Were all Scottish really: Investigating the Tension between Claimed Identity and Linguistic Behaviou

1
Were all Scottish really investigating the
tension between claimed identity and linguistic
behaviour in Berwick upon Tweed Heike Pichler
Dominic Watt Centre for Linguistic Research,
University of Aberdeen
2
Location of Berwick

• Englands northernmost town
• 3 miles south of the Scottish/English Border
• on the main road and rail routes between
  Newcastle and Edinburgh

3
History

• status ambiguous and complex
• Scotlands largest town during mediaeval period
• changed hands 14 times between the two kingdoms
• finally came under English control in 1482
• only fully incorporated under English
  jurisdiction in 1836
• retains numerous Scottish characteristics

4
Internal divisions

• Berwick proper on the north bank of the Tweed
• was at times part of Scotland
• Tweedmouth and Spittal on south bank
• only incorporated into Berwick in C19
• didnt change hands along with Berwick
• informants discriminate between the two sides of
  the town

5
(No Transcript)
6
Identity

• Although jurisdictionally English
• high presence of Scottish economic institutions,
  e.g. banks
• Presbyterian/non-conformist religious
  affiliations
• Berwick Rangers other sports clubs in Scottish
  leagues
• River Tweed under Scots jurisdiction
• county of Berwickshire entirely within Scotland

7
Local and national identity

• Kiely et al. (2000)
• Berwickers avoid explicitly articulating a
  definitive nationality. Instead they mobilise a
  specific identity strategy of localism
• although 41 of their 54 respondents described
  themselves as Scottish at least some of the time
• informants to north and south attributed
  opposite national identity to Berwickers,
  principally on basis of accent
• Our data suggest that
• Berwickers do indeed classify themselves first
  and foremost as Berwickers
• younger women tend to identify less with
  Scotland than do older women

8
Social significance of the border

• access to Edinburgh/Newcastle A1 dualled north
  but not south of the border
• cross-border clubs have impact on the identities
  of participants and audiences (Gil, ms)
• heavy inflow and outflow connections with
  Scotland
• Scottish Borders the main commuting
  destination outside the district
• also accounts for 69 of commuters into
  Berwick
• (source National Census 1991)
• regular short-term linguistic contact between
  speakers from both sides of the border
• increasing levels of long-term contact?

9
Linguistic significance of the border

• underplays historical permeability of border in
  linguistic, cultural and economic terms

10
Research aims

• investigating linguistic consequences of
  Berwicks proximity to border (defined both
  politically and sociopsychologically)
• effects of cross-border interaction on
  Berwickers speech patterns
• detection of patterns of convergence/divergence
  as a result of changing statuses of varieties on
  either side of border (cf. Glauser 2000)
• establishing whether observed speech patterns
  correlate with speakers claimed (national)
  identities

11
Data elicitation

• method designed by Llamas (1999, 2001) for the
  Survey of Regional English (SuRE)
• Identification Questionnaire (IdQ)
• Language Questionnaire (LgQ)
• provides a means of accessing the informants
  metalinguistic discourse
• reading passage and word lists
• Wells (1982) list list to elicit SVLR
  alternations

12
Phonological variation I Scottish Vowel Length
Rule (SVLR)

• 18 speakers 8 recorded in 2000 (Watt Ingham
  2000) 10 from current corpus (Watt Pichler
  2003)
• hypothesised that SVLR alternations giving way
  to Voicing Effect (VE) only, as per Anglo-English
• SVLR operant for /i/ and /u/ for all speakers,
  though only marginal for some
• around ½ of 18 speakers show SVLR effect for
  /ai/
• little suggestion of age effect (contra Watt
  Ingham 2000)

13
Phonological variation II rhoticity

• overtly commented on trilled r and uvular ?
  stereotyped
• highly variable ? R ? R? ? r
• 20 speakers, male/female, N/S of river, age
  range 14 78
• pre-vocalic and word-medial intervocalic (r)
• c. 1600 tokens 30 gt N gt 166 x? 76
• post-vocalic, linking intrusive (r)
• c. 2300 tokens 44 gt N gt 189 x? 99

14
variants of (r) by speaker age ()
15
non-rhoticity () by speaker age
16
Phonological variation summary

• taken in combination, SVLR and rhoticity
  patterns suggest attrition of Scottish
  phonological features in BwE
• ties in with perception among young informants
  that BwE is becoming less Scottish-sounding
• much more work to be done on these and other
  variables

17
Language Questionnaire (LgQ)

• we will gain further insights into the
  language/identity nexus by accessing our
  informants perceptions of the geographical and
  social distribution of the linguistic features
  under investigation
• e.g. do informants perceive particular forms as
  being (a)typical of, e.g., their own age group,
  sex or speech community?

