162 Giovanna Marotta
- Discussion
The parameters we have proposed exhibit a qualitative and discrete nature,
instead of a quantitative and gradual one, as is more usual within the framework
of sociophonetics.
Some general remarks are then needed in order to show the possible advan-
tages of adopting such model of sociophonetic variation. A first benefit derives
from the lack of redundancy: each parameter is distinctive and independent from
the others. In other terms, it is not possible to predict the value for a parameter
x from the one relative to a parameter y. At the same time, there is not a strict
correlation among the different parameters, but only a lax relation (for instance,
between thickness and weight).
Furthermore, the metaphorical parameters allow an easier and more trans-
parent comparison between the various sociophonetic indexes: a process a may
receive a higher value than a process b, but a lower value than a process c, with
respect to a specific parameter. In such a way, the adoption of a point of view
based on the notion of discreteness permits to directly compare the sociophonetic
indexes, with the result of being able to forecast their spreading or compression
in space and time.
Despite their qualitative and discrete nature, our parameters could be rep-
resented in terms of a multi-factorial scale. In particular, with reference to the
typical phonological processes occurring in the Tuscan varieties considered so far,
each parameter could be assigned a different value going from a minimum up to
a maximum. However, in order to decide the exact numerical value to assign to
each parameter with respect to the phenomena considered, we would need new
and more accurate analyses, especially on the perceptual side. Therefore, at least
for the moment, we prefer to dispense with explicit multi-factorial scales.
Our model tries to conjugate the fine-grained description of phonetic events
with the adoption of a systemic analysis, thus leaving aside any theoretical per-
spective exclusively oriented towards the surface outputs of the single speaker’s
behavior. An example of such surface-oriented perspective is the one based on
Exemplar Theory which is often enrolled by socio-phoneticians (e.g., Johnson
1997; Foulkes & Docherty 2006; Carlson & Hawkins 2007). As a matter of fact,
the basic tenet of Exemplar Theory is that phonetic categories map directly onto
phonological categories; the basic elements of our phonetic knowledge are rela-
tional, dynamic, self-organizing and entirely context-sensitive (Pierrehumbert
2001, 2006; Hawkins 2010); consequently, it seems that we do not need phonology
any more. In this approach, the cognitive process of evaluation and classification
is driven by the degree of resemblance between the single and concrete manifes-
tations of the category in a word and the members assumed as prototypical for a