INTRODUCTION
In many bilingual and multilingual communities – understood here
to include communities in which two or more varieties (e.g. standard and
nonstandard) of a single normatively defined “language” are
spoken – certain communicative practices may be said to be
code-specific in that they conventionally require, and are
constituted in part through, the speaker's use of a particular code.
Where two or more languages or varieties coexist in a hierarchical
relationship, and where functional differentiation of codes is a salient
feature of everyday language use – as is the case in most creole
settings and many other situations of language contact (Thomason 2001, Garrett
2004), particularly those where diglossia in Fishman's (1967) expanded sense may be said to obtain –
speakers may assert that a particular code in their community
repertoire1
This notion of community
repertoire is not meant to suggest a communal repertoire; neither
should it be taken to suggest that all speakers have equal control of, or
even access to, all of the codes comprised. On the contrary, as suggested
by Pratt's (1987) critique of
“linguistic utopias,” it should be assumed that this is
not the case.
is particularly well suited for certain
types of communicative practices, but ill suited for others. Typical
examples of the kinds of communicative practices in question include
joking, scolding, gossiping, storytelling, insulting, and cursing,
2 This brief list of verba dicendi is not
intended to privilege the culturally specific communicative practices
typically recognized and named by speakers of (standard) English, nor to
exclude from consideration those of other groups.
thus running
the gamut from what
Goffman 1981 characterizes
as “self-talk,” such as cursing aloud to oneself (or perhaps
at an inanimate object or someone not co-present) as a form of
“response cry,” to more overtly performance-like modes of
language use that presuppose the co-presence and relatively focused
attention of at least one addressee, such as telling a joke or a story. In
most bilingual/multilingual settings, code-specific communicative
practices such as these simultaneously constitute and partake of what I
shall call
code-specific genres: normative, relatively stable,
often metapragmatically salient types of utterance, or modes of discourse,
that conventionally call for the use of a particular code.
In this article I suggest that the notion of code specificity, and the
related notion of code-specific genre, can be useful ones for theorizing
the relationship between code and communicative practice in
bilingual/multilingual settings – particularly settings in which
language shift and other contact-induced processes of linguistic and
cultural change tend to highlight that relationship. In order to
demonstrate this, and to show how the notions of code specificity and
code-specific genre can be applied to a specific ethnographic case, I
examine how young children in a St. Lucian community are socialized to
“curse” and otherwise assert themselves by means of a code
that under most circumstances they are strongly discouraged from using.
Among the questions that arise are the following: How and why does a
particular set of code-specific communicative practices come to be singled
out and given special emphasis in language socialization interactions? In
a situation of ongoing language shift, how and why do children manage to
acquire just enough of the obsolescent code to engage in these specific
practices (but not others)? Finally, what might be the implications of
these language socialization activities for processes of language shift?
In order to address these questions, I begin by developing further the
notions of code specificity and code-specific genre. I then examine some
typical language socialization practices involving code-specific genres
that I observed and recorded in a St. Lucian community where language
shift is currently underway.
FROM DIGLOSSIA TO CODE-SPECIFIC GENRE
A striking commonality across bilingual/multilingual communities
is that speakers, when engaging in communicative practices such as those
mentioned above, tend strongly to use the language or language variety
that is generally regarded as occupying the “lowest” position
in the local hierarchy of codes. In many cases, the language or variety
that occupies this position is nonstandard, lacks official status, and is
unwritten; or else it is considered to be “broken,”
“corrupted,” or “impure,” and therefore something
less than a “real language.” By virtue of these broad but
potentially significant commonalities, the aforementioned communicative
practices are an excellent point of departure for examining code
specificity and code-specific genres as widespread phenomena that demand
to be accounted for theoretically.
Why should particular communicative practices in
bilingual/multilingual contexts call for the use of a particular code
– in the case of cursing, scolding, insulting, and other practices
like those considered here, the “Low” code?3
The focus here will be on genres that entail use of the
vernacular/nonstandard/“Low” code, but speakers
typically have equally strong feelings that other genres, such as lectures
and sermons, necessitate use of the
formal-official/standard/“High” code (Ferguson 1959).
For
Ferguson 1959 and numerous others since (
Hudson 1992) who have described such contexts in terms
of diglossia, the explanation lies in broad correlations, empirically
observable and readily demonstrable by means of quantitative analyses,
between speakers' code selections and particular domains of usage (or
“situations,” in Ferguson's original formulation). Such
correlations between code and domain reflect what Ferguson (
1959:328) characterized as “specialization of
function for H[igh] and L[ow],” the first of his
nine defining characteristics of diglossia. Other early work on language
choice or code selection in bilingual/multilingual settings (e.g.
Fishman
1965,
1972)
likewise took broad correlations between code and domain as their focus,
usually also taking into account conventional associations between code
and domain that were expressed by (or could be elicited from) native
speakers in response to questionnaires and/or structured
interviews.
Other investigators, particularly those with an ethnographic
orientation, have tended to regard the same kinds of empirically
observable, relatively stable patterns of communicative practice as
deriving from the choices and behaviors of individuals (rather than vice
versa, as in domain-centered approaches). While acknowledging that
language use is always to some extent constrained by and responsive to
“higher-order societal structures” (Milroy
& Muysken 1995:6), these investigators have considered code
selection and other aspects of communicative practice to be fundamentally
open-ended and underdetermined, and they have emphasized the agency of the
individual speaker. As pioneers of this type of approach, Gumperz and his
colleagues (e.g. Blom & Gumperz 1972, Gumperz 1982) provided an enduringly influential
alternative to correlational and domain-centered approaches to
code-switching, functional differentiation of codes, and related
phenomena. More recent linguistic anthropological approaches (e.g. Woolard 1989, Rampton 1995b,
Urciuoli 1996, Zentella
1997, Bailey 2002) have continued to
build on this foundation, particularly in investigations of
code-switching, which is regarded as “systematic, skilled, and
socially meaningful” use of two or more codes (Woolard 2004:74).4
See
Stroud 1992 for an insightful critique of
certain assumptions that tend to inhere in such approaches.
Recent ethnographic works such as these reveal an important
shortcoming of the diglossia model and other domain-centered approaches:
They tend to overlook (if not ignore altogether) the facts that there is
always some slippage between code and domain, and that speakers can and do
exploit that slippage as a communicative resource in and of itself. In
other words, domain-centered approaches do not take adequately into
account the fact that codes are not actually tied in any hard and fast way
to the domains with which they may be conventionally associated by native
speakers, or correlatively and statistically associated by investigators
(particularly those who uncritically espouse the approaches in question).
Consequently, such approaches are unable to account for much of the
situated communicative behavior that can be observed at almost any given
time in any community. Constantly and in myriad ways in everyday
interaction, speakers actively and creatively exploit the looseness of fit
between code and domain in order to achieve specific communicative
effects. (The porosity of code boundaries in many language contact
situations opens up further expressive possibilities, as Woolard 1998 demonstrates in her discussion of
“bivalency.”) To take a simple example,5
See Morgan (1996:409–14ff.) on
“reading dialect” for an intriguingly more complex
example.
an august personage, expounding publicly in the
“High” variety, may make a humorous aside in the
“Low” variety – just as a jokester in a neighborhood bar
the next day may use the High variety in lampooning the same august
personage. In both cases, it is in large part the incongruity, the lack of
“fit,” between code and domain that renders a readily
apprehended humorous effect. Domain-centered approaches occasionally make
passing note of phenomena such as this, but they fall well short of
providing satisfactory explanations of these and other aspects of situated
language usage.
As noted previously, cursing, scolding, and other such code-specific
communicative practices – which are to some degree conventionalized
and a focus of metapragmatic attention in most if not all communities
– can usefully be regarded as what Bakhtin (1986:60 et passim) calls speech
genres: “Each separate utterance is individual, of course, but
each sphere in which language is used develops its own relatively
stable types of these utterances. These we may call speech
genres” (original emphasis).6
Presumably Bakhtin, a literary scholar, refers here to
“speech genres” (rather than simply “genres”) in
order to stress that his concerns extend beyond literary genres and
written texts.
On the surface, this basic definition, with its
emphasis on “spheres” of usage, might suggest a
domain-oriented approach that is as limited and limiting as those
mentioned above. But actually, Bakhtin's notion of genre is much more
sophisticated and dynamic in that it explicitly addresses the tension
between centripetal and centrifugal forces that, in Bakhtin's (
1981) view, are constantly acting on language.
7 Briefly, centripetal forces are those that tend
to have centralizing, homogenizing, regimenting, normative (often
prescriptive, standardizing) effects on language and language use.
Centrifugal forces are those that tend toward decentralization,
heterodoxy, and diversity, and often involve the lifting or breaking (or
breakdown) of normative constraints.
To understand the dynamics
of that tension is one of Bakhtin's foremost theoretical concerns.
Another is to find principled ways of coping with the seemingly
overwhelming multiplicity of utterances – that is,
langage,
“the heterogeneous mass of speech facts” that was adumbrated
and then almost immediately bracketed and set aside by Saussure (
1966:7–15
et passim). Genres, as
“relatively stable and normative forms of the utterance”
(
Bakhtin 1986:81), provide a principled means of
cutting through that multiplicity and bringing order to the heterogeneity.
For Bakhtin, the utterance, “with all its individuality and
creativity, can in no way be regarded as a completely free combination of
forms of language” (
1986:81). Genres,
irreducibly social in origin, normative and conventionalizing in nature,
are the key constraint on the forms that utterances take and the ways in
which those forms are combined and recombined, effectively setting
parameters for the individual speaker's situated use of language at
any given moment (Saussure's
parole). At the same time, from
an analytic perspective, the notion of genre provides a basis for
classifying and categorizing utterances, and ultimately for developing a
taxonomy of utterances that is amenable to theorizing.
