Mahowald Kyle. An LFG approach to Old English constituent order

University of Oxford, 2011. — 129 p.Old English (OE) differs from Present Day English (PDE) in many ways although the most noticeable is perhaps consituent order. Many generative analyses from van Kemenade (1987) to Pintzuk (1999) have sought to explain Old English word order, but no purely transformational account has successfully explained the data. In this thesis, I show how the multi-projection architecture of Lexical-Functional Grammar (LFG) can explain the data in a way unavailable to transformational accounts. While the thesis treats several different areas of grammar, there are consistent themes that emerge throughout. I will first survey the basic data that an analysis of Old English should seek to explain. This will be the focus of of Chapter 1. From there, I will consider previous analyses of OE including head-first accounts, head-final accounts, the competing grammars account, and Clark’s 2004 OT-LFG analysis in Chapter 3. Chapter 4 will introduce the basics of LFG with specific reference to Old English. The main body of the paper will focus on elements of my own lexical-functional analysis of Old English, which has been developed and tested using XLE (Crouch et al., 1998). Through the interplay of the f-structure and c-structure, LFG improves our understanding of OE syntax. Chapter 5 builds upon Clark (2004) by proposing a novel hierarchical verb cluster. Chapter 6 offers a discourse-based account of OE constituent order within the LFG framework. And Chapter 7 points the way towards an improved formal analysis of word order freezing in LFG. My conclusion offers final thoughts and directions for future research.Although this project grew out of a desire to analyze Old English systematically and without stipulation, it has largely become an effort to develop and analyze the relationships between the multiple projections available in LFG, specifically the effects of these relationships on the constituent structure. Besides just offering a fascinating laboratory for testing the bonds that hold together various parts of syntactic structure, Old English is an important language to study for its diachronic implications. If we can develop a coherent picture of Old English syntax and understand how and why consitutents get ordered in OE in the way that they do, we are a step closer to understanding how and why the language has changed in the way that it has. For example, it is not insignificant that the verb cluster I propose in Chapter 5 is rare in Middle English and vanishingly rare by Shakespeare’s time. In the space available in this thesis, I will be unable to delve too deeply into these diachronic considerations. But I believe that much of what is presented here is essential to that project and can point to exciting new directions in diachronic syntax and ultimately contribute to a clearer picture of language change. And language change is a powerful window into the way that language works as a whole.