'Ane end of an auld song?': macro and micro perspectives on written Scots in correspondence during the Union of the Parliaments Debates

Abstract

This thesis examines the relationship between political identity and variation from a diachronic perspective. Specifically, it explores the use of written Scots features in the personal correspondence of Scottish politicians active during the Union of the Parliaments debates. Written Scots by 1700 had steadily retreated from most text-types in the face of ongoing anglicisation, but simultaneously the Union debates sparked heated discussion around questions of nationality and Scotland's separate identity. I consider the extent to which the use of Scots features may have been influenced by such discourse, but also how they may have become indexical markers used to lay claim to these ideologies. Drawing from the frameworks of First, Second and Third Wave perspectives on variation, and combining quantitative, macro-social methods with micro-social analysis, broad socio-political factors are explored alongside plausible stylistic intentions in conditioning or influencing the linguistic behaviour of these writers. The first analysis examines variation in the corpus temporally, using the chronologically-organised clustering technique VNC - Variability-based Neighbor Clustering (Gries and Hilpert, 2008), to measure Scots features over time. The crucial years of the debates (1700-1707) are compared with correspondence either side, and the VNC analysis identifies heightened use of Scots falling within the key years of the debates. The following macro-social analysis explores the factors driving this variation quantitatively, using a number of different statistical models to examine the data from various perspectives. Probabilities of Scots are found to correlate with certain political factors, though in complex and multilayered ways that reflect the composite nature of the historical figures operating in the Scottish parliament. The third analysis focuses on the features of written Scots itself and how these pattern in aggregate and across the individual authors who comprise the corpus. Findings suggest the persistence of written Scots was not being driven by a singular feature or set of tokens, rather, authors varied widely in their range and proportion of different variants. Finally, the micro-analysis examines the intra-writer variation of four individuals representing different political interests, exploring their Scots use across various recipients. Close-up inspection of features within particular extracts and letters suggests the subtle social and stylistic functions Scots had acquired for these writers. Its occurrence was found to reflect but also constitute the macro-social patterns identified earlier.

Taken together, results indicate the use of Scots features was both influenced by, and contributed to, the political and ideological loyalties these writers harboured. Moreover, they tentatively suggest a process of reinterpretation was underway, in which Scots features were becoming a resource that could be selectively employed for particular indexical and communicative purposes.