8 This is not to suggest that genres are static. As Ben-Amos
(1976:xxxi) puts it, genres are “evolving
forms … not merely analytical constructs, classificatory categories
for archives, file cabinets, and libraries … but distinct modes of
communication.”
Drawing on the insights of Bakhtin and others, Bauman (2000:84) has recently defined “genre” as
“a constellation of systemically related, co-occurrent formal
features and structures that serves as a conventionalized orienting
framework for the production and reception of discourse.” (This
definition succinctly encapsulates some of the major ideas about genre
developed in Briggs & Bauman 1992; see also
Bauman 1992.) As Bauman's definition
suggests, in order for the notion of genre to serve as a coherent unit of
analysis that is of explanatory and classificatory utility in the study of
situated utterances, certain minimal commonalities of form and/or
function must be identified at some level.9
As an anonymous reviewer for Language in Society
points out, the notion of “genre” as used here has much in
common with the notion of “speech style” as used by Hymes
1989[1974] and Urban 1985.
Bauman notes that the concept of genre
“emphasizes conventionality and textuality” (85); that is, it
relates a given text to prior texts. This intertextual aspect of genre is
a crucial one in that it facilitates the recognition and interpretation of
a particular utterance or performance as belonging to (i.e., as being a
token of) a particular genre. Clearly, in a bilingual or multilingual
setting, the code conventionally used for a particular communicative
practice may be an important consideration in these regards: to the extent
that a code-specific communicative practice is constitutive of its
corresponding “primary” genre, in Bakhtin's (
1986) sense
10 This is
not to suggest that communicative practices logically (or ontologically)
precede genres; the two should be regarded as mutually
constitutive.
– for example, to the extent that the
practice “joking” is constitutive of, and coterminous with,
the genre “joke” – the genre in question can likewise be
described as code-specific. But just as clearly, the code conventionally
used is not sufficient in and of itself to be definitive of a given genre;
this is true even of the most singular and most rigidly code-specific
genres that can be imagined, such as an incantation that requires rote use
of an archaic or otherwise esoteric ritual language (which may be
referentially opaque even to the few who have knowledge of it and are
deemed qualified to use it).
It is therefore necessary to look to other features. In the case of
code-specific genres, domain may present itself as a likely candidate; to
continue with the example just given, an incantation is unlikely to be
heard outside the specific ritual context in which its functionality and
performativity are realized (and realizable). But as is suggested by the
hypothetical example given previously of the distinguished orator and the
mocking jokester, a genre that is code-specific (e.g. the oration or the
joke) is not necessarily domain-specific as well. Indeed, it would seem
that there can be no such thing as a truly domain-specific genre (though
doubtless some genres can be said to be more domain-specific than others;
the ritual incantation and the joke, for example, can perhaps be located
at two different points on a continuum of domain specificity). Although a
particular code-specific genre may well be strongly associated with a
particular domain, genres by definition have the potential, indeed the
tendency, to traverse domain boundaries. Code-specific genres are no
exception in this regard; hence the vernacular joke that punctuates the
otherwise solemn oration. In order to avoid the problems presented by the
presumption of an overly tight linkage between code and domain, and in
order to be able to account for those instances in which a given genre
slips the bounds, if only momentarily, of the domain with which it is
typically associated, utterances “belonging to” that genre
must be identified and classified as such with reference to other criteria
and other levels of analysis. The same is true if particular genres are to
be grouped together in sets and placed under higher-level rubrics such as
the one proposed here, code-specific genres.
To these ends, it may be necessary to look beyond formal features (the
usual criteria for identifying and defining genres) and to take pragmatics
into account.11
Bauman (1992:138) observes, “As suggestive as they are
… Bakhtin's writings engender a certain amount of frustration
… his perspective seems to demand a dimension of formal analysis,
but he never provides it.”
One salient characteristic that
cursing, joking, and the other code-specific communicative practices under
consideration here have in common is that each involves a self-assertive
stance on the speaker's part. In each case, the speaker is
effectively calling attention to himself or herself, either as someone who
has been offended or slighted in some way (as with cursing, scolding, and
insulting of the earnestly vituperative sort) or as a performer of some
kind (as with joking, gossiping, storytelling, and various forms of ritual
insulting). Furthermore, the speaker is presenting himself or herself as
someone whose words are straightforward, candid, and uncensored,
12 The objection might be raised that such terms
as these do not apply to ritual insults, which clearly are not meant to be
taken at face value. But consider that ritual insults are situated on the
tense boundary between self-censorship and self-abandon – between
what can and cannot be said freely and openly. One who engages skillfully
in ritual insults is displaying expert knowledge of precisely where that
boundary lies.
and are to be understood and evaluated in terms of
local, in-group values and vernacular frames of reference. (In the case of
certain types of verbal performances, such as telling folktales, notions
of authenticity may also be relevant.) In other words, in invoking one of
these code-specific genres, the speaker is claiming for himself or herself
– at the moment of speaking, at least – those positively
evaluated characteristics that are typically associated with the
vernacular or “Low” code and its community of speakers, and is
at the same time implicitly rejecting, or at least distancing himself or
herself from, the pretense, subterfuge, and artifice (or merely the
heightened self-monitoring and self-censorship) that are commonly
associated with the use of, and users of, the official-standard variety or
“High” code.
13 For an
interesting examination of how the tensions generated by such a
polarization of codes have been dealt with in the Haitian context, see
Schieffelin & Doucet 1994.
The
self-assertive stances that are typically associated with and indexed by
the genres under consideration here, I suggest, are a function of the code
specificity of those genres. Or to approach the matter from a different
and perhaps more productive angle: The tendency for certain genres in
bilingual/multilingual settings to be code-specific, I suggest, is a
function of the fact that the invocation of those genres is a conventional
means of indexing and displaying certain kinds of self-assertive
stances.
CODE-SPECIFIC GENRES AS AN ASPECT OF
HABITUS
Considerations such as the above raise the possibility that for
bilingual or multilingual speakers, selection of genre may sometimes take
precedence over selection of code, and may effectively preempt it. When a
code-specific genre is selected by a speaker, the code conventionally
called for by that genre is presupposed and thus entailed by the genre
itself, rather than being selected (or “chosen”) in a separate
exercise of pragmatic judgment on the speaker's part. That is, when a
speaker formulates an utterance so that it conforms to and partakes of a
code-specific genre, the speaker does not also make a separate (even if
closely and integrally related) pragmatic judgment as to which of the
codes in his or her repertoire is the most appropriate one for invoking
that genre. The code selection is predetermined, inasmuch as it is a
constitutive aspect of the genre – a matter of social convention
rather than of individual choice or preference, stylistic or
otherwise.14
This is not to suggest that the
code selection is overdetermined. It is possible for a speaker to
“bend” a generic frame without breaking it – and in a
bilingual/multilingual context, use of a nonconventional code might
well be one way of doing so. In Briggs & Bauman's (1992) terms, this would constitute a
“widening” of the “intertextual gap.” It is also
worth noting here that the code conventionally used for a code-specific
genre may shift, either gradually or abruptly. (An example of abrupt shift
would be the shift from Latin to local languages in the Roman Catholic
liturgy after the Second Vatican Council.)
What makes code specificity a matter of theoretical interest, I
suggest, is that it inheres in the genre itself as a constitutive feature
thereof; it is not contingent upon, nor a function of, the individual
speaker's situated invocation of that genre. Thus code specificity is
not reducible to, nor is it merely an epiphenomenon of, code-switching,
register-shifting, or style-shifting. (As this suggests, much of what has
been regarded as “code choice” or “code selection”
in the literature on code-switching is perhaps better regarded as genre
selection.) Consequently, code specificity, and the situated invocation of
code-specific genres by speakers, require a different mode of analysis
from these other phenomena. Where a code-specific genre is concerned,
“code choice” or “code selection” (if such terms
even apply) is not a matter of the individual speaker's making
context-sensitive modulations in his or her communicative behavior in
response to factors such as the relative status of his or her addressee(s)
or the presence of potential overhearers. If anything, code specificity
can be expected to have a constraining effect on code-switching and
style-shifting, particularly situational switching (Blom & Gumperz 1972).15
In contrast to situational switching, metaphorical switching
(Blom & Gumperz 1972) may be exploited quite
extensively by the speaker within the parameters of those code-specific
genres that can be regarded as “secondary genres” in
Bakhtin's (1981) sense. Gossip is a prime
example: Although the vernacular invariably serves as the matrix code
(Myers-Scotton 2002), the gossiper may very well
code-switch when quoting someone, or perhaps as a way of characterizing
someone's manner or behavior (either verbal or nonverbal). This
underlines the larger point that code specificity need not constrain the
speaker's creativity and expressiveness; indeed, in this and other
ways, a genre's code specificity can itself be exploited to these
ends.
A speaker who is conversing with someone of higher social
status in the High code or in an honorific register, for example, will
most likely refrain from cursing or gossiping in the first place
16 See Haas 1964 on
“interlingual word taboos” for a description of an interesting
related phenomenon. A brief example: Haas noticed that Thai students in
the United States, when in the presence of English speakers, “tend
to avoid certain words of their own language which bear a phonetic
resemblance to English obscene words,” such as fàg
‘sheath, (bean-)pod’, khán ‘to crush,
squeeze out’, and phríg ‘(chili)
pepper’.
; and a speaker who becomes aware of the approach
of potential overhearers is more likely to cease cursing or gossiping
altogether than to switch to an alternate code and continue doing so.
17 But cf. Reisman (1970:139) on what he characterizes as “the
fading effect” in Antigua: “By the fading effect I mean the
tendency of Creole speech to fade or disappear without any marked or
noticeable shift in the manner of speaking when outsiders, particularly
whites, and even white investigators of language,
approach.”
