172 Rosanna Sornicola and Silvia Calamai
Starting from an archive of dialectal speech is of crucial importance for the study
of urban dialectology in the southern territories of Italian peninsula. The tra-
ditional models of urban dialectology, mostly elaborated in an Anglo-American
and German framework, focus on the analysis of variation in the sub-standard
language. However, north-American, British and German towns are substantially
different from the big urban areas of Southern Italy. The former are interested in
processes of linguistic standardization that involve large parts of the social classes.
The latter still present high levels of dialectal uses that (i) are employed in func-
tionally different contexts; (ii) are particularly vital in code-switching; (iii) inter-
fere with regional and sub-standard Italian.
Dialectological and ethnographic sound archives may have different goals
from those of current sociophonetic research (but it is useful to recall that the
corpora of data collected by the French and Swiss dialectologists of the end of the
19th and the beginning of the 20th century prompted the pioneering analysis of
linguistic variability that later on was to become a source of inspiration for socio-
phonetic research). If investigated in the light of the basic principles of variation
and variability of European linguistics, they offer a considerable amount of new
data for sociophonetics. In particular, ADICA has been constructed according to
some general principles that clearly illustrate the potential of combining dialectol-
ogy and sociolinguistics:a. the principle of relativity of variation;
b. the principle of the centrality of the individual;
c. the principle of microscopy in the study of variation;
d. the principle of oscillation and the principle of linguistic context;
e. the principle of having recourse to spontaneous speech.The principle of relativity of variation assumes that the range of variation is not
defined by one absolute theoretical unit with respect to which all other variants
are to be considered as ‘alterations’; on the contrary, it is defined by the presence
of more than one variant. For this reason, the archiving procedure of all items
is based on the underlying etymological lexeme. In this respect, the historical
perspective proves the most flexible for storing the multiple outputs of a segment.
The principle of the centrality of the individual speaker assumes that the indi-
vidual has to be considered as the basic unit of analysis, according to a series of
arguments developed by several scholars from Schuchardt to Jespersen, Mathesius
and the founders of Romance dialectology. The reflections in Mathesius (1911) are
particularly revealing in this respect. Mathesius observes that some linguistic phe-
nomena, such as the length of English stressed vowels, are characterized by a range
of variability defined by specific boundaries. This range of variability is associated