506 JOURNAL OF LAW AND POLICY
Regarding the variables, the IIS is concerned with discrete
variables^26 that occur in the idiolectal style of the two speakers
or writers under analysis, and they all show variation, which is
structured in two main variants, either variant A or B, or the
presence or lack of the process, following the most standard
formulations of linguistic variation analysis.^27 For example, the
variable that deals with the process of yod-coalescence explained
above contemplates two variants: 1) the presence of the process,
by which all instances where yod-coalescence occurs are
calculated and 2) the lack of process, by which all the instances
where yod-coalescence could occur but does not, are calculated.
- Corpora
Different corpora have been used to test the formulated
hypotheses, and all, in one way or another, have involved the
elicitation of semispontaneous speech,^28 except for the
morphosyntactic module of Spanish, which was analyzed by
using a written corpus. Moreover, all the corpora (except that of
the discourse-pragmatic module of Spanish) contain data from
the same adult men and women collected in two measurement
times (“MT1” and “MT2,” respectively) with a lapse of ten to
twenty years depending on the module, in order to investigate
the subjects’ idiolectal style throughout time.
The corpus of study for the Catalan modules contains data on
Eastern Catalan and consists of sociolinguistic interviews
recorded in La Canonja, a Catalan speech community in the
(^26) In statistics, variables may be a) discrete, meaning that they take a
limited number of values, such as gender (either male or female) or social
class; and b) continuous, which implies any value within a range of values on
a scale, such as age, for example.
(^27) See, e.g., LABOV, supra note 1, at 192–93; WILLIAM LABOV, THE
SOCIAL STRATIFICATION OF ENGLISH IN NEW YORK CITY 31 (2d ed. 2006).
(^28) Semispontaneous speech implies the speech resulting from an
interview, where the electronic equipment such as microphones or cameras
may make the speaker aware of the situation and inhibit them from using
completely spontaneous speech, or their vernacular, as it is referred to in
sociolinguistics.