Even speakers who code-switch extensively in everyday interaction may
be observed to curse, joke, or scold predominantly if not exclusively in
one particular code. Asked to reflect on the matter, they may go so far as
to assert that it would be impossible to do so in any other available
code. The other side of the coin is that speakers may feel that the code
in question is good only for such practices18
As an anonymous reviewer for Language in Society
urges me to point out, native speakers of a language commonly make
metapragmatic assertions that are demonstrably untrue and/or
contradictory, that tend to perpetuate stereotypes about the language and
its speakers, and that may be discrepant with their own everyday use of
the language. It is not my intent to propagate, condone, or even
countenance any of the evaluations of specific languages or language
varieties that I describe, report, or allude to in this article.
– or that it is iconic of those practices, and of those who
characteristically engage in them (
Irvine & Gal
2000). The poet David Dabydeen, who grew up in Guyana before
migrating to England as a teenager, begins an essay with this
observation:
It's hard to put two words together in creole without
swearing. Words are spat out from the mouth like live squibs…. The
creole language is angry, crude, energetic. The canecutter chopping away
at the crop bursts out in a spate of obscene words, a natural gush from
the gut. (1990:3)
Commenting on the canonical forms of English poetry as well as the
English language itself, Dabydeen ultimately asserts, “I cannot feel
or write poetry like a white man” (13). Authentic artistic
expression for Dabydeen necessitates using the creole, “a language
that is both lyrical and barbaric” (5).19
Analogous assertions extolling the expressive, poetic and
other virtues of creole languages are made by other creolophone writers,
including Brathwaite 1984 and Bernabé, Chamoiseau, & Confiant 1993; the
works of writers and performers such as Louise Bennett and Mutabaruka
(both renowned for their use of Jamaican Creole) are also worth
consideration in this regard. See also Ngugi
1986 on the use of African indigenous languages in literature; and
cf. Gates (1990:45–46) on what he
characterizes as the “mockingbird” posture. For a discussion
of the use of creole languages in the novel and other written genres, see
Mühleisen 2002.
Dabydeen's dissatisfaction with English as vehicle of poetic
expression for his own sentiments, life experiences, and artistic
sensibilities – notwithstanding the fact that it is, overwhelmingly,
the usual medium of written and printed communication in England as well
as in Guyana and elsewhere in the “anglophone” Caribbean
– can be regarded as one manifestation of the intuitive and deeply
felt (though generally tacit and unreflective) sense that speakers have of
how the two or more codes at their disposal are to be deployed, under what
circumstances, and to what ends.20
Hymes
(1989[1974]:451) notes that
“fluent members of communities often enough themselves evaluate
their languages as not equivalent. It is not only that one language, or
variety, often is preferred for some uses, another for others, but also
that there is experience with what can in fact be best done with one or
the other.”
For creole speakers in Guyana, St. Lucia, and
elsewhere, pragmatic, stylistic, and aesthetic judgments such as these are
expressions of an ideology of language (
Woolard &
Schieffelin 1994,
Schieffelin et al.
1998) that is rooted in speakers' subjective experience of the
creole as the language of intimacy and sentiment – of local life as
locally lived. Attempts to convey these sentiments and experiences
(whether in poetry and other literary genres or in talk) through the
official-standard or High variety fail to ring true; the results
invariably seem forced, artificial, inauthentic.
21 Clearly this broadly generalizing statement is in no way
intended to dismiss or discount the formidable and well-known
accomplishments of the many Caribbean writers, orators, and others who
have made, and continue to make, brilliant use of these
languages.
The code itself lacks the immediacy and transparency
of its vernacular counterpart. As a second language for many speakers,
shot through with prescriptive regimentations, it over-mediates, demanding
attention for its own sake and impeding free, “heartfelt”
expression. In addition to impediments such as lexical gaps (real or
perceived),
22 Commenting on a different but
related issue – the “negative definition” of Creole,
vis-à-vis English, in Antigua – Reisman (1970:137) notes, “In speaking with non-Creole
speakers, or in any formal context that uses English as a main code,
villagers are almost universally both unwilling and unable to cite Creole
forms, to translate from English to Creole, or to repeat Creole utterances
in even vaguely the same form.”
there is a deeply felt
sense that the insult would not be as cutting, the gossip not as juicy,
the scold not responded to with sufficient alacrity; that the joke would
fall flat,
23 Michaels
2003 provides an example to demonstrate the semi-futility of
translating a Yiddish joke into English:
The rabbi says, ‘What's green, hangs on the wall,
and whistles?’
The student says, ‘I don't
know.’
The rabbi says, ‘A herring.’
The student says, ‘Maybe a herring could be green and
hang on the wall, but it absolutely doesn't
whistle.’
The rabbi says, ‘So it doesn't
whistle.’
Michaels comments, “The joke is inherent in Yiddish, not any
other language. It's funny, and, like a story by Kafka, it isn't
funny.”
or that uttering the curse would be, at best, less
than satisfying. In Austin's (
1962) terms,
the speaker's use of the “right” code thus qualifies as a
felicity condition for these speech acts; to use the “wrong”
code (if the possibility of doing so can even be entertained by native
speakers) would be to render the speech act infelicitous.
24 For present purposes, Austin's notion of
speech act can be regarded as broadly equivalent to Bakhtin's speech
genre in that both authors are concerned with the recurrent,
conventionalized aspects of language usage. (It is perhaps worth noting
here that Austin has been critiqued by linguistic anthropologists for
allowing his thinking on this issue to be limited by the verba
dicendi of Western languages.)
Even in situations where the
link between code and genre is a looser one, and where the speaker has
somewhat more flexibility as to code selection – a teacher scolding
an unruly student in the schoolyard during recess, for example
25 An observation made by Reisman (1970:140) on language use in Antigua is interesting to
consider in this regard: “Creole is intrinsically felt to be the
code of the genuine. School teachers, even head teachers, may, or may be
forced to, move into Creole to convince the children that they really mean
what they are saying. Thus other forms of speech carry some aura of
falseness.”
– the speaker's selection of one
code over another may contribute substantially to the perlocutionary force
of the utterance.
Observations such as these highlight the embodied, even visceral
character of bilingual/multilingual speakers' subjective
experience of codes and code-specific genres, and the intimately related
“practical sense” of language that guides their usage of the
codes at their disposal. Hanks (1996:246)
regards genres as “modes of practice,” observing that
“they are among the best examples of habitus as a set of enduring
dispositions to perceive the world and act upon it in certain ways
… they embody just the kinds of schemes for practice that
constitute the habitus.” Further evidence that knowledge of genres
is part of habitus (Bourdieu 1977, 1991)26
See Bourdieu 1991 for focused discussion of
“linguistic habitus” – the more or less coherent,
integrated set of “socially constructed dispositions … which
imply a certain propensity to speak and to say determinate things (the
expressive interest) and a certain capacity to speak, which involves both
the linguistic capacity to generate an infinite number of grammatically
correct discourses, and the social capacity to use this competence
adequately in a determinate situation” (1991:37).
is the fact that such knowledge is
largely preconscious and tacit. As Bakhtin (
1986:78) comments, “Our repertoire of oral (and
written) speech genres is rich. We use them confidently and skillfully
in practice, and it is quite possible for us not even to suspect
their existence
in theory” (original emphasis). Lacking the
benefit of Bourdieu's theoretical formulations (see
Hanks 1987 for a particularly useful statement of how
genres can fit into a practice theory framework), Bakhtin imputes to the
speaker a “speech plan” or “speech will,” which is
“manifested primarily in the
choice of a particular speech
genre” (
1986:77–78ff., original
emphasis). Whatever the ontological or phenomenological underpinnings of
this “speech will,” and however the speaker's
“choice” (or selection) of genre comes about in practice, the
speaker conforms his or her utterance more or less rigidly to that genre
through the process characterized by
Briggs &
Bauman 1992 as “calibrating the intertextual gap”
– a process that is as fundamental to everyday conversational
exchanges as it is to verbal artistry and performance in both oral and
written modes. But there are limits to the flexibility afforded to the
speaker by any given genre. These are best thought of as parametric
constraints (as opposed to hard and fast limits) on the individual
speaker's idiosyncratic use of language: If an utterance or
performance departs too radically from its generic antecedents, the
intertextual gap widens to a point where the “conventionalized
orienting framework” begins to give way, with all of the
consequences for intersubjectivity and mutual intelligibility that this
entails.
Habitus, in Bourdieu's well-known formulation, is
“inculcated,” that is, socialized. But how is it
socialized? In particular, how do children in bilingual/multilingual
communities acquire the subtle, finely tuned, embodied sense of when and
how to deploy the codes available to them? Bakhtin's observations in
this regard would seem to point in the right direction:
We speak only in definite speech genres … We are given
these speech genres in almost the same way that we are given our native
language, which we master fluently long before we begin to study grammar
… The forms of language and the typical forms of utterances, that
is, speech genres, enter our experience and our consciousness together,
and in close connection with one another. (1986:78)
As these observations suggest, and as the literature on language
socialization amply demonstrates (Schieffelin & Ochs 1986a, 1986b, Garrett & Baquedano-López 2002), linguistic
and sociocultural knowledge are acquired simultaneously and are
inextricably bound up with one another; thus language socialization is in
large part a matter of the child's or novice's acquiring the
particular configuration of historically, socially, and culturally
specific “dispositions” that, taken collectively and as an
integrated whole, constitute his or her habitus. A speaker's
knowledge of the genres of his or her community is inseparable from his or
her knowledge of the various other aspects (semantic, syntactic, etc.) of
that community's language(s). A language socialization approach to
the issue of code specificity thus has the potential to enhance our
understanding of how children and other novices in
bilingual/multilingual settings learn to cope with code-specific
genres, as well as how those genres are reproduced and transformed in the
process. This in turn has the potential to illuminate how children's
emergent mastery of code-specific genres articulates with their emergent
socially grounded knowledge of the codes themselves.
