Our use of language is an everyday affair – but we still do not have a clear picture of how we ‘encode’ complex thoughts etc. into linguistic signals and how we interpret such signals in appropriate ways. The project Meaning and Understanding across Languages addresses three core questions relating to this problem:

• How can the interaction between syntax-semantics on the one hand and pragmatics on the other be spelled out in a theory of the relation between language and meaning, i.e. what is the interplay between the linguistic ‘code’ and the context in which language is used? We approached this central problem of the semantics-pragmatics interface partly from an overarching theoretical point of view, partly from an empirical perspective with a special focus on perspective in reported speech, specificity of reference, (adversative, additive and causal) discourse particles, and non-restrictive modification (Research area 1: coherence, competition and compensation).
• What impact do specific grammatical features of a speaker’s mother tongue have on how they conceptualizes reality and organizes their description of it, i.e. in how far does so-called ‘linguistic relativism’ hold as far as grammar is concerned. Research in this area has had a renaissance recently due to important methodological progress in cognitive and experimental psycholinguistic research. It has centred on tense and aspect in oral and written narrative. The present project extended experimental research to other text types and other grammatical categories (Research area 2: Language-specific perspectivation in text production).
• How do specific structural features and patterns of the speakers’ mother tongue affect the way they process and interpret texts in their second and third languages? Such questions have been investigated experimentally on the basis of single sentences, by and large out of context. The aim of the present research group was to develop precise hypotheses and robust methods of investigating speakers’ processing of second and third language texts (Research area 3: Languagespecific constraints on text comprehension).

Research areas 2 and 3 take up cognitively oriented issues concerning mother tongue impact on the productive and receptive construal of meaning. A cross-linguistic approach is essential in both areas. Research area 1 involves the working out of a theoretical model for how complex linguistic expressions are assigned an interpretation. Research in the past has demonstrated quite clearly that a contextoriented cross-linguistic approach is an essential condition for real progress even in theory design.

Work carried out in area 1, supported by related work in area 2, has added substantially to the evidence that considerations belonging to the domain of pragmatics, in particular, attunements to interlocutors' perspectives, have impacts in the domain of semantics, and thus contributed to the growing realization that the borderline between semantics and pragmatics is more porous than has generally been supposed.

The study on tense in subordinate clauses resulted in a compositional semantic theory for temporal constructions and an explicit theory for the syntax-semanticsmorphology interface.

Cooperation between researchers from different theoretical and linguistic backgrounds shed new light on presupposition-triggering particles like ‘again’ and ‘too’ as a means of discourse management, ensuring that the interlocutors keep track of the whole communication; and they have increased our understanding of the impact of cross-linguistic differences on so-called free indirect speech – and literary writing in general.

Cross-disciplinary and cross-linguistic collaboration brought up a novel perspective on the interaction of linguistic code and context in language production and comprehension (research areas 2 and 3). The main focus of the joint enterprise was on the interpretation of expressions of causality (e.g., because) and coordination (e.g., and, or). In addition, a new experimental design based on a combination of acceptability judgments and sentence completion was developed to examine temporal interpretation of conjoined clauses embedded in context. This design was applied to test cross-linguistic differences in various languages including Norwegian, German and English.

The time at CAS was an extremely fruitful and positive research experience for all members of the group, leading to new insights in all three areas, new collaborations, new research plans for the future – and new professional and personal friendships. In addition, the CAS administration, taking efficiently care of all practical matters, and the ambience, made daily life at CAS very pleasant.