Linguistic vs. Non-linguistic Conditioning of Linguistic Variables

Résumé

Using as its data some results of dialect contact (Trudgill 1986) in the history of Newfoundland English, this paper attempts to test the following two hypotheses.

(1) It is naturally easier to distinguish linguistic from non-linguistic conditioning if these two types of conditioning produce opposed effects rather than similar effects. 

(2) Linguistic conditioning is likely to be stronger in cases of structural variation than in cases of lexical variation, since structural systems (or subsystems) are usually more tightly organized than are lexical systems (or subsystems).

We will examine the Newfoundland fates of selected structural and lexical variants brought from one or more of Newfoundland's three main source areas CDevonia', southeastern Ireland, and 'Dorsetia') in the Old World (Mannion 1974 and 1977; Handcock 1989). The results clearly demonstrate the crucial role of non-linguistic (social or socioeconomic) conditioning in some of the relevant contact situations (Thomason and Kaufman 1988). They also suggest that linguistic conditioning can be both powerful and complicated, sometimes involving subtle interplays of FORM and MEANING for both structural variants (Paddock 1988 and 1991) and lexical variants (Story, Kirwin and Widdowson 198211990).