THE CASE OF ST. LUCIA
St. Lucia's Afro-French creole, known locally as Patwa or
Kwéyòl,27
The label
Patwa (or patois) is of colonial origin, and some St.
Lucians – though by no means all – consider it pejorative.
Those who hold this view now prefer to call the language
Kwéyòl instead. Because the latter appellation is now widely
recognized (though still not widely used colloquially) among the general
public in St. Lucia, I refer to the language as Kwéyòl
here.
dates from the island's early French colonial period
(c. 1642–1814), which after numerous military and diplomatic
skirmishes between rivals France and England was followed by a British
colonial period (1814–1979). For most of its brief history,
Kwéyòl has been regarded, even by its own speakers, as a
sort of “broken” or “corrupted” French: the
coarse, unrefined tongue of slaves and, later, peasants – in short,
something less than a “real language.” Today, after several
decades of coexistence with English, Kwéyòl continues to be
spoken by the majority of St. Lucians, most of whom are now bilingual to
varying degrees. But this situation is changing fairly rapidly: Many St.
Lucian children today are acquiring only limited proficiency in
Kwéyòl, if any. There are clear indications that
Kwéyòl is going into decline and that it is being replaced,
or rather displaced, by a simultaneously emerging vernacular variety of
nonstandard English, which I refer to as Vernacular English of St. Lucia,
or VESL (Garrett
2000,
2003). Although English prevails in formal and
official domains, relatively few St. Lucians control the standard variety;
by and large, it is the emergent VESL, not standard English, that is
effectively displacing Kwéyòl as the vernacular of many
members of today's younger generations. VESL is heavily influenced by
Kwéyòl in its lexicon, syntax, phonology, and other
subsystems (
Garrett 2003).
Language shift in St. Lucia has progressed to a point where some
children, particularly in and around the capital city of Castries (where
VESL has made the greatest headway), are growing up as monolingual
speakers of VESL. In rural areas, meanwhile, both Kwéyòl and
VESL are spoken. In the rural village where I conducted fieldwork, Morne
Carré, many of the oldest residents are virtually monolingual
speakers of Kwéyòl, while most persons middle-aged and
younger are bilingual (to varying degrees) in Kwéyòl and
VESL. Many of their children are also bilingual to some extent, their
proficiency in Kwéyòl and VESL broadly reflecting the extent
to which each language is used with them by their caregivers and other
members of their households. This varies considerably: In some households,
caregivers are diligent about speaking only “English” (really
VESL) with children, while in others, Kwéyòl is used almost
exclusively. Most Morne Carré households fall somewhere between
these two extremes, with adults speaking mostly Kwéyòl among
themselves but showing a strong preference for the use of
“English” with children. The preference for the use of English
by children is even stronger; children are expected to respond in
English even when addressed in Kwéyòl by an angry or
impatient caregiver.28
When the annual
cultural festival Jounen Kwéyòl ‘Creole
Day’ was observed during my fieldwork, one of the teachers in the
Morne Carré school, as part of the day's special activities
for the occasion, asked her class of 8- to 10-year-olds to tell her some
things that they knew how to say in Kwéyòl. Getting only
puzzled looks, she then asked the students to tell her some things that
they hear their mothers and fathers say in Kwéyòl. This got
an enthusiastic response, as the giggling children called out phrases such
as Pa fè sa! ‘Don't do that!’,
Sòti la! ‘Get away from there!’, and Sa ki
wivé'w?! ‘What's the matter with
you?!’.
Some language socialization activities that adults commonly engage in
with children, such as calling-out and greeting routines, can be regarded
as characteristically “creole” regardless of the code used;
such routines thus highlight important historical continuities in local
notions of personhood, sociability, and morality, despite the fact that
they are now used in ways that privilege displays of communicative
competence in VESL rather than Kwéyòl. But other
socialization routines that foster the development of locally valued
social and verbal skills, particularly self-assertion, seem to require
Kwéyòl – despite the fact that children are otherwise
discouraged more or less systematically from using the language.
What role might these varied language socialization practices be
playing in the ongoing process of language shift that was outlined above?
It is perhaps best to begin by reversing the elements of the question and
asking what impact this process of language shift is having on cursing and
other genres that St. Lucians at present continue to constitute, virtually
without exception, through the use of Kwéyòl. In this and
any other case of language shift, one clear possibility is that the code
toward which speakers are shifting – the “new” code, as
it were – will simply be pressed into service for these
“old” vernacular functions. There are likely to be impediments
to such a process, however, such as the high overt prestige of that code
and its associations with pedagogy and other relatively formal domains of
usage, and the prescriptive regimentation that these entail.29
In a case of relatively recent contact between
two languages and cultures with extremely disparate histories, the
challenges of adapting a vernacular to the exigencies of a novel genre may
be even more formidable, as Schieffelin 1996,
2000 demonstrates. The “hybrid” or
“boundary” genres of colonial Yucatán described by
Hanks 2000 are another interesting case to
consider in this regard.
Another, perhaps more likely possibility
is that speakers will persist in using the obsolescent code (the code that
speakers are shifting away from) for the particular communicative
practices that call for it, and in the particular domains of usage in
which these practices figure prominently, even after they have ceased to
use the code for other practices and in other domains. Various observers
of language shift and obsolescence have noted that certain words and
phrases, such as greetings and exclamations, may continue to be used as
salient markers of ethnic identity and group membership – as
“diacritica,” in Barth's (
1969)
sense – long after the process of shift has gotten underway. These
forms may be used as such even by semi-speakers, near-passive bilinguals,
and others whose overall proficiency in the language is quite limited.
Dorian 1982 provides a well-known example of
this phenomenon, making the larger point that the ability to use even a
very limited repertoire of such words and phrases appropriately, in
accordance with local norms of interaction, is a crucial aspect of
communicative competence (
Hymes 1972) – so
much so that it may allow for full participation and
“membership” in the speech community even on the part of
persons who have quite limited proficiency in the obsolescent code. Dorian
further notes that it may actually have the effect of masking the
(semi-)speaker's relative lack of productive competence and
fluency.
30 In other cases, knowledge and use
of such verbal formulae serves as an important means of conveying respect
or goodwill toward members of a group to which one does not belong (Heller 1982, Jackson 1974),
and may even make it possible for persons who clearly are not (and cannot
become) members of a particular group to interact with members of that
group on an intimate level (Rampton 1995b, Spears 1998, Urciuoli 1996,
Zentella 1997).
In his pioneering study of language contact, Weinreich (1953:95) notes, “An obsolescent language seems
destined to acquire peculiar connotations and to be applied in special
functions even after it has lost its main communicative role.” One
likely outcome identified by Weinreich is for the language to
“acquire a certain esoteric value” – that is, to become
a sort of “secret” code that speakers can use to exclude
community outsiders and/or members of the younger generations of their
own community. A second tendency is for the language to develop
“comic associations”; Weinreich notes that among the children
of American immigrants, “the mere utterance of a word in their
parents' language easily evokes laughter.”
It is the third likely outcome noted by Weinreich that will be the
focus here: the retention, maintenance, and reproduction of certain
pragmatically salient lexical items and expressions in the speech of
members of the younger generations whose proficiency in the obsolescent
language is otherwise quite limited. As Weinreich puts it:
The stylistic specialization of an obsolescent language and
the association of it with intimate childhood experiences is conducive to
the borrowing of its lexical elements into the younger people's
speech, especially in discourse that is informal and uninhibited by
pretensions of high social status. Particularly apt to be transferred are
colorful idiomatic expressions, difficult to translate, with strong
affective overtones, whether endearing, pejorative, or mildly obscene.
(1953:95)
For present purposes, I shall dispense with endearments31
Kwéyòl endearments (e.g.
doudou ‘sweetie’) are in fact commonly used by St.
Lucian caregivers.
and focus here on items that fall under
Weinreich's categories of “pejorative” and/or
“mildly obscene.” However, I depart from Weinreich's
conceptualization of this phenomenon as a process of “stylistic
specialization”
32 via
“borrowing” or “transfer” on the part of younger
speakers. I regard it rather as the retention and ongoing reproduction,
through language socialization, of such items, and of the larger
communicative practices in which they are embedded.
KWÉYÒL AND
“ENGLISH” IN MORNE CARRÉ
The data excerpts to be examined here are drawn from a large corpus
(over 100 hours) of naturalistic audio-video recordings gathered as part
of a language socialization study (Schieffelin & Ochs 1986a, 1986b; Garrett & Baquedano-López 2002) that I
conducted in St. Lucia during 16 months of fieldwork in 1996–1997.
The study focused on five children in five different households whose ages
at the beginning of the study ranged from 18 months to 3 years. Each
child's language acquisition and social development were followed
longitudinally as I examined how the children and their caregivers
interacted in culturally specific socializing routines and activities
whereby the children learned to recognize, negotiate, and indeed to create
and co-construct various kinds of locally meaningful interactional
contexts. A central goal was to examine these micro-level, locally
situated developmental processes in the contexts of everyday home and
community life, and to relate them to various macro-level linguistic,
ideological, and sociocultural changes currently underway in St. Lucia. In
addition to the process of language shift outlined previously, these
include such transformations as the increasing accessibility of primary
education and the gradual transition from agriculture to international
tourism as the island's economic mainstay.
Morne Carré is a rural village with a population of just under
300. It is situated atop a low but steep mountain ridge that rises above
the relatively remote and sparsely populated northeastern coast of the
island. Most Morne Carré residents earn their living, or a
significant portion thereof, as small farmers; production of bananas for
export is the primary source of income for many households, and indeed
makes up the bulk of St. Lucia's economy as a whole. (Tourism remains
a distant second, but the gap has been gradually closing in recent years.)
