Thesis (Ph.D.)--Chulalongkorn University, 2007

This dissertation explores within the Principles and Parameters theory the syntactic variation of English adverbs in the interlanguage of Thai learners with respect to the range of positions of adjunction relative to the clause. It also investigates how this develops during a period of two years, divided into four stages. In this study, adverbs were categorised into twelve positions of adjunction and 37 semantic types. The subjects were five advanced and five intermediate Thai learners of English. The adjacency parameter was first applied. According to the parameter, English is a [+/-strict adjacency] language, whereas Thai is a [+strict adjacency] language, and so clause-medial adjunction is permitted only in the former. That is, Thai grammar is a subset of that of English. Thus, in order for Thai learners to acquire the range of positions of adjunction in English, parameter resetting from the [+strict adjacency] value to the [+/-strict adjacency] value is needed, based on positive evidence. Regarding the adjacency condition, the native group placed adverbs quite equally between the clause-initial and the clause-medial positions. On the other hand, the learner groups put much more adverbs clause-initially. This indicates the likelihood that the [+strict adjacency] setting of Thai was being transferred and parameter resetting in this respect was partial. Despite this fact, the advanced group was on a par with the native group, placing adverbs in all the twelve positions being investigated, whereas the intermediate group put adverbs mostly in nine positions. The development in both regards was generally gradual for both groups, although the adjacency condition fluctuated more for the intermediate group, while the advanced group attempted more new positions of adjunction. In addition to the adjacency parameter, the lexical parameter was adapted. According to this parameter, English adverbs, regardless of their semantic categories, are allowed to adjoin in many clausal positions, whereas those in Thai can generally adjoin only clause-initially or clause-finally, depending on their semantic types. That is, the syntactic specifications of Thai adverbs form a subset of those of English adverbs. Thus, the acquisition of the range of positions of adjunction again requires parameter resetting based on the L2 data, but this time at the lexical level. It was found that the natives placed adverbs in the majority of the 37 semantic classes in more positions than the advanced learners, whose range of positions of adjunction was broader than that of the intermediate learners. However, both groups of learners put adverbs in more positions than possible in their L1. This likely shows the effect of L1 transfer and partial resetting to accommodate the lexical parameter of the L2. An analysis of the markedness of different positions of adjunction reveals that the less marked positions were acquired before the marked ones. Also, the degree of markedness and thus the acquisition order seemed to correspond with the degrees of adjunction in those positions identified in the native data, suggesting some correlation between frequency and the extent of markedness. Finally, several interlanguage aspects might play a part in shaping the Thai learners’ interlanguage, namely language transfer, transfer of training, avoidance behaviour, and overgeneralisation. The first three affected the intermediate learners to a greater extent than the advanced learners, whereas the last was a major difficulty confronting both groups of learners.