Ever since its introduction by Christian Lehmann in 1982, the term ‘degrammaticalization’ has been the subject of much controversy. There has been considerable disagreement about what kind of phenomena the term refers to, and indeed if degrammaticalization exists at all. In this paper I will argue that degrammaticalization is not grammaticalization in reverse, but a separate kind of change worthy of study in its own right. The purpose of this paper will be to show how degrammaticalization changes are best identified, by using existing linguistic tests as diagnostics for establishing changes in the morphological status of certain degrammaticalization suspects, and for assessing the nature of the subchanges involved. One of the stock examples of degrammaticalization literature, i.e. the s-genitive in Swedish, will serve as a case study.

The paper attempts to determine the motivating causes of palatal vowel harmony (PVH). Previously suggested causal explanations of PVH are critically evaluated, especially the progressive palatal assimilation view of the origin of PVH. Data on PVH restrictions from Finnish and Turkish are examined against a set of perceptually motivated working tendencies. It is shown that PVH and “labial harmony” are only special cases of a single, unitary type of restriction, statable acoustically in terms of the frequency of the second formant. A perceptual theory of the causes of PVH is proposed and general phonetic conditions for and against the development of PVH in a language are suggested. On the whole, the paper is an argument for a substance-based approach to phonology.

Northern Norwegian (NN) allows verbs to precede adverbs in non-V2 contexts, whereas in Standard Norwegian (StN) verbs have to follow adverbs. These facts are discussed with respect to three different approaches to clausal structure. NN is problematic for a head movement account (cf. Cinque 1999) because multiple verbs may precede a given adverb, leading to violations of the Head Movement Constraint. A multiple positions account (cf. Ernst 2002, Svenonius 2002) would assume that any adverb in StN and NN can be adjoined to high positions, which may be problematic with respect to scope relations. A remnant movement approach (cf. Nilsen 2003) can account for both StN and much of the NN data by means of one generalisation, but a separate generalisation is needed for finite verbs in NN. Thus, all three approaches are faced with challenges with respect to the Norwegian data. However, it is argued that the remnant movement approach seems the most promising of the three approaches.

The aim of the present study is twofold. First, it addresses the question of how well Norwegians are able to identify Norwegian dialects. Second, it investigates the role which intonation plays in the identification of Norwegian dialects. In a number of listening experiments, 15 groups of young Norwegians from different parts of Norway were asked to identify 15 different dialects. The dialects were presented in the original form as well as in a version that was manipulated by computer in such a way that the intonation was removed. In this way it was possible to see how important intonation is for the identification of the dialects. The results are put into perspective by comparing them to results of similar experiments with Dutch dialects. Norwegians turned out to be less good at recognizing Norwegian dialects than Dutchmen. Linguistic and extra-linguistic explanations are given. Furthermore the results show that intonation is more important for the identification of Norwegian dialects than for the identification of Dutch dialects.

The paper discusses the process of adverbialization in English, focusing on one type of adverb, subject-modifier adverbs such as sadly, thoughtfully and pinkly. It is also shown that the -ly suffix in English (unlike its cognates in the other Germanic languages) has become an extremely versatile adverb suffix. Finally, it is argued that in English, the manner adverb category is prototypical, whereas other adverb types, notably subject-modifier adverbs, are less central adverbs.

Various linguists working within the theory of Government and Binding (e.g. Hornstein & Weinberg (1981), Kayne (1981)) have attempted to provide a unified account of preposition-stranding. This article uses evidence from Icelandic to show that preposition-stranding is not a unified phenomenon. Although Icelandic freely allows preposition-stranding in wh-movement constructions, it lacks prepositional passives in which the prepositional object of an active verb corresponds to the grammatical subject of a passive verb. Various syntactic tests which distinguish between grammatical subjects and topicalized NPs are used to demonstrate this. Our conclusion is that while lexical reanalysis is needed to account for prepositional passives, no such reanalysis is warranted for preposition-stranding due to wh-movement; hence, an adequate theory of preposition-stranding must allow for two separate parameters.

In contemporary informal Estonian, the negative verb form ei tea ‘don't know’ has become a routinized part of generic questions, in which the agent is left unexpressed. This pattern is in accordance with the general impersonal and reference-avoiding style of conversing in Estonia. The study outlines a continuum of synchronic usages from the original expressions sa ei tea ‘you don't know’ and ma ei tea ‘I don't know’ to the epistemic usages of (ei) tea, which are specifically tied to the speech act of questioning. The data is interactional and the analysis relies on the interpretation of (ei) tea-questions by the participants themselves, following the methodology of conversation analysis. It is demonstrated that the development of (ei) tea displays phonological and semantic erosion, pragmatic strengthening, subjectification, and decategorialization. Thus, grammaticalization theory is here combined with interactional linguistics in order to display the emergence of a grammatical structure from a discourse pattern.