Most Morne Carré families also grow traditional subsistence crops
such as cassava, yams, and taro in small provision gardens. Many of
today's young adults, preferring to avoid these more traditional ways
of making a living, opt to make the arduous daily commute to wage-earning
jobs in Castries. As for education, there is a primary school in the
village, but it was not established until the mid-1970s. Today's
parents of young children are therefore members of the first local
generation to have had access to primary education. Most older villagers
never had the opportunity to attend school at all, and many are
monolingual speakers of Kwéyòl.
In Morne Carré and throughout St. Lucia today, there is a
deep-seated ambivalence toward both Kwéyòl and English. Now
more than ever before, English is undeniably the language of education,
social mobility, and communication with the wider world. But it is also
the language of colonial and postcolonial domination, and of persistently
asymmetric power relations which are experienced by many St. Lucians as
denigration and rejection when they fail to meet the standards of the
formal education system (in which Kwéyòl has no sanctioned
place) and the fairly rapidly modernizing labor market.
Kwéyòl, on the other hand, has always been the language of
home, community, and sentiment, and is the only language in which many St.
Lucians can express themselves fully and comfortably. But it is also
considered by many to be the language of backwardness, rurality, poverty,
and lack of social mobility. Attitudes toward Kwéyòl have
changed appreciably in recent years, particularly since St. Lucia's
independence in 1979, owing to the concerted efforts of a fairly small but
dedicated group of pro-Kwéyòl activists to
“preserve” and “instrumentalize” the language as a
sine qua non of St. Lucia's cultural heritage and national
identity in the postcolonial era. But as explained in Garrett 2000, there remains a sizable gulf between
attitude and practice, particularly where children are concerned. Most St.
Lucians subscribe to a sort of locally adapted notion of subtractive
bilingualism: Even among those who unreservedly express pride in
Kwéyòl, it is widely and quite firmly believed that the
simultaneous acquisition of Kwéyòl has a profoundly
detrimental effect on children's acquisition of English. As a result
of Kwéyòl's putative potential for interference –
“contamination” would not be too strong a word – adults
insist that children must learn to speak “English” (really
VESL) before they are allowed to begin speaking Kwéyòl.
While “English” must be explicitly taught to children,
Kwéyòl need not be taught at all – it will come
naturally, adults say, and they assume that children will eventually begin
speaking Kwéyòl of their own volition.
This does not necessarily prove to be the case, at least not to the
extent that parents and other caregivers assume. In households where
caregivers are especially vigilant about their children's use of
English, and where the children consequently receive limited
Kwéyòl input and have few opportunities to speak the
language, only occasionally do adults seem to notice that their
children's Kwéyòl proficiency is surprisingly lacking
– for example, when a school-age child fails to understand something
said to him or her by an elderly visitor, or even by one of the
child's own grandparents. Most St. Lucians dismissed my occasional
musings that such instances may indicate that Kwéyòl is
going into decline. “Kwéyòl is everywhere” in
St. Lucia, they assured me; children can't help but “pick it
up.” There is some truth in this: Most St. Lucian children,
particularly those living in rural villages such as Morne Carré, do
indeed “pick up” some Kwéyòl –
particularly in late childhood and adolescence, when they begin to spend
more time out of the earshot of adults and community-wide vigilance over
their “verbal hygiene” (Cameron
1995) decreases. Adults do not always realize, however, that what
children are acquiring is not necessarily full conversational proficiency
in Kwéyòl. What they are “picking up,” in many
cases, are a few relatively small sets or subsets of Kwéyòl
words and phrases, by no means random, that reveal certain culturally and
historically specific tendencies in local language socialization
practices.
LEARNING TO CURSE
Despite adults' concern that young children, from the time they
first begin to produce intelligible single words, should acquire
“English,” many do from time to time engage children in
certain kinds of Kwéyòl exchanges that provide limited
opportunities for the children to acquire some Kwéyòl as
well. The focus of the language socialization practice that will be
examined here is cursing. The communicative practice known in St. Lucia as
“cursing,” and the genre associated with it, are invariably
constituted through Kwéyòl and hence can be regarded as
code-specific; the relevant Kwéyòl verb is
jiwé ‘to curse’.33
The Kwéyòl verb jiwé is clearly
a reflex of French jurer ‘to swear; to make a vow or oath;
to blaspheme’. St. Lucians translate jiwé into
English as “curse,” however, and use it in the colloquial
sense of the English term: ‘to use vulgar, obscene, or abusive
language; to denounce or execrate by means of such
language’.
Sporadically but with some frequency, adults
actively encourage and provoke young children to “curse” in
Kwéyòl,
34 Cf. Sidnell's
(1998) examination of use of the verb
kos ‘curse’ as a “metapragmatic framing
device” among children and their caregivers in a Guyanese
community.
despite the fact that in most other contexts they
discourage or prohibit children from using Kwéyòl at
all.
Below is an excerpt from one such interaction involving a child named
Tonia. This interaction occurred about two weeks before Tonia's third
birthday; also present were Tonia's mother and Tonia's
mother's younger sister (Tonia's maternal aunt), Noelicia. I too
was present, videotaping. The setting is a standpipe where women gather,
often bringing their preschool-age children along, to launder clothes and
socialize. As this excerpt begins, Noelicia, who has been conversing with
Tonia's mother in Kwéyòl, suddenly directs Tonia
– also in Kwéyòl – to fling a curse in my
direction, apparently just for the sake of general amusement. Thanks to
just this sort of coaching on Noelicia's part, Tonia had by this time
already acquired the semi-habit of referring to me by the
Kwéyòl nickname Kaka Bouwik ‘Donkey
Shit’. (I hasten to add – and not merely to underscore a
central theme of this article – that a certain something is lost in
the translation. As with most such nicknames, I never heard Noelicia or
anyone else translate it into English. Despite their brevity,
Kwéyòl nicknames themselves can perhaps be considered a sort
of code-specific genre.35
As Bakhtin (1986:81–82) notes, genres as “forms of
utterances” are “extremely diverse in compositional structure,
particularly in size (speech length) – from the single-word
rejoinder to a large novel.” A relevant consideration here is
whether or not the vocative use of a nickname can constitute a
“complete utterance” in Bakhtin's sense. Aceto's
(2002) analysis of nicknames as a socially
embedded communicative practice in Caribbean (and other) contexts suggests
that it can and often does, as do Butler's (1997) remarks on naming or addressing as a form of
interpellation (in Althusser's sense).
) References to
Paya in this excerpt are nickname references to Tonia's
mother; unlike
Kaka Bouwik,
Paya is not also a common
noun and therefore cannot be translated. Kwéyòl utterances
are shown in the standardized orthography set forth in
Louisy & Turmel-John 1983; VESL utterances are
shown in phonemic transcription
36 The system
of phonemic transcription used here for VESL is that used by Rickford 1987 for Guyanese; Rickford's system, in
turn, is based on that devised by Cassidy 1961
for Jamaican.
and are underlined.
Tonia's response to Noelicia's initial prompting (in
utterance 1) to tell me to beat my ass – a response that evidently
takes both Noelicia and Tonia's mother by surprise – is to tell
me to beat my mother's ass (2). (Here again, something is lost in the
literal translation. The expression in 1 is rather similar to the
colloquial North American English expression Shove it up your
ass.) Mention of a person's mother is a conventionalized means
of escalating virtually any Kwéyòl curse or insult; Tonia is
displaying here a verbal skill that the women may not have been aware that
she has yet acquired. (Then again, this interpretation may give Tonia too
much credit; having heard this expression before, she may simply be
repeating Noelicia's model inaccurately, and may be innocent of how
her inadvertent embellishment alters it.) Both Paul and Mother respond
with appropriate expressions of shock (3, 4, 5). Gamely, and no doubt
encouraged by these responses, Tonia then produces another (slightly
nonsensical) variant on Noelicia's model (6), which Noelicia
gleefully echoes (7).
Mother, although certainly amused by this, takes the stance that would
be expected of any good mother by scolding Tonia, first in VESL, then in
Kwéyòl (8). Mother's threat here that I will beat Tonia
is typical; it is generally expected that the punishment for such an
offense will be meted out by the person who has been offended, regardless
of that person's relationship to the child (and without need of
explicit permission from any caregiver present).37
Caregivers generally look on approvingly if someone from
outside the household, such as a family friend or neighbor, delivers a
sound slap to a child who is misbehaving toward him or her; the prevailing
sentiment is that it serves the child right. On numerous occasions I was
encouraged to slap a child who was judged to be misbehaving toward me;
when I good-naturedly declined, as I always did, I was sometimes
admonished that children (not merely the child in question, but children
in general) would pwan'w pou papicho ‘take you for
nonsense’, that is, become accustomed to treating me as someone who
need not be shown proper respect.
Although the scold is delivered
in a serious tone of voice, Tonia, having engaged in this type of routine
before, knows that no punishment will actually befall her (certainly not
from me). Furthermore, she has a good idea of what Noelicia and the other
amused onlookers want from her, so she takes the initiative here and
flings an appropriately sassy Kwéyòl reply at her mother
(9). Her mother scolds her this time in a slightly more serious tone,
again in Kwéyòl; this is followed by a
half-Kwéyòl, half-VESL reproach (10). Note that the epithet
meaning ‘nasty dirty-mouthed thing’ is in Kwéyòl
(as in 8), but the warning that immediately follows – which is meant
to be heeded – is in VESL. (The literal translation ‘Paul is
not your friend’ fails to convey the sense of this VESL expression,
which adults use frequently in correcting children. Mother is not
suggesting that I will withdraw my friendship, as might be construed by
most speakers of North American English; rather, she is reminding Tonia
that I am not one of her age-mates, not someone with whom she can be
playful with impunity, and that as an adult I must be shown proper
respect.) Mother likewise uses VESL in 12, when she points out a potential
real-life consequence of Tonia's behavior – namely, the
possibility that I will not come to her birthday party (bearing a gift, to
be sure).
