Northern English & Scots phonology and Syntax: towards a Linguistic Atlas of Northern England and Scotland ('NESPS')

The Northern English and Scots, Phonology and Syntax project (NESPS) project investigated two well-known (but still quite poorly understood) features/variables of Northern English and Scots (the 'Northern Subject Rule' and 'T-to-R'), trialing a methodology which aims to offer a means to discover the structural patterning of geolinguistically restricted linguistic features, in a sociolinguistically balanced sample of speakers.

Although standard varieties of English have been quite well described, many non-standard features, such as the ‘T-to-R rule’ and the ‘Northern Subject Rule’, are still not fully understood. T-to-R involves the realisation of /t/ as an ‘r-sound’ when it occurs at the end of only certain words (with a following vowel-initial word). The NSR involves verbs taking the suffix –s for all persons in the present tense in only certain circumstances. Previous research has not fully described these structural constraints, and, moreover, we do not know precisely how widespread the two phenomena are, either geographically or socially. This project investigated T-to-R and the NSR using a novel methodology to elicit intuitions from a range of speakers of different backgrounds in two localities in Scotland and Northern England.

(i) The questionnaire-based methodology was successful in eliciting judgements about the features - for example, T-to-R had previously been described for words such as 'not, what, but' and never for words such as 'dot, pit, lout', and this was replicated in the questionnaire results; (ii) sentences with NSR-type agreement were accepted in both Hawick and Newcastle, showing that it is a widespread feature which straddles the Scottish-English political border, while sentences with T-to-R were widely accepted in Newcastle, but were not accepted in Hawick, showing that this phonological feature has a different distribution to the syntactic NSR; (iii) NSR patterning is slightly different in Hawick and Newcastle in statistically significant ways: in Hawick, more older speakers accept it than younger speakers, whereas in Newcastle, less older speakers accept it than younger speakers, and in T-to-R there is a significant difference between the ratings given by the two age groups: older respondents give a lower average rating than younger respondents; (iv) T-to-R is not possible for any speaker in 'fat, pit, hut, dot' (showing that not all words with a short vowel allow it), and in 'battle, attack, kitten, little, cotton, habit' (showing that morpheme-medial environment may be dispreferred), and 'boat, lout, loot' (showing that a long preceding nucleus may discourage T-to-R, although it is possible in 'light, feet, repeat, boot, flirt, about, bought, eat, meet, caught', so this constraint is not absolute, as had previously been thought); (iv) the ‘Proximity to Subject Constraint’, according to which subject pronouns have been claimed to tend to favour the NSR only when there is intervening material between the subject and the verb, is not significant in either locality, and verbs of communication (e.g., 'ask' and 'say') favour the acceptance of NSR most, followed by verbs of cognition ('think' and 'remember'), then other verbs (we tested 'eat' and 'see') and then verbs of emotion ('like' and 'feel').