We give an overview of the phonological properties and processes that define the categories of the prosodic hierarchy in Swedish: the prosodic word (ω), the prosodic phrase (φ) and the intonation phrase (ι). The separation of two types of tonal prominence, big accents versus small accents (previously called focal and word accent, e.g. Bruce 1977, 2007), is crucial for our analysis. The ω in Swedish needs to be structured on two levels, which we refer to as the minimal ω and the maximal ω, respectively. The minimal ω contains one stress, whereas the maximal ω contains one accent. We argue for a separate category φ that governs the distribution of big accents within clauses. The ι governs the distribution of clause-related edge phenomena like the initiality accent and right-edge boundary tones as well as the distribution of nuclear big accents.

Based on a corpus of spontaneous production data, this paper compares the word order of wh-questions in two Norwegian dialects, Kåfjord and Tromsø. While the choice of word order (V2 or non-V2) in Tromsø is dependent on information structure, the Kåfjord speakers produce considerably more non-V2 in questions with monosyllabic wh-elements. The majority of questions with multisyllabic wh-constituents, on the other hand, occurs with V2. This synchronic variation is given a diachronic analysis within a Split-CP model of clause structure and a cue-based approach to acquisition and change, where an economy principle (head preference) also plays an important role. Furthermore, an information structure drift from V2 to non-V2 is argued to cause the cue for verb movement to fall below a critical level in the input to children, the result being that V2 only survives in lexically marked cases in Kåfjord, i.e. with the verb ‘be’.

Parts of a new model of phonology-morphology-lexicon interplay is presented to account for the complex distribution of the Modern Danish stød (a syllabic prosody). Stød, which is sometimes productive for speakers, is analysed as a signal of the second mora of bimoraic syllables not subject to the Non-Stød Principle (NSP). The author's cross-language model for Systematically Graded Productivity of Endings (section 3) is shown to account for the application of NSP (section 4), and a detailed typology of lexemes with respect to stød-alternations, derived from the model, is presented in section 5. In section 6, a simple case of stød-alternations in inflection, viz. regular plurals of nouns, is given, and section 7 exemplifies stød and non-stød as a key to morphology for the addressee.

This paper defends the analysis of Scandinavian ‘pancake sentences’ as semantic or referential agreement (Enger 2004). Alternative analyses assuming that pancake sentences are to be analysed as syntactic agreement (Josefsson 2009), have some drawbacks. Notably, the distinction between different kinds of gender is not well motivated; the connection between different uses of the neuter is lost, and the use of ‘light verbs’ and of invisible elements is problematic. The referential agreement analysis, in combination with the Agreement Hierarchy, yields diachronic predictions that turn out to be correct; the syntactic analysis does not yield anything equivalent. Furthermore, the agreement pattern for very ‘general’ nouns follows more naturally under the referential agreement analysis.

The Danish stød, a kind of glottal prosody associated with certain syllables, as in barʔn ‘child’ (cf. stødless barnlig ‘childish’), has long been the target of intense phonological investigation. In this paper, we show that its analysis requires an understanding of the prosodic constituent structure of Danish, and of the essential role of the perfect prosodic word (coextensive with one foot). After motivating this notion on independent grounds, both in other languages and in the context of acquisition, we show that the Danish stød system, analyzed in Optimality Theory, provides a window on the workings of the perfect prosodic word, regulating the presence and absence of stød in some of the much-discussed cases in the literature. In conclusion, we discuss the status of the perfect prosodic word in the light of recent developments in phonological theory, such as Match Theory.

In this paper, we consider Swedish Finite VP-Topicalization and Yiddish Finite-Verb Topicalization, which share certain interesting features that have implications for Germanic syntax and for syntactic theory in general. Although neither construction contains a gap, we argue that both are products of Movement, but with a lexically realized trace, and that this phenomenon is crucially related to Movement of finite verbs. The arguments are based on the distribution of aspectual types, lexical and phonological phenomena, the V/2 Constraint, Subjacency, and discourse phenomena.

The presentational focus construction in Norwegian can have an agent in object position, as in Det arbeidet en mann i skogen ‘It worked a man in the-woods”. This creates problems for linking theory, in which a robust generalization says that agents are not realized as objects. Optimality Theory makes possible a new approach to this classical problem in Scandinavian generative grammar. The constraint that agents are not realized as objects must compete with other constraints. In Norwegian, it is ranked below a requirement that a presentational focus must be realized in object position. The Norwegian situation is compared to languages with different constraint rankings.