The reply to this that is suggested and then modeled by Noelicia (13)
is an appropriately coarse Kwéyòl imperative, Las bat
djòl ou. (The coarseness derives primarily from the combined
effects of the verb bat – literally ‘beat’,
used here as in the colloquial North American English expression
“Quit beating your gums” – and the noun
djòl, an impolite way of referring to a person's
mouth.) But Tonia instead produces the same insult that she had used with
me previously: Bat tjou mama'w (14). Mother's response,
probably expressing real exasperation now, is in VESL (15).
Noelicia's response, however, is an attempt to “correct”
Tonia (16), whose attention has apparently wandered to a pair of panties
(17); or it may be that she has now become uncomfortable with the tension
created by this fairly lengthy routine and is trying to change the
subject. In either case, Mother is having none of it; she follows up with
a reproachful rhetorical question in Kwéyòl (18). Noelicia
follows up on her previous correction, but it is not clear that Tonia
understands the relatively complex instructions, entirely in
Kwéyòl, that she gets from Noelicia (19); she now somewhat
listlessly reverts to the tried and true Bat tjou'w (20).
Tonia has by now become uncooperative, but Noelicia presses on, so
preoccupied that she ignores Tonia's utterance in 22, a VESL calque
on Kwéyòl Ba mwen fè sa that ordinarily
would have been noticed and corrected immediately. (VESL giv is
an overly literal translation of the Kwéyòl verb
ba.) Mother likewise ignores this, perhaps not noticing it
because Tonia is annoying her by trying to wrest a clothespin out of her
hand. This prompts a relatively lengthy Kwéyòl rebuke, the
referential meaning of which Tonia is not necessarily expected to
understand; Mother's tone of voice here, and the gestures that
accompany it, are more than sufficient to bring about the desired
response. What is perhaps most noteworthy about these final exchanges is
that Tonia, in what seems to be an attempt once and for all to break out
of the Kwéyòl routine begun by Noelicia, does not merely
attempt to change the subject (as she did in 17, unsuccessfully); she
switches from Kwéyòl to VESL, thus reverting to the code
that is normally expected of a child her age, and in which she is more
proficient.
Significantly, children older than Tonia are not engaged by caregivers
in this type of routine; Tonia's older sister and brother
(approximately 2 and 4 years older, respectively) never were, for example.
The cutoff point is about the time that the child is becoming old enough
to understand the meanings (both literal and figurative) and the pragmatic
impact of the strong words being bandied about – and old enough to
produce them without prompting. In fact, Tonia's fairly formidable
display of skills in this instance is a good indicator that she is quickly
reaching the point at which this type of routine will no longer be
age-appropriate. An older child is thought to be naturally inclined to be
disobedient and disrespectful, and therefore needs no encouragement to be
cheeky and self-assertive, verbally or otherwise. Self-assertiveness
itself does not need to be learned; on the contrary, children need to be
socialized to keep their self-assertiveness in check. But at the same
time, expressions of self-assertiveness are valued and at times rewarded,
even in children – provided that they do not violate or exceed
certain parameters such as those that become discernible in routines like
the one just examined.
As all of this suggests, there is a certain element of contradiction
or paradox in villagers' ideology of language. No one in Morne
Carré, least of all Tonia's mother, doubts that children need
to master “English,” or that they must be taught it more or
less explicitly. But at the same time it is recognized, at least
implicitly, that children also need to be able to fend for themselves in
Kwéyòl. Verbal assertiveness is an important and respected
skill in all areas of social life, as necessary within one's own
household as it is in the workplace or on the streets of Castries. And
more often than not, verbal assertiveness requires the use of
Kwéyòl, especially when it might be necessary to insult or
to curse one's opponent (either in jest or in earnest).38
Cursing is considered a prosecutable offense
in St. Lucia, as in various other West Indian societies. In Morne
Carré, a heated verbal exchange that culminates in a cursing may
result in the offended party's making a complaint to the police. But
it may take the police a matter of days to follow up on such a complaint,
by which time the complainant's sense of aggrievement may have
subsided.
Notwithstanding the undisputed value of English in more
and more domains in a rapidly “modernizing” St. Lucia, it
seems clear that a certain level of proficiency in Kwéyòl,
and above all in certain code-specific genres such as cursing, remains
useful and even necessary. A complaint occasionally heard from St. Lucian
adults who have minimal or no proficiency in Kwéyòl is that
they are at a real disadvantage: Although they may be able to understand
the gist of the abuse being hurled at them in Kwéyòl by an
antagonist, they are not able to return it nearly as effectively (which
makes it risky to even attempt to do so – for if one attempts and
falters, one opens oneself up to further abuse and ridicule). Even in
cases where the exchange is a less affectively charged one, such as a
bargaining situation, one who does not speak Kwéyòl may be
at a disadvantage. A woman from Castries who reported having been reared
in an English-only household – her father, a Barbadian, did not
speak Kwéyòl and forbade its use in the home –
half-jokingly told me that she would like to learn Kwéyòl so
as to be able “to keep up with the market women.” She was
referring to the vendors of fresh produce and other goods in the Castries
marketplace, who speak Kwéyòl almost exclusively among
themselves and with their Kwéyòl-speaking customers –
and who are reputed to give less favorable prices to non-speakers of
Kwéyòl.
LEARNING TO RECOGNIZE AND RESPOND TO
INDIRECTION
In most social situations, of course, it will not do to fling an
unprovoked curse or insult at an unsuspecting addressee, as Noelicia
encourages Tonia to do in the opening of the excerpt examined above. Even
more important than the ability to formulate such an utterance verbally
– and a rather more difficult thing both to teach and to learn
– is the ability to monitor and evaluate a social interaction as it
is unfolding, and to judge, while immersed in that interaction, when and
how it might be necessary, fitting, or desirable to deploy such an
utterance.
Often, such an utterance is called for as a retort to a conversational
move made by another, or as a response to a stance taken (either verbally
or nonverbally) by another. The excerpt below is taken from the same
recording as the previous one and occurs shortly thereafter. The location
is the same, but the group has now been joined by Andrea, another of
Tonia's maternal aunts, and Sharice, Tonia's 7-year-old cousin.
As the excerpt begins, Andrea, gazing at Tonia, initiates a teasing
routine by casually venturing the opinion (in Kwéyòl) that
Tonia's older sister Brita is pli bèl ‘more
pretty/prettier’. (Brita, the older sister referred to, is not
present.) This invidious comment – ostensibly directed to the adults
present, but actually intended to provoke Tonia and get a response out of
her – fails to do so until Tonia's mother begins coaching Tonia
in how to recognize and respond to this oblique verbal affront:
Andrea's invidious remarks are at first made without explicit
reference to Tonia (utterances 1, 3), although the construction pli
bèl ‘more pretty, prettier’ indicates that she is
making a comparison to someone. (This type of provocative indirection is
common, and is quite similar to “signifying” as described in
the literature on African American Vernacular English, e.g. Mitchell-Kernan 1972, Morgan 1996.) For Tonia, this is part of the lesson to be
learned – how to recognize when she is the intended recipient of
such an obliquely directed remark. In this instance, despite an additional
cue from Mother (4), Tonia gives no indication that she understands
Andrea's intent. This leads Andrea to reformulate her comment in a
more explicit fashion (5); even here, however, she addresses the comment
not to Tonia but to Mother and Noelicia. Sharice, Andrea's 7-year-old
daughter, attempts to join in the routine (6); but misapprehending what is
going on and thereby displaying incompetence, particularly with regard to
the relevant participation framework (Goodwin
1990), she is ignored by all.
Tonia still has shown no sign of understanding what is going on, so
her mother briefly explains – in Kwéyòl, but in
relatively simple, direct terms that Tonia can be expected to grasp
– and urges her to take immediate verbal action (7). Mother does not
specify what action would be appropriate, however. Tonia responds by
calling Andrea ugly (8), but it is quite possible that she is merely
responding to Mother's prompt, rather than returning Andrea's
previous verbal volleys; it is still not clear that she understands what
is going on. Sharice again tries to join in, this time making a somewhat
more germane contribution (9); but by the implicit “rules” of
the game being played, it is for Andrea to make the next move here, not
Sharice, and she is again ignored by all.
Indirection is now abandoned by Andrea, who is obliged to respond to
Tonia's direct insult with a direct insult of her own (10). This time
Tonia is prompted to respond by Noelicia, who offers a model utterance for
Tonia to repeat (11–12) – the same one that was directed
toward me in the previous routine. Sharice again tries to join in, again
blunders – this time by bringing up another, irrelevant party,
Tonia's preverbal infant sister Jessie (as well as by intruding on,
or at least trying to insinuate herself into, the exchanges between Tonia
and Noelicia) – and continues to be ignored by all (13). Andrea
responds to Tonia with a simple reversal of Tonia's insult (14). She
is purposely keeping the game down at Tonia's level; had an adult
issued the same insult, it would have been incumbent upon Andrea to come
up with a more creative and virtuosic response. Noelicia now suggests that
Tonia raise the stakes a bit by mentioning Andrea's mother (15).
Andrea's mother is Tonia's grandmother and one of her primary
caregivers, but that is irrelevant here; mention of a person's mother
in a playful exchange such as this is understood to involve a certain
degree of abstraction. Now the routine gets hurried along, probably
because the adults are growing bored with it and realize that they have
reached the limits of Tonia's present abilities anyway. Without
waiting for Tonia to repeat Noelicia's prompt, Andrea responds to it
as if Tonia herself had delivered the insult (16). This occurs one more
time (17–18) before the women abandon the routine and return to
their own conversation.
