Pernilla Hallonsten Halling

Results II: Overlaps at the root, form, and lexeme levels

1. Introduction

In this chapter and the next, I will show that attributive adjectives, predicative adjectives, and adverbs are interrelated in various ways that have not received due attention before. In the language sample of this study, this is manifested in the encoding strategies found, and the extent to which they overlap with each other. As described in chapter 4, overlaps are patterns of identical encoding in two or more functions. The present chapter describes the results of examining the encoding of attr, pred, and adv on three different levels: the root, the word form, and the lexeme, as introduced in chapter 4. The overlaps attested at the root level are discussed in section 6.2. The overlaps at the word form level, described in section 6.3, are less important for this dissertation, as they are attested in very few sample languages. Still, as discussed in chapter 4, it is important to take the form level into account as a potential basis of analysis. The most significant level for present purposes is the lexeme level, discussed in section 6.4.
The most important findings of encoding analysis at the root and lexeme levels are the following, from most to least common:

• 
  All three functions show identical encoding in a large majority of languages at the root level and in over half of the languages at the lexeme level (see sections 6.2.3 and 6.4.4).
  
• 
  attr and pred, as opposed to adv, show identical encoding in over a third of the languages at the root level and in almost two thirds of the languages at the lexeme level (see sections 6.2.1 and 6.4.1).
  
• 
  pred and adv, as opposed to attr, show identical encoding at the root level in only five languages, and at the lexeme level in six languages (see sections 6.2.2 and 6.4.2).
  

In the following sections, the overlaps are described in the same order at each level of analysis: from attr and pred to pred and adv, to attr and adv, and finally to the overlap of all functions. Throughout the sections, one or two language examples from each attested overlap are given, followed by a table summarizing all languages that instantiate the overlap at the end of each section. Just as in the preceding chapters, individual examples are marked in the right margin for the function that they instantiate (attr, pred, or adv). Any overlap that an example is argued to instantiate is indicated in the right margin of the example heading within square brackets (e.g. [attr pred]). Any

overlap at the lexeme level implies at least the same, or a greater overlap, at the root level.

2.