Two variants of what looks like disagreement between a subject and a predicative adjective are explored:

1. (i)

2. (ii)

This article attempts to give a uniform and coherent analysis of the semantics and syntax of negation, with special emphasis on negative concord (NC) constructions. It is argued that negation is semantically a tripartite quantifier binding the event variable in its restriction. In syntax, this logical form corresponds to a structure where a NegP is present with Neg ° (normally) hosting the negative particle and a specifier as a checking position in the Minimalist sense. As a consequence, negative indefinites are held to be weak indefinites with a formal feature NEG which has to be checked away (thus obtaining Neg absorption). Additionally, some new data are introduced showing that even English style negative indefinites are analysable as non-negated weak indefinites.

Most Norwegian declaratives are subject-initial verb second (V2) clauses. This paper discusses declaratives that can be construed as non-V2, two constructions that have traditionally been analyzed as left dislocation phenomena: the (adjunctive) så-construction and the Copy Left Dislocation (CLD) construction, where the ‘copy’ is a weak pronoun. Both constructions share an affinity to root clauses, have particular scope effects, and employ a prosodically light particle between the topicalized phrase and the finite verb in V2 (så and a weak pronoun, respectively). The paper attributes these properties to the fact that the relevant particles are topic markers of a particular kind; they mark A-topics. A-Topics signal a topic-shift in the conversation and are confined to clauses with illocutionary force (Bianchi & Frascarelli 2010). The aforementioned particles are much more frequent in spoken contexts than in written prose, and I propose that this is because they depend on prosody. They are obligatorily light, and they occur in the part of the clause that has traditionally been described as ‘the Wackernagel position’. Wackernagel (1892) proposed that certain prosodically light elements (clitics in particular) tend to occur in the second position in Indo-European languages. Although the resumptive elements of the så-construction and especially of CLDs may not be fully-fledged clitics, like clitics, they appear in the second position of declaratives.

The aim of this paper is to shed new light on the tonal grammar of Icelandic and to complement the tone inventory as previously described in the literature (Árnason 1998). Specifically, types of pitch accents and edge tones and their combinations in neutral declaratives and questions, and in utterances containing narrow focus are addressed. Two pitch accent types (H* and L*) and two edge tones (H- and L-) are identified, for which evidence has not been found in previous research. Moreover, the paper shows for declaratives, that along with downstep, Icelandic has upstep across Intonational Phrases. Upstep applies to a series of pitch peaks. It may occur in neutral declaratives and in utterances with final narrow focus. Overall, the results of this study provide a substantial addition to our knowledge of Icelandic intonational phonology.

In a recent study of the clause-initial position in verb-second declaratives (the prefield), Bohnacker & Rosén (2008) found significant differences between native Swedish and German concerning the frequencies with which constituents occurred in the prefield, as well as qualitative differences concerning the mapping of information structure and linear word order: Swedish exhibited a stronger tendency than German to place new information, the so-called rheme, later in the clause. Swedish-speaking learners of German transferred these patterns from their L1 to German. Their sentences were syntactically well-formed but had Swedish-style prefield frequencies and a strong pattern of Rheme Later, which native Germans perceive as unidiomatic, as an acceptability judgment and a rewrite-L2texts task showed. The present study extends Bohnacker & Rosén's work in three ways. Learners of the reverse language combination (L1 German, L2 Swedish) are investigated to see whether similar phenomena also manifest themselves there. Secondly, written and oral data from highly advanced learners are examined to see whether the learners’ persistent problems can be overcome by extensive immersion (3, 6 and 9 years of L2 exposure). Thirdly, besides investigating theme–rheme (old vs. new information), some consideration is given to another information-structural level, background vs. focus. The learners are found to overuse the prefield at first, with non-Swedish, German-style frequency patterns (e.g. low proportions of clause-initial expletives and high proportions of clause-initial rhematic elements). This is interpreted as evidence for L1 transfer of information-structural or discourse-pragmatic preferences. After 6 and 9 years, a substantial increase in clause-initial expletive subjects, clefts and lightweight given elements is indicative of development towards the target. The findings are related to current generative theorizing on the syntax-pragmatics interface, where it is often maintained that the integration of multiple types of information is one of the hardest areas for L2 learners to master.

The Extended Projection Principle (EPP) has remained a controversial topic in generative grammar. This article proposes to derive the EPP from a generalized theory of nominal case and verbal agreement. According to the proposal presented in this article, morphosyntactic features such as case and verbal phi-features are valued uniformly by the closest asymmetrically c-commanding element, whereas the PF interface is constrained so as to prevent verbs from being valued nominal case and nominals by verbal phi-features. This constraint together with a new theory of valuation explains the appearance of the EPP. The theory is applied to the investigation of negative clauses in Finnish and other languages, Finnish (elliptical) non-finite negative clauses, expletive constructions, multiple wh-movement in a variety of languages, multiple agreement both in the finite and nominal domains, and asymmetries between finite and non-finite clauses.