As mentioned previously, adults no longer engage children in such
routines once they become slightly older than Tonia is in this excerpt.
Notice that Sharice makes three attempts to participate (6, 9, 13), but is
simply ignored by the others. At 7 years of age, her time for such things
has passed. Her primary concern should be to learn English well; indeed,
her limited use of Kwéyòl here might not even have been
tolerated had it not occurred within this particular interaction, or
rather on the periphery of it (Lave & Wenger
1991). Children Sharice's age are regarded by adults as
needing no encouragement in being self-assertive, nor in the use of
Kwéyòl; on the contrary, they need containment and rigid,
no-nonsense discipline, and their developing English needs to be monitored
and reinforced. Furthermore, older children do not need or deserve any
assistance in fending for themselves in verbal confrontations, which occur
frequently among peers; if such a confrontation is overheard by an adult,
she or he is more likely to warn the children involved not to use overly
strong language (such as Kwéyòl “curses”) than
to squelch the confrontation altogether. For these and related reasons,
the blunders that Sharice makes in trying to participate in the routine
examined above go ignored and uncorrected. On one hand, her evident lack
of skill notwithstanding, she is a few years too old for the kind of
verbal coaching that Tonia receives; on the other hand, she is several
years too young to join the women in engaging Tonia in this kind of
routine (or otherwise to be part of their Kwéyòl
conversations). As party to this interaction, then, the only
age-appropriate thing for Sharice to do is to learn by silent observation.
Her learning, it is worth noting, can be on either of two levels (or
perhaps both): She can learn the same lessons being taught to Tonia (how
to interpret and respond to indirection, how to manage a confrontation,
how to be verbally assertive), and/or she can learn how persons older
than herself engage children younger than herself in this type of
socialization routine.
It is not so much the Kwéyòl language that is being
socialized here as it is a particular affective stance – a way of
comporting oneself in certain types of confrontational situations that
arise quite frequently in local social life. (The notion of
“teaching” Kwéyòl to a child, whether directly
or indirectly, would be a rather odd one to most St. Lucian adults, and
anathema to some.) The fact that the code deemed most appropriate for
communicating and displaying the affective stance in question is
Kwéyòl, it would seem, is little more than incidental. The
crucial lesson here is that one must be able to fend for oneself, to make
known one's needs and wants, to stand up for one's rights, to
demand respect, and to be prepared to give as good as one gets. This is
taught more or less explicitly to young children in various other ways
that are less directly dependent on language. Preschool-age children are
often teased by adults, both verbally and physically, and are encouraged
by their caregivers to stand up to their mock tormentors. Just as a mother
will look on approvingly if an unrelated person (e.g. a family friend, or
even a casual acquaintance of the mother whom the child has never before
encountered) delivers a bit of justly deserved punishment to a 3-year-old
child who has misbehaved toward him, she will stand by as he teases the
child, verbally and/or physically (e.g. by threatening to take away a
bit of food or a toy), to the point that the child begins to cry and hide
behind the mother's legs. At that point she will either ignore the
child or encourage him or her to act assertively, often modeling an
appropriate utterance or physical action.
INDEXICALITY OF CODE-SPECIFIC GENRES
In effect, what is being socialized in the excerpts examined above is
how to conform one's utterances to particular genres – genres
that are code-specific and that highlight self-assertion. Like all genres,
those examined here bear the imprint of the sociohistorical contexts out
of which they emerge; and like most, they are colored by affect. As Kulick
& Schieffelin (2004:352) note,
“Everyone has ideas about and conventions for displaying, invoking
and interpreting affect … [T]he ability to display
culturally intelligible affective stances is a crucial dimension of the
process of becoming a recognizable subject in any social group.”
Genres are surely a crucial means by which these ideas, conventions, and
stances are organized, understood, and enacted by speakers, and by which
they are socialized.
Bakhtin 1986 discusses the affective
dimension of genre in terms of “expressive intonation.”
Utterances, according to Bakhtin, are produced by the speaker and
apperceived by others not as sentences (which as mere “units of
language” are expressively neutral, in his formulation), but as
“kinds of speech genres that embody typical expression … One
of the means of expressing the speaker's emotionally evaluative
attitude toward the subject of his speech is expressive intonation, which
resounds clearly in oral speech” (1986:85). Developing further his analytic distinction
between the sentence and the utterance, he notes, “A speech genre is
not a form of language, but a typical form of utterance; as such the genre
also includes a certain typical kind of expression that inheres in
it” (87; emphasis added). Bakhtin does not propose a mechanism
by which this inherence comes about, however. The existence of
code-specific genres, as has been proposed here, suggests that those
genres that, taken collectively, typify the characteristic ways in which a
particular code is used in a bilingual/multilingual setting leave a
certain “imprint” on that code via the code-specific genres
with which it is associated; and that this is a central mechanism in the
semiotic process by which particular codes in bilingual/multilingual
settings come to be laden with specific, more or less conventionalized
(though still open-ended and negotiable, and hence fundamentally emergent)
social meanings.
But what is the nature of that semiotic process? Code specificity of
genres is, fundamentally, a matter of indexicality of the type that Silverstein 1976 refers to as
“nonreferential” indexicality. Nonreferential indexes are
“features of speech which, independent of any referential speech
events that may be occurring, signal some particular value of one or more
contextual variables” (1976:29).
Silverstein uses as his examples language-internal sets of differentiated
forms, such as the speech levels of Javanese by means of which speakers
index differences of status, rank, age, and other culturally salient
variables. Likewise, in virtually any sociolinguistic setting in which two
or more distinct codes are in use, speakers tend to exploit code selection
for broadly analogous nonreferential indexical functions; as Ochs (1990:293) notes, various categories of sociocultural
information (genre and affective stance among them) may be indexed in this
way. Ochs goes on to note that such indexing may be accomplished by means
of either a single linguistic feature or a set of such features. In the
latter case, which she refers to as “collocational” indexing,
“contextual information is indexed through a set of co-occurring
structures … Here social identity of speaker or addressee, genre,
communicative activity taking place, and the like are indexed through a
set of linguistic features that systematically co-occur rather than
through a single feature.”
Code-specific genres such as have been described here provide prime
examples of collocational indexing: Code and genre are tightly linked, and
they index accordingly. That is, it is not just a matter of code indexing
genre (although certainly this is the case); code and genre
together index some “local contextual dimension”
(Ochs 1990:293) – a particular culturally
salient affective stance, in the examples examined here. As Ochs observes,
the implications for language socialization are significant:
That context is indexed through collocation of indexes as
well as through single indexes is a sociolinguistic generalization that
must be incorporated into a developmental model of how sociocultural and
linguistic knowledge interface in the course of language socialization. We
must represent the fact that children come to understand constraints on
co-occurrence and ordering of indexes and, further, that they come to
understand how indexes interact to signal contextual information. (1990:294)
As this suggests, language socialization practices may be a
particularly important site for examining code-specific genres, across
social and cultural contexts as well as across historical contexts –
for example, as a means of shedding light on historical accounts from the
Caribbean colonial era that mention (usually with shock and disapproval)
the children of European planter families speaking the creole language of
their African caregivers and playmates. Likewise in contemporary contexts
in the USA, as Schieffelin notes,
[W]e need to acknowledge … and take into
account the many nonstandard English speakers and non-English speakers who
are employed to take care of the children. We cannot ignore the fact that
these caregivers speak to other people's children similarly to the
ways in which they speak to their own, using the language(s) in which they
are most comfortable and that they feel to be the most appropriate to
express what they need to say. (1994:38)
Schieffelin's primary concern here is with the implications of
code-switching for language socialization, particularly the ways in which
code-switching constitutes a source of information for the child about
language, social context, and relationships between the two. The
implications of code-specific genres are worth consideration along similar
lines, in that genres constitute an intermediate level in the organization
of talk – between code and context, or in broader terms, between
language and social structure – by means of which children learn
about the codes in their environment, the social order within which they
function, and the varied ways in which the former constitute and organize
the latter. Studies of language acquisition in bilingual/multilingual
settings, including those in which code-switching is prevalent, have
demonstrated that code-switching and other types of mixed input do not
present special problems for the language-acquiring child, and that the
child is able to differentiate languages from a very early age (Döpke 2001). Just how the child is able to do so,
however, remains to be addressed. Attention to genres, particularly
code-specific genres, makes possible a more adequate account of how
code-switching serves as a source of information to the child in the ways
that Schieffelin suggests.
CONCLUSION: CODE-SPECIFIC GENRES AND THE
TENSION BETWEEN PERSISTENCE AND SHIFT
Research concerned with the question of why nonstandard and
“low-prestige” varieties persist in the face of strong
hegemonic pressures from dominant and standard varieties has often
stressed their function as “markers” of group identity, and
the issue of “solidarity” more generally. Too often this has
led to simplistic conclusions such as that offered by Ryan (1979:155): “[L]ow-prestige speech
varieties persist basically because people do not want to give them
up.” Other accounts, more attentive to situated language use, find
in the persistence of these varieties a “seamless discourse”
of resistance; but such approaches also have their shortcomings,
particularly when the result is “an uncritical romanticism of
resistance everywhere” (Limón
1996:188). Meanwhile, some of the most sophisticated ethnographic
studies dealing with nonstandard varieties and their speakers (e.g. Heath 1983, Goodwin 1990) do
not explicitly address the issue of persistence as such, focusing instead
on the richly complex ways in which speakers draw upon these varieties in
constructing and sustaining their culturally distinctive social worlds.
Although these studies tend to focus on communicative practices much like
those that have been considered here, they do not explicitly problematize
the relationship between these practices and the language varieties
through which they are constituted and enacted.
