LINGUIST List 15.3282
Mon Nov 22 2004
Sum: English Dialect Alternation: Was/Were
Editor for this issue: Jessica Boynton <jessicalinguistlist.org>
To post to LINGUIST, use our convenient web form at
English Dialect Alternation: Was/Were
Message 1: English Dialect Alternation: Was/Were
From: Mark Jones <markjjoneshotmail.com>
Subject: English Dialect Alternation: Was/Were
Regarding query http://linguistlist.org/issues/15/15-2728.html
I recently posted a query about variationist literature on 'was/were' usage
in the non-rhotic accents of northern England. My concern was that forms
taken to be 'were' could actually be 'was', as suggested to me by a number
1) My own interpretation growing up bi-dialectally;
2) Vowel quality (open back rounded CLOTH vowel, cardinal vowel 13);
3) Written data from native speakers using 'wa' for 'were';
4) Negative forms lacking fricatives, e.g. 'wan't' for 'wasn't';
The possibility of intrusive [r] complicates matters, though even here a
tag form like 'worrit' for 'was/were it' has the open back round cardinal
I asked whether those familiar with the literature could tell me if the
problem of interpretation had been thoroughly investigated. I had two
responses only, which could suggest that the problem of interpretation has
not been thoroughly investigated in the relevant literature. Any further
insights, please let me know.
Neil Tipper, another former Sheffield resident, disagreed with my intuitive
interpretation of dialect forms as being 'was' instead of 'were' (though my
interpretation was supported by some spontaneous native speaker written
forms). Neil pointed out that there may be considerable variation with
accents south of Sheffield having more definite CLOTH vowel forms at the
Kirk Hazen of West Virginia University referred me to similar patterns in
North Carolina English which he has investigated instrumentally. The past
tense forms were lacking in rhoticity (low F3), suggesting a 'was'
interpretation. See Hazen, Kirk. The birth of a variant: Evidence for a
tripartite negative past be paradigm. 1998. Language Variation and Change
10:221-244. Of course, Yorkshire varieties of English are by and large
non-rhotic, so my original post referred to variation in vowel quality only.
Further to Neils' point, the Survey of English Dialects (SED) records many
cardinal 13 CLOTH vowels in words from the midlands and north of England
which have the NURSE long mid vowel in standard varieties, e.g. words like
'turnip'. Coincidentally this indicates that a lexically anomalous word
like 'chonnocks' for 'turnips', recorded in the SED in Staffordshire, may
actually be derived from 'turnips' by error or mini-sound change.
The occurrence of CLOTH vowels for standard NURSE forms complicates the
origins of 'was/were' variation further, as the forms with the CLOTH vowel
(which I take to be possibly 'wa(s)') could be 'were' form which have
undergone this change.
In any event, it appears that the literature has not given sufficient
attention to the problems of determining which form is in fact observed in
these patterns of variation, and whether or not the variation is therefore
phonological or morphosyntactic.
Thanks to Neil and Kirk for their help.
Mark J. Jones
Department of Linguistics
University of Cambridge
Linguistic Field(s): Language Description; Morphology; Phonology; Sociolinguistics
Respond to list|Read more issues|LINGUIST home page|Top of issue