Almost all existing accounts of the change of the 2nd person singular verbal agreement ending -s to -st in Old English and Old High German attribute the development to some combination of reanalysis of forms with enclitic subject pronouns ending in -stu and analogy based on a handful of mainly preterite-present verbs that already had 2nd person singular -st in the present indicative in Proto-(West)-Germanic. These factors retain a role in the analysis presented here, but I argue that the operative mechanisms are essentially phonological in nature, licensed ultimately by certain neutralizing processes such as degemination, rather than involving the spread of an existing -st morpheme, grammaticalization of an enclitic subject pronoun, or relocation of a morphological boundary. This analysis also sheds light on the relationship of the 2nd person singular change to the more general phenomenon of word-final t accretion seen in dozens of words such as German Axt ‘ax’ < Middle High German ackes or English against < earlier agains.*

This paper concerns the superlative forms of the words many, much, few, and little, and their equivalents in German, Dutch, Swedish, Norwegian, Danish, Dalecarlian, Icelandic, and Faroese. It demon-strates that every possible relationship between definiteness marking and interpretation is attested. It also demonstrates that different kinds of agreement mismatches are found under relative and proportional readings. One consistent pattern is that under a relative interpretation, quantity superlatives with adverbial morphology show neuter singular agreement even if the target noun is plural. In contrast, under a proportional interpretation, quantity superlatives always agree in number. This evidence is taken to show that quantity superlatives are not structurally analogous to quality superlatives such as tallest on either a relative or a proportional reading; however, depending on their interpretation, quantity superlatives depart from a plain attributive structure in different ways. On relative readings, they can have a structure akin to that of pseudo-partitives (as in two liters of milk), while on proportional readings, they tend to have a quantificational structure, sometimes involving a true partitive (as in some of the children). Furthermore, I suggest that the agreement features of a quantity superlative depend on the domain from which the target is drawn (the Target-Domain Hypothesis).

This paper investigates the use of modal particles in spoken Dutch imperatives. Two types of particles are differentiated: mitigating, which are often used as a politeness strategy, and reinforcing, which add extra force to the utterance (Vismans 1994). Our findings show that in Netherlandic Dutch, the use of mitigating particles is determined by the type of occupation that the speaker has: Speakers in service-oriented occupations use mitigating particles significantly more often than speakers in nonservice-oriented occupations, and it is argued that this is a function of their need to be more polite in their role as a service provider. Since the data do not come from the speakers’ workplace interactions but from informal conversations with friends and family, it is suggested that speech patterns of speakers’ professional and private language practices influence each other. The effect of occupation is not observed in Belgian Dutch, however, where mitigating particles are significantly less frequent. Moreover, an important methodological consideration arises from this analysis: There is the need for researchers to examine the data beyond the standard sociolinguistic categorizations made available by large corpora.

Resultatives in English and Dutch have developed special degree readings. These readings stem from a reinterpretation of the resultative predicate as indicating a high degree rather than an actual result. For example, when a parent says I love you to death, one need not call the cops, since the sentence is not about love turning lethal, but merely indicative of a high degree of affection. Such cases have often been noted in the literature as idiomatic, but this view ignores the fact that these are not isolated cases but productive constructions that can be used with a variety of verbs. We explore various resultative constructions in English and Dutch, and give a classification of the subtypes involved as well as their diachronic development from ordinary to degree interpretation. We link these subtypes to lexical semantic classes of verbs. Both English and Dutch show a steady growth in the lexical and structural diversity of degree resultatives throughout the early modern and contemporary periods (1600-2000). We focus in our paper on the period 1800-2000, for which we did an extensive corpus study using the Corpus of Historical American English (COHA) and Delpher (a collection of digitized Dutch newspapers, journals, magazines, and other resources). One of our findings is that, similar to other types of expressive language, such as degree modification and emphatic negation, taboo expressions play a role in degree resultatives; in fact, their role is excessive. We outline a number of the commonalities among the semantic domains of expressive language used in resultatives.

This article undertakes the first systematic examination of Frank’s (1979, 1981, 1987, 1990, 2007b, 2008) claim that Old Norse influence is discernible in the language of Beowulf. It tests this hypothesis first by scrutinizing each of the alleged Nordicisms in Beowulf, then by discussing various theoretical considerations bearing on its plausibility. We demonstrate that the syntactic, morphological, lexical, and semantic peculiarities that Frank would explain as manifestations of Old Norse influence are more economically and holistically explained as consequences of archaic composition. We then demonstrate that advances in the study of Anglo-Scandinavian language contact provide strong reasons to doubt that Old Norse could have influenced Beowulf in the manner that Frank has proposed. We conclude that Beowulf is entirely devoid of Old Norse influence and that it was probably composed ca. 700, long before the onset of the Viking Age.

This paper explores V2 variation in West Flemish and French Flemish dialects of Dutch based on an extensive corpus of authentic spoken data. After taking stock of the existing literature, we probe into the effect of region, prosodic integration, form and function of the topicalized constituent, form of the subject, and the number of constituents in the prefield on (non)inverted word order. This is the first study that carries out regression analysis on the combined impact of these variables in the entire West Flemish and French Flemish region, with additional visualization of effect sizes. The results show that noninversion is generally more widespread than originally anticipated, with unexpected higher occurrence of noninversion in continental West Flemish and lower frequencies in western West Flemish. With the exception of the variable number of constituents in the prefield, all other variables had a significant impact on word order: Clausal topicalized elements, elements that have peripheral functions, and elements that lack prosodic integration all favor noninverted word order. The form of the subject also impacted word order, but its effect is sometimes overruled by discourse considerations.

Gothic alew ‘oil’ is ultimately derived from Latin oleum. Its phonological features, however, seem hardly reconcilable with those of the Latin word. This has prompted scholars to postulate that the Latin word was not borrowed directly into Gothic but rather via a third language: continental Celtic, Illyrian or Raetic. This article examines the weaknesses of these theories and proposes that the unexpected features of the Gothic item may be explained in terms of proper Gothic or Latin developments, making direct derivation of alew from oleum the most plausible and parsimonious hypothesis.

This paper discusses three typological differences between the North Germanic DP and the West Germanic DP. While North Germanic has suffixal definite articles leading to cases of double definiteness, weak adjective endings regulated by definiteness, and doubly-filled definite DPs, West Germanic does not. These three properties cluster together in that they all have to do with definiteness. It is claimed that they can be subsumed under one more general difference. Assuming various subcomponents of definiteness, it is proposed that these components originate low in the structure. North Germanic arranges these components into several individual feature bundles. Some of these bundles move to D, while others remain lower in the structure. Consequently, definiteness components are spelled out separately in different positions. In contrast, West Germanic involves one complex feature bundle containing all definiteness components. In this language family, all of the components move to D as one bundle and, as a consequence, they are all spelled out as one determiner.