140 Giovanna Marotta
the acoustical signal have necessarily a corresponding perceptive value for the
listener, although the sociophonetic indexes are necessarily carried on by some
phonetic substance.
Sociophonetic indexes are normally conceived of as gradient rather than cat-
egorical. In sociolinguistic research notions like continuum and gradient scale are
traditionally employed instead of category and discreteness: e.g., idiolects and lin-
guistic varieties merge one into the others without a clear definable boundary; the
linguistic phenomena sensitive to the socio-cultural factors show a gradual nature
more than a binary one; phonemes cannot be conceived any more as granitic enti-
ties, identical in every situation and in all the speakers of a language or a dialect (see
for instance Labov 1972, 1994; Thomason & Kaufman 1992; Coulmas 1998, 2001).
However, in our opinion, sociophonetics cannot entirely dispense with a sort
of discrete representation of sounds, if its goal is to capture the linguistic compe-
tence of the speakers (see below, §9 and §10). The perception of speakers has to
be grounded on some kind of categorical units, in the sense of cognitive entities
located in the mind of the speakers/listeners of any linguistic community.
Indexicality is a double-face entity: on the one hand, it can be considered as the
measure of correlations between phonetic variable forms and social factors; on the
other one, it may also refer to the awareness of these correlations by speakers. More
or less, this difference covers the distinction made by Silverstein (2003) between
First Order Indexicality and Second Order Indexicality: in First Order Indexicality,
reference to patterns of correlation between specific linguistic forms and social
factors is meant, whereas in Second Order Indexicality the awareness of the socio-
phonetic correlations by the speakers and listeners of a given social community is
involved. Moreover, the awareness can be overt or tacit, according to the degree of
prestige of the linguistic variation (Trudgill 1972; Labov 1966, 1975, 2001).
As a matter of fact, we believe that a deep inspection into the nature of socio-
phonetic indexes is needed in order to catch their crucial aspects. First of all, the
phonetic features encompassing sociolinguistic variation do not have the same
status, both in the speaker’s awareness and with reference to the linguistic sys-
tem; therefore, they cannot be considered in the same way. Their status can be
different, because:involving the two entities (sign and object); therefore, there is a causal relation between signans
and signatum (Peirce 1972, 1991). Let us quote two relevant passages with specific reference to
this notion: “an index is a sign which would, at once, lose the character which makes it a sign
if its object were removed, but would not lose that character if there no interpretant. Such,
for instance, is a piece of mould with a bullet-hole in it as sign of a shot” (Peirce 1991: 239);
“an index is a sign really and in its individual existence connected with the individual object”
(Peirce 1991: 251).