66
individual‟s network ties are with his/her local community, the closer his/her
language approximates to local vernacular norms. Milroy and Milroy‟s research
rejects the Labovian premise that prestige is the primary agent of language change
and instead claim that the individual‟s social network is the determining factor in
maintaining or precipitating change. They do not consider ethnicity as
sociolinguistically relevant to Northern Irish English as their research failed to
discern any systematic phonological differences between Protestants and Catholics
(see Milroy and Milroy, 1985; Milroy, 1987). Their rationale for this was that
making any ethnic-based linguistic differentiations was difficult due to generations
of high-level Protestant/Catholic residential segregation. They further claim that
community members in Northern Ireland referred to accents in geographical terms,
for example east Belfast, as opposed to marking accents in terms of Protestant or
Catholic. Therefore, the chief dimensions of variation in Belfast are associated with
urban location and with locally constituted networks (Milroy and Gordon, 2003).
However, in contrast to this, McCafferty (1998, 1999) maintains that ethnicity does
indeed play a role in language variation and change in Northern Ireland. His studies
of Londonderry/Derry English (LDE) demonstrate that strength of social network is
not grounds enough to overlook ethnic differences. McCafferty (1999: 264) claims
that Catholics, of whatever class, are more likely to use variants characteristic of
LDE, whether these are local innovations or older ones that are dying out in other
parts of Northern Ireland. Conversely, Protestants are more willing to adopt
variations from the rest of the North, mainly Belfast, and are less likely to use
localised forms, old or new.
Social networks are relevant to the present study because they provide a set of
procedures for studying small groups, such as ethnic minorities, migrants or rural
populations, where speakers are not discernable from one another in terms of any
kind of social class index (Milroy, 2002: 556). Analysis of these networks has
highlighted a disjunction between the networks of male and females. Eckert (2000),
Dubois and Horvath (1998) and Chambers (1995) have all noted
that women seem to
form ties across a wider social spectrum than men. Male speech often appears more
localised and conservative than female speech and Milroy (1999) attributes this
behaviour to the particularly constraining effect of male peer networks. Some
language differences between Traveller men and women may be best explained by
68
with the settled community than Traveller men (Gmelch, 1989). Therefore, these
women are at the forefront of establishing a Traveller identity in settled contexts.
Although social network theory offers much to the study of family discourse, it is
deemed unsuitable for this particular study for a number of reasons. The first is that
a social network is, by definition, concerned with a limitless web of different social
ties stretching through a society. However, this study is concerned primarily with the
ties that exist within the family – those that exist between family members and
individuals outside the family are incidental to the study. Secondly, Holmes and
Meyerhoff (1999: 180) maintain that one of the fundamental differences between
social network theory and the community of practice is that
a social network requires
Do'stlaringiz bilan baham: