This paper deals with a classical problem in Scandinavian grammar, so-called ‘pancake sentences’, nicknamed after examples likePannekaker er godt‘Pancakes are good’ where there seemingly is disagreement between the plural subject and the predicative adjective in the neuter singular. Our aim is twofold. From the theoretical point of view, we shall argue that there are advantages with a construction-based approach, and that such an approach is superior to previous analyses within various generative frameworks.The main reason for this is that the data require generalizations over combinations of subjects and predicative adjectives at a rather specific level. From a more empirical point of view, we shall argue that Scandinavian displays a range of different, but related pancake constructions. For the first time, corpus data are brought into the debate. We show that a construction type that has not received much attention previously is in fact the most frequent type, namely constructions where the subject is a deverbal noun.

A classical topic in the syntax of the mainland Scandinavian languages is so-called pancake clauses where there seemingly is disagreement between the subject and the predicative adjective, as in Pannekaker er godt ‘Pancakes(f):indf:pl be:prs good:n:sg’; the subject is in the plural, whereas the predicative adjective is in the neuter singular. According to one of the several approaches, these clauses display a type of semantic agreement. Recently, it has also been argued that there are at least four different types of pancake constructions. In this article, the semantic relationship between the different constructions is investigated further. It is argued that, diachronically, pancake agreement started with subjects interpreted as virtual, ungrounded processes, and that the absence of grounding has been reinterpreted as absence of spatial boundedness in the latest kind of pancake construction. The analysis is supported by a diachronic corpus investigation. The emphasis on virtual reference is a new feature with the current paper, and it enables us to set aside an objection against the semantic agreement analysis. The diachronic corpus investigation enables us to revise, empirically, earlier suggestions as to when the pancake constructions originated: They are well attested from the mid-1800s, in both Swedish and Norwegian Nynorsk.

SamandragKorleis kan funksjonell grammatikk skape grunnlag for djupnelæring i norsk-faget? I norskfaget er det ein lang tradisjon for å skilje tydeleg mellom språket som system og språket i bruk, og ein gjennomgang av nyare lærebøker retta mot lærarstudentar viser at ein framleis i stor grad held fast på dette skiljet. Dei tilrådde grammatiske termane til bruk i skuleverket bygger på ein strukturalistisk og dels generativ grammatikktradisjon der tekstar blir sett på som uinteressante som studieobjekt. I læreplanen i norsk er det derimot nettopp tekstar som er det sentrale, og det er såleis lite samsvar mellom teoretisk tilnærming til det grammatiske systemet og vektlegginga av grammatikken som metaspråk for arbeid med tekst i læreplanen. Funksjonell grammatikkundervisning er ikkje det same som funksjonell grammatikkteori, og i artikkelen blir det argumentert for å legge ei funksjonell tilnærming til språksystemet til grunn for arbeidet med grammatikk i skulen. I funksjonell grammatikk ser ein språksystemet og tekstlege realiseringar av systemet som to sider av same sak, og ein vil såleis ha eit betre grunnlag for å sjå ulike delar av faget i samanheng. Med ei djupare forståing for tydinga til språklege kategoriar vil ein også skape større potensial for overføring av kunnskap mellom ulike teksttypar og mellom ulike språk.Nøkkelord: funksjonell grammatikk, metaspråk, djupnelæringFunctional grammar as metalanguage in school– an opportunity for deep learning when working with language and textAbstractHow can functional grammar create a foundation for deeper learning in Norwegian studies? In Norwegian schools, there is a long tradition for distinguishing clearly between the language system and language in use, and a review of newer textbooks for teacher students shows that this approach is still valid. The officially recommended grammatical terminology for use in schools is based on a structuralist and partly generative grammar tradition in which texts are seen as uninteresting as objects of study. In the curriculum, on the other hand, texts are the most central objects of study, which means that there is little harmony between the theoretical approach to the grammatical system and the emphasis on grammar as metalanguage for work with texts in the curriculum. Functional grammar teaching is not the same as functional grammar theory, and in the article it is argued for a functional approach to the language system as a basis for the teaching of grammar in schools. In functional grammar, the language system and textual realisations of the system are two sides of the same coin, and it therefore provides a good basis for seeing connections between different subject areas. With a deeper understanding of the meaning of language categories, one will also create greater potential for transferring knowledge between different text types and between different languages.Keywords: functional grammar, metalanguage, deep learning

The article offers new data to the debate on lexical vs. constructional approaches to valency. Research on valency has mainly been concerned with verbs, and in this article it is argued that the neglected area of adjectival valency can shed some new light on this fundamental question. Among the evidence in favour of a constructional approach are the considerable differences in the possibilities of valency realisation between adjectives in attributive and in predicative function. It is also argued that a constructional approach allows for a more principled treatment of polysemy between valency variants.

No abstract