Pernilla Hallonsten Halling

Part II. Results

The second part of this dissertation presents the results of the study over the course of three chapters. Chapter 5, which focuses on adverbs and adverbials, presents the encoding attested in the adverbial function in the languages of the sample. Here, I show that simple adverbs are attested in a majority of sample languages, even in a number of languages that do not have simple adjectives. Chapter 6 discusses the results of examining attributive adjectives, predicative adjectives, and adverbs on three different levels within the word: the root, word form, and lexeme levels. As the root and lexeme levels are the most important ones for the present study, they are examined in particular depth. Analysis at the root level shows encoding overlaps of all three functions in a great majority of languages. Analysis at the lexeme level shows that it is fairly common for languages to have a class of lexemes used for all three functions. In chapter 7, I describe the results of examining attributive adjectives, predicative adjectives, and adverbs at the level of the construction. At this level, important connections between adverbs and predicative adjectives are discerned for a substantial number of languages, and between adverbs and attributive adjectives for a portion of languages.

3. Part III. Discussion

The third part of the dissertation is devoted to discussion. It opens with chapter 8, in which I discuss semantic types of adverbs, much like Dixon (1982 [1977]) discusses


semantic types of adjectives. Here, speed is presented as a core semantic type for adverbs, whereas noise and care are peripheral semantic types. value is found to be a semantic type for both adjectives and adverbs. Chapter 8 also discusses semantic shift of adverbs. In this context, the term shift does not refer to diachronic shift, but to the potential meaning change of a lexeme used in different functions. This leads to a discussion of how property-denoting adjectives and adverbs differ in their semantics. In chapter 9, I turn to the highly disputed issue of adverb as a part of speech. Adverbs have been described as, for instance, “the least understood large class of words in natural language” (Cresswell 1981: 21) and “the most heterogeneous of all word classes” (Haser & Kortmann 2006: 66). The discussion of the part of speech status of adverbs is intentionally left until the end of the dissertation, in order to have good grounds for drawing reliable conclusions. Despite their peculiarities, I conclude that the simple adverbs attested in this thesis constitute a stable prototypical part of speech from a cross-linguistic perspective. Part III closes with a concluding discussion in chapter 10.

4.