is better preserved although variation occurs. Finally, pronominal reference is less conditioned by grammatical gender agreement and more conditioned by semantic features than might be expected based on the German heritage of Pennsylvania Dutch. We conclude that these changes are impacted both by language-internal developments and influence from English.
We focus on the use of subject clitics and subject pronouns in speakers of linguistic islands in northern Italy, with an Upper German dialect as L1 and Italian as the surrounding majority language and L2. German requires an explicit subject pronoun or clitic, while in Italian the explicit subject is typically omitted. Further contact varieties such as Friulian, Piedmontese, and Francoproven & ccedil;al display subject clitics. In contrast to previous studies, which were limited to one or two speech communities (of the same idiom) (see e.g. Z & uuml;rrer , Dal Negro , Kolmer , Bidese & Tomaselli ), we compare three speech communities with two different idioms: Alemannic Walser varieties spoken in Gressoney and Issime and Zahre German, a Bavarian idiom spoken in Sauris. Our findings show that linguistic islands in northern Italy of both Bavarian as well as Alemannic origin develop in the same direction. Moreover, Issime and Gressoney differ notably from each other: Gressoney German indeed in some respects displays greater similarities to Zahre German, probably due to its different language contact situation as compared to Issime German. Finally, the grammaticalization status of the clitic is more advanced in Zahre German than in Walser German (at least than in Issime German).
Abstract This study examines how Uniform Information Density and the lexical verb position, predicting valency knowledge and disambiguation, influence the order of nominal dative and accusative objects in German from the fifteenth to twentieth century using corpus data and regression analysis. As a measure of sentence uniformity, DORM (Deviation of the Rolling Mean, Cuskley et al. 2021) is applied on unigram lemma surprisal values of each clause containing dative and accusative objects. The verb position is determined using the POS-information from the corpora. Further factors included in the analysis are the givenness status of the two objects, their length difference, and the period in which the texts were published. In historical data and with uneven information profiles, Acc>Dat is expectedly more frequent in SVO while Dat>Acc is more common in SOV. In modern German, Dat>Acc is dominant independent of the verb position with uneven information profiles. The known influence of givenness is also attested. A balance between the frequency of grammatical construction and the lexical uniformity of the information profile can explain the results.*
