1.4 Limitations of the study
In comparison to the corpora mentioned thus far, MICASE, CLAS, BNC and LCIE, the
datasets from the settled family and the Traveller family that comprise the corpora to be
analysed in the present study are relatively modest. While the context-type – intimate
discourse – is, as a rule, only available on a smaller scale, and, arguably, more suited to
analysis on this scale, it is important to acknowledge that with small datasets issues of
representativeness
and
generalisability
arise. In order to offset these criticisms in some
way, the quantitative data generated by the corpora in the present study, are, where
possible, compared to findings from corpora such as LCIE and the BNC (see, for
example, Sections 5.2, 6.2.1 or 8.2.3). Moreover, any qualitative examination of the
quantitative insights offered by the corpora are, in the main, supported or disputed by
evidence from both families and reference to findings from other studies of family
discourse. In addition, the present study has endeavoured to ensure that a range of
variables is comparable across the two families. As Section 4.1 demonstrates, both
families are from the Limerick City region, have a similar gender profile and represent a
wide range of ages (in both the parents and the children). Both families could also be
labelled with the traditional ‘nuclear family’ tag as they consist of two parents, married
to each other, and their biological children. Finally, the recording context of the family
home was kept constant.
Neither does the present study make any claim that the two families featured are
representative of their respective communities. The Traveller Community, while
arguably more homogenous than the settled one, consists of, for example, Travellers
that live in mobile accommodation and those that are referred to as ‘settled Travellers’
who often live in houses. Similarly, ‘the settled community’ represents a wide range of
people from differing social, educational and ethnic backgrounds. In addition, both
communities contain so-called traditional families but also an increasing number of non-
traditional family constellations. Therefore, the data collected is not presented as
representative of or generalisable to the wider Irish community, rather it is maintained
that the data is representative of the pragmatic reality of the everyday linguistic
17
practices of two distinct families. To extend Hunston’s (2002: 23) argument that ‘a
statement about evidence in a corpus is a statement about that corpus’, so too any
statement about evidence from the present study should be accompanied by a caveat: it
is evidence of the particular and not of the general.
Do'stlaringiz bilan baham: |