The order of premodifiers in English nominal phrases

Abstract

The research reported in this thesis sought an explanation for the order of premodifiers in English nominal phrases. It aimed to establish what validity there is in the quite divergent earlier explanations, to find any other valid forms of explanation that might exist, and to integrate them all. The method was to make a wide survey of as many varieties of current English as possible, by observation; to then analyse the order at all levels (semantics, syntax, and so on); and to check the accuracy of the results against the 100-million-word British National Corpus. From that research, the thesis asserts that parts of most past approaches can be integrated into a comprehensive explanation; and that there is a new and important element of the full explanation, namely that of words' semantic structure, which is the combination of types and dimensions of meaning that make up the sense of each premodifier. Other new elements in this treatment of the subject are analysis of long groups of premodifiers (up to 10 words), consideration of why premodifiers regularly occur in different positions in the order, and explanation from the historical development of premodifier order. After an introductory chapter and a survey of the relevant literature, the thesis argues that the explanation of premodifier order in English nominal phrases is as follows. There are four positions for premodifiers, as in "your (1) actual (2) tinny (3) round (4) percussion instrument" [i.e. a tambourine] (chapter 3). The regular, unmarked order (illustrated in the phrase just quoted) has several elements of explanation: primarily, the semantic structure (chapter 4); secondarily, the syntactic structure (chapter 5). In a second type of order (when two or more words occur in one position), stylistic considerations control the order, not grammatical ones (chapter 7). In a third type of order, a marked one, a premodifier may be put in a position different from the position that the word's usual semantic structure would require, changing its meaning and stylistic effect (chapter 8). Some features of all three types of order are to be explained partly by their historical development - for example, the existence of borderline uses (chapter 9). There are some supporting explanations, from discourse structure and psycholinguistics, for example (chapter 10). The relevance of the previous chapters to wider issues, such as grammaticalisation, is discussed (chapter 11); and conclusions are drawn (chapter 12).