18
(No Transcript)
19
Speaker sample
20
Variables

• verbal negatives in declarative sentences
• negation of periphrastic do
• present tense forms of BE
• modal can
• negatives of past tense operators (only 1/236
  formed with non standard negative particle)
• variants standard English nt and not
• Scots enclitic na(e) and clitic no
• special forms dinna(e), divvent, cannot
• found both sides of Border, but in different
  quantities
• (ScottishGeordie shorthand labels)

21
Circumscribing the variable context

• every context of negative present and past tense
  modal, auxiliary and full verbs was extracted
  from the corpus
• negative know after 1st person sg. subjects
  removed - discourse marker?
• negative 3rd person periphrastic does removed
  lack of variation relative to do

22
Overall distribution of variants

• predominance of standard variants
• stigmatisation of non-standard variants?
• gender-bias?
• divergence of varieties north and south of
  Border?

23
Internal constraints

• Data analysed in terms of
• subject type
• verb complement
• clause type
• phonotactic constraints
• ? as yet no discernible pattern

24
Age do

• dinna(e) hardly used by anyone below 60
• divvent more frequent than dinna(e) used to
  about same degree by all age groups
• Glauser (1974) recorded dinna in/around Berwick
• ? data suggest change in progress

25
Realization of dinna(e)
26
Age - BE

• freestanding no seems salient feature of BwE
• marked drop between seniors and adults
• ? data suggest that variant is becoming rare

27
Age - can

• cannot more frequent than cannae, esp. among
  adults
• cannae, however, also a feature of BwE, esp.
  among young adults and seniors

28
Age - can
29
Variation across age

• standard variants predominate among teenagers
• vernacular/localised ones among seniors
• support for Glausers (2000) divergence
  hypothesis?

30
Residence

• vernacular variants (esp. Scottish ones) used
  markedly more often south of the Tweed

31
Perceptions of BwE as more Geordie vs. more
Scottish

• speakers who think of BwE as more similar to
  Scottish English are more likely to use
  Scottish variants

32
National identification

• research question does claimed national identity
  correlate with the frequency with which
  informants use Scottish or Geordie linguistic
  forms? If so, to what extent?
• speakers were grouped by looking at the
  qualitative data, in particular
  nationality-related questions
• 3 groups (1) no clear national identification
• (2) identification with England
• (3) identification with Scotland

33
National identification

• do, BE Scottish variants used significantly
  more often by speakers who affiliate with Scotland

34
National identification

• can variation does not support
  inter-relationship of variants with national
  identification

35
Metalinguistic comments (LgQ)

• dinna(e), s no perceived by some informants,
  esp. those who use them, as forms typical of
  Berwick
• divvent perceived by most as a Berwick form

36
Synthesis

• metalinguistic comments speakers consider
  Geordie divvent and Scottish dinna(e), s no
  as Berwick forms
• also consider tapped/trilled R r as typically
  Berwick
• usage of these variants possibly not an
  expression of speakers national allegiance
• instead, the use of these variants might be a
  strategy of localism - ties in with Kiely et
  al.s (2000) findings
• cannot not included in LgQ but plausibly might
  follow same pattern as divvent
• ? re-interpretation of the quantitative findings
  (e.g. in terms of the north/south of the river
  variable?)

37
Conclusions

• attitudinal data gathered using IdQ and LgQ show
  ties between linguistic variation and claimed
  identities
• findings show benefits of accessing speakers
  attitudes toward and perceptions about usage and
  distribution of vernacular and localised forms
• forms commonly perceived as Scottish vs.
  Geordie used in Berwick to express speakers
  local affiliations as Berwickers
• present findings similar to Llamas (2001) for
  Middlesbrough
• ? evidence of phonological and morpho-syntactic
  variation being used by speakers as a means of
  (local) identity construction
• attrition of Scottish features surprising in
  light of hypothesised increase in levels of
  long-term contact with Scottish English speakers

38
(No Transcript)
39
pre- and intervocalic tap usage () by speaker age