The St. Lucian case suggests that investigators seeking to account for
the persistence of a particular code in a situation of contact would do
well to focus less on the code itself than on the genres through which
that code is instantiated in the course of everyday interactions. Insofar
as those genres persist, particularly those that are code-specific, the
code persists as well. But depending on the number of code-specific genres
and the level of linguistic proficiency that they require – cursing,
for example, does not require a particularly high level – the code
may persist only in an attenuated form, and may remain vulnerable to
processes of language shift. This seems to be the case at present in St.
Lucia. Bakhtin (1986:80) notes, “Many
people who have an excellent command of a language often feel quite
helpless in certain spheres of communication precisely because they do not
have a practical command of the generic forms used in the given
spheres.” As this suggests, command of a code and command of a genre
are of two different orders. The St. Lucian case, like those described by
Dorian 1982 and Jackson
1974, shows that in some bilingual/multilingual settings,
speakers who have good command of a particular set or subset of
code-specific genres may get along quite comfortably and effectively in
everyday interaction despite having very limited command of the code
itself. As Hanks points out,
In order for two or more people to communicate, at whatever
level of effectiveness, it is neither sufficient nor necessary that they
‘share’ the same grammar. What they must share, to a variable
degree, is the ability to orient themselves verbally, perceptually, and
physically to each other and to their social world…. This kind of
sharing – partial, orientational, and socially distributed –
may be based on common schemes of perception that go far beyond language.
(1996:229)
As Hanks suggests, we must look beyond code to other dimensions of
communicative practice, broadly conceived, in order to understand how
intersubjectivity is achieved and sustained in such situations. Genre,
understood to comprise affective stance and other suprasegmental and
paralinguistic features of language use, is one such dimension (Hanks
1996:242–46).
The notion of code-specific genre can help account for why
“low-prestige” varieties persist as well as why, in the St.
Lucian case and numerous others, they tend to persist (if only in
attenuated form) through cursing and other genres in which self-assertion
is highlighted and solidarity is created. (The paradox here is more
apparent than real, as various studies of ritual insulting and other
vernacular genres suggest, e.g. Mitchell-Kernan
1972, Limón 1996, Morgan 1996.) At present in St. Lucia, notions of what it
means to be St. Lucian are a matter of ongoing negotiation and
contestation – as they have been for some time, particularly since
the local dawning of the postcolonial era in the 1970s. Language-related
issues, particularly the relationship between Kwéyòl and
English and the relative importance of Kwéyòl as an aspect
of St. Lucian identity, figure prominently in current debates, though not
always explicitly (Garrett 2000). As mentioned
earlier, many St. Lucians of the current, largely bilingual generation of
young adults take for granted that all St. Lucians speak
Kwéyòl, and that their own children likewise will eventually
speak it. But the growing numbers of young St. Lucians who in fact do not
speak Kwéyòl proficiently, and may not even understand it
well, is a phenomenon with which all St. Lucians, even those living in
rural villages such as Morne Carré, are now confronted from time to
time. Must one speak Kwéyòl in order to be regarded as a
“real” St. Lucian by other St. Lucians? The answer at present
would seem to be a qualified no. That is, a St. Lucian need not be a
fluent speaker of Kwéyòl, but in order to avoid the
sometimes disadvantageous or ambiguous status, particularly in
(mock-)confrontational situations, of the St. Lucian who does not speak
any Kwéyòl at all, she or he must at least be able to use
Kwéyòl in what Jackson 1974
characterizes as an “emblematic” fashion. And in order to use
Kwéyòl emblematically, one need not master the code per se;
it is sufficient to master a few salient code-specific genres such as
those that have been examined here.
As objects or units of analysis, genres, like languages, are
reifications – abstractions from the dynamic flux of communicative
practice. Thus, the notion of genre offers advantages as well as potential
pitfalls. Analytic emphasis on the conventionality, stability, and
durability of genres is of great theoretical utility in that it allows one
to discern areas of relative stability in the aforementioned flux and to
render them amenable to analysis and theorizing. But it must always be
borne in mind that the stability and durability of any given genre are
only relative, and that there is considerable variation among genres along
these very dimensions; some genres remain fairly resistant to innovation
and transformation for remarkably long periods, while others morph more
readily and more rapidly into new forms. In Bakhtinian terms, genres are
subject to the same centrifugal and centripetal forces as languages, and
their coherence (and hence their identifiability and utility as units of
analysis) ultimately derives from the same kinds of resultant tensions
(Bakhtin 1981, 1986).
Bakhtin (1986:65) asserts that genres are
“the drive belts from the history of society to the history of
language.” Situated between social structure and linguistic
structure, partaking of both and sensitively registering changes in both,
genres function as conduits linking the two in complex dialectical
relationships. As this suggests, genres are centrally involved in
processes of sociocultural and linguistic persistence, stability, and
continuity as well as in processes of transformation and change. This
makes them particularly valuable as units of analysis for investigating
cases of linguistic and cultural contact, which often present intriguing
combinations of profound conservatism and radical innovation – as
can be seen in the sorts of dynamic processes commonly characterized as
creolization, hybridization, and syncretism. The language socialization
routines that have been examined here can perhaps be read broadly as a
form of “resistance” or, taking a more nuanced view, as a
means by which speakers open up an alternative social space in which they
collaboratively “negate the alienating constraints of the
historically given social order … and affirm the possibilities of a
different social order,” as Limón (1996:193) puts it in his insightful study of
working-class Mexican-American men in south Texas. Such interactively
constructed social spaces can even be regarded as sites of potentially
profound and far-reaching social change, as Rampton 1995a, 1995b suggests in his
study of “crossing.”
But St. Lucians' code-specific practices, and the genres and
social spaces that they constitute, can perhaps just as well be regarded
as sites of continuity – a means by which St. Lucians are holding
onto and reproducing a distinctively St. Lucian sense of personhood, even
as much else is changing in their society. Urban (1985:326) suggests that “the contexts and
subject matters most focal in a given cultural system will be those
highlighted by the most salient speech styles.” The features that
characterize a given speech style, Urban proposes, will be those
“whose indexical meaning in the everyday code or elsewhere is
appropriate to the overall indexical function of the speech style as sign
vehicle” (324); thus the salient constituent features of Shokleng
ritual wailing, such as “creaky voice” and “broken
voice,” are those that occur as “pragmatic alternatives to
ordinary voice in the everyday code” and that index intense feelings
of grief (322). Essentially the same semiotic process is at work when
(semi-)bilingual St. Lucian adults who generally avoid speaking
Kwéyòl with children use that very code – which in
everyday usage is associated with directness, straightforwardness,
frankness, and emotional expressiveness – in certain types of
socialization activities that involve the playful instigation, enactment,
and display of self-assertive stances. Recent changes in the
sociopolitical status, “prestige,” and symbolism of
Kwéyòl notwithstanding, when St. Lucians socialize children
to “curse” or otherwise to speak and act self-assertively by
means of Kwéyòl, the central issue at stake is less what it
means to be St. Lucian than what it means to St. Lucians to be a person
– that is, a socially competent individual and community member.
While the former has changed significantly in recent years and remains a
point of negotiation and contestation, it would seem that the latter has
remained quite stable.
In all communities, certain core cultural values, which may generally
remain tacit, are socialized through language. Such values are
instantiated in and through particular genres, which may or may not be
code-specific (see Field 2001 for a case in
which they apparently are not). If the relevant genres are
code-specific, as in the St. Lucian case, caregivers can be expected to
make certain focused efforts (though not necessarily conscious efforts) to
ensure that both code and genre are being reproduced to the extent
necessary for children to be able to demonstrate their awareness of that
value in conventionally expected, locally intelligible ways. Where
language shift is underway, this in turn may have an overall retarding
effect on the process; or it may present a sort of snag or sticking point,
such that language shift proceeds somewhat unevenly, giving rise to a
situation in which certain vestiges of the obsolescent code linger on and
perhaps ultimately become assimilated into the code toward which the
community is shifting.39
It would seem that
this has been the case in Trinidad, Grenada, and possibly other Caribbean
islands where varieties of French-lexified creole closely related to that
of St. Lucia were formerly spoken (Garrett
2003). In these islands the French creole has now been almost
entirely supplanted by local varieties of vernacular English or
English-lexified creole, but numerous vestiges of French creole remain in
common use. Significantly, prominent among these are local labels for
precisely the kinds of genres that have been considered here, such as
konmès ‘gossip; slander; a state of noisy confusion
or disturbance’ and mako, a verb meaning ‘to pry into
or be overly attentive to others' affairs; to be nosy’, and
also a noun meaning ‘busybody; nosy or overly inquisitive
person’ (Allsopp 1996). Not surprisingly,
Kwéyòl words such as these are also common in
VESL.
Language shift is never a simple or straightforward process, and is
notoriously unpredictable. Even so, the study of language shift and
obsolescence, particularly ethnographically oriented research since the
late 1970s (e.g. Gal 1979, Schmidt 1985, Hill & Hill
1986, Dorian 1989, Kulick 1992, Errington 1998,
Tsitsipis 1998, Jaffe
1999), has yielded important insights that help delimit the range
of outcomes to which a given case of language shift may give rise. As
these studies demonstrate, a multitude of factors – some
self-evidently language-related, others much less so – can
accelerate or decelerate the process of shift, and still others can
introduce unexpected twists and turns. The value of a language
socialization approach to language shift and other phenomena associated
with language contact has been demonstrated in several recent studies
(Garrett & Baquedano-López 2002); and
as Kulick & Schieffelin point out still more recently,
language socialization studies can document not only how and
when practices are acquired, but also how and when they are acquired
differently from what was intended, or not acquired at all. Hence,
reproduction is not assumed, and unintended consequences of socializing
practices, or change, can be documented and accounted for in empirically
delineated social contexts. (2004:352)
A language socialization approach that brings to bear the notions of
code specificity and code-specific genre can help elucidate such processes
and their implications for larger issues of linguistic and cultural
reproduction and change.
