The Government-Binding theory cannot account for reflexives in Slavic languages. We may guess that the more a language differs from English with its quite rigid word-order, the worse are the predictions made by the theory.

One cannot exclude Slavic reflexives as non-anaphors in a non-arbitrary way while keeping the spirit of Chomsky et al. The Slavic reflexives behave “as they ought to” in tensed clauses, too. An attempt to exclude them would also be another step on the self-destructive path that started by excluding the Japanese reflexives in order to cope with the facts clashing with the Government-Binding theory. Many interesting cases of English reflexives would also have to be ignored for the sake of the theory.

This paper presents a detailed analysis of the behavior of the Icelandic conjunction enda. It turns out to be possible to explain all the puzzling facts about this conjunction by reinterpreting the traditional division between coordination and subordination by breaking it up into two binary features, [±CO] and [±SUB]. Furthermore, it is argued that this approach gives support for the analysis of Icelandic sentence structure proposed in Rögnvaldsson (1984) and elaborated in Thráinsson (1984, 1986).

An interesting restriction on Norwegian identity-of-sense anaphoric reference when the anaphor occurs as second object is discussed. Fretheim (1974) proposed to account for this by a global rule constraining pronominalization in such contexts, and claimed that the necessity of this constraint renders any interpretive account of identity-of-sense anaphoric reference untenable. The present paper purports to demonstrate the contrary, that Fretheim's observations in fact provide support for an interpretive theory of identify-of-sense anaphora as far as Norwegian is concerned.

Written Norwegian has two words, the co-ordinating conjunction og and the infinitive marker å, both of which are pronounced [o] in Spoken Norwegian. In this paper, an attempt is made to show that [o] in Spoken Norwegian constitutes a single polysemous word. Furthermore, it is argued that Spoken Norwegian [o], in all its usages and meanings, is a descendant of Old Norse ok “and”, while the Old Norse infinitive marker at has not survived in Spoken Norwegian.

This article examines the semantic structure of the Norwegian preposition mot ‘towards, against’, and suggests an analysis within cognitive semantics where it is argued that the various senses of this word are best accounted for in terms of a family resemblance category, with two interrelated image schemas, SOURCE-PATH-GOAL and FORCE DYNAMICS, as its basic senses. Other senses are treated as extensions, for the most part metaphorical, of these two basic senses. The study builds on a corpus of Norwegian texts from various genres.

Shortly after arriving in Copenhagen five years ago, I realized what many linguists have long understood: the case situation in the Nordic languages is formidably complex. Of course, the broad outlines of inter-speaker (or, cross-linguistic) variation in Nordic nominal case inflection are well known. Within two major language families, (North) Germanic and Uralic, there are dozens of closely related language varieties. The Finnic and Sami languages of Uralic have adpositional case systems, while the North Germanic languages can be further subdivided into the Mainland and Insular groups, partially on the basis of their different case systems. The latter group, namely Icelandic, Faroese, and Älvdalian (which is spoken in a fairly isolated rural community in the interior of Sweden), has ‘rich’ inflectional case morphology on a range of elements comprising nominal phrases, including articles, determiners, demonstratives, nouns, pronouns, wh-words, and more. The former group, namely Danish, Norwegian, and Swedish, are ‘case-poor’, maintaining only a vestige of their historically rich case morphology on a subset of personal pronouns, which have Nominative, Oblique, and Possessive forms. Furthermore, certain varieties of Swedish and Norwegian retain vestigial Dative forms of clitic pronouns.

It has been reported that speakers of Danish understand more Swedish than vice versa. One reason for this asymmetry might be that spoken Swedish is closer to written Danish than vice versa. We hypothesise that literate speakers of Danish use their orthographic knowledge of Danish to decode spoken Swedish. To test this hypothesis, first-language (L1) Danish speakers were confronted with spoken Swedish in a translation task. Event-related brain potentials (ERPs) were elicited to study the online brain responses during decoding operations. Results showed that ERPs to words whose Swedish pronunciation was inconsistent with the Danish spelling were significantly more negative-going than ERPs to words whose Swedish pronunciation was consistent with the Danish spelling between 750 ms and 900 ms after stimulus onset. Together with higher word-recognition scores for consistent items, our data provide strong evidence that online activation of L1 orthography enhances word recognition of spoken Swedish in literate speakers of Danish.