Re: [code-switching] Re: Language alternation and code-switching
>>>Establshing definitions for any widely used terminology is always difficult. I
think one of the key reasons why descriptive categories for behaviour, such as
our topic 'code-switching' or 'code switching', are often applied to quite
different groups or phenomena, and with attention by researchers to differing
aspects of the environment, or context, in which such phenomena occur is that
different data sets call attention to different areas of interest.
'Code switching' research incorporates studies which approach mixed language
behaviour from the social and pragmatic perspective, looking perhaps at the
distribution and functions of language varieties (codes) in a particular
community and investigating what particular roles and meanings these varieties
and patterns of their use encode in local discourse. 'Code switching' research
also incorporates studies which approach mixed language behaviour from a
structural perspective, investigating formal relations between varieties.
The division of topic between these subfields may seem to justify closer
distinctions in the terminology we use to describe such research, but I feel
that maintaining the use of a single cover-term is practical and useful.
Research into mixed language behaviour is now in a phase where different
paradigms are becoming established and starting to cross-fertilize. Under the
rubric of 'code-switching' we have seen research into patterns of multilingual
and multilectal expression flourish across the fields of linguistics.
Admittedly, formal and functional studies have rarely gone hand in hand. Myers-
Scotton, for example, chose to publish her social and structural analyses or
mixed language behaviour, and specifically code-switching, in East Africa as
separate works in 1993. However, if we choose to pick separate terms for
sepaprate areas of research we are in danger of unravelling the network of
associations which are flourishing in this field.
While some 'code-switching' research focuses on pragmatics and some on
structure there is shared ground in the apparant distinction between
linguistic expression which calls upon the resources of a single 'code' and
that which utilizes the resources of different systems.
On the question of language varieties and dialects it is this issue of
systemacity which is most relevant. Where we can define a discrete and
systematic 'code' we have a variety which can be distinguished from
other 'codes' where they are used discretely. All code-switching research
selects as its topic behaviour where different 'code' systems are used
I would resist presenting 'Language Alternation' as a more appropriate super-
category to 'code-switching' on the grounds that to use the word
and 'language' to define an area of research which naturally distinguishes
between more discrete subjects as 'languages' and more similar subjects
as 'dialects' within the same language in the process of establishing similar
patterns in their combination is to limit ourselves unduly.
Similarly, 'alternation' does not seem an appropriate super-definition for
phenomena which can include insertional patterns and congruent
grammars. 'Language Alternation' seems an ideal title for a particular
category of 'code-switching' research, just as 'dialect alternation'
or 'language insertion' might be.
The more neutral term 'code-switching' remains valuable to because it allows
for researchers into, for example, dialect mixing at the phonological level to
network their theories and results with, for example, researchers into the
expressive function of mutlilingual discourse in constructing particular
identities. As a super-category, 'code' is neutral with respect to the kind of
system under study, and 'switching' does capture the perceptual salience of
those moments when we notice language varieties in contact, while still
remaining neutral as to whether the general patterns the 'switching' takes are
alternational, or insertional, congruent, etc.
Personally, I like to think of my broad field of research interest as
being 'Mixed Language Behaviour', a term I have used in my comments above,
without being particular about whether I am considering languages, or language
varieties, or dialects, or exactly what kind of behaviour is under the
microscope. Under this category I consider a range of differing subjects, such
as second language acquisition, pidgin formation and decreolisation, and code-
switching as seperate subfields resulting from their psycholinguistic, social
developmental, and pragmatic and strucural focus. Each of these established
fields differ in the paradigms they follow. I resist a definition of 'code-
switching' as being limited to dialects or varieties within a single system on
the grounds that the term 'code' seems to have a more pwerful scope than that.
What is clear, however, is that for 'code-switching' research to fulfill its
potential, individual practitioners need to think for themselves about the
paradigm that they are a part of, and how it relates to the philosohpy and
practices of other researchers in the field. Celso's category of 'Langauge
Alternation' I would still consider a part of 'code-switching' research, as I
consider 'language' a type of 'code' and 'alternation' a type or 'switching',
but I think that he is right to make a distinction for his own work as he
establishes a stronger paradigm for his research.
I think that the established super-category encourages shared thinking, and
this should not be diminished. However, we should all be aware that 'code-
switching' research is diverse and comewhat fragmented. We are all in the
process of defining our own paradigms. Giving them names is part of a process
which helps us find our way.
2 kina blo mi tasol
Peace and love
Quoting "cdnilep <Chad.Nilep@...>" <Chad.Nilep@...>:
> --- In firstname.lastname@example.org, Celso Alvarez Cáccamo--
> <lxalvarz@u...> wrote:
> >I personally present the view that we
> > should revisit and reframe the "code-switching" notion, and I
> propose to
> > distinguish between, precisely, language alternation (LA) and
> > code-switching (CS)
> I find this a facinating, and potentially very useful distinction.
> However, I have a few questions or issues relating to your
> description of the problem.
> First, you assert, "[C]ode-switching theoretically involves the
> alternation of systems which are not distinct 'languages', but
> dialects, registers, etc." It is not my understanding that code-
> switching necessarily refers to dialects or registers of a single
> language. While seminal work such as Blom and Gumperz (1972)
> sometimes deals with alternation among dialects, Gumperz (1982) goes
> to some trouble to differentiate code-switching from diglossia
> (Ferguson 1964). Did you mean to say that CS can involve switching
> either between distinct languages OR between varieties within a
> language? (At any rate, it has always been difficult to
> define 'language' in a way that consistently differentiates sister
> languages from, for example, dialects.)
> As you point out, your view of language alternation versus code-
> switching is not the orthodox view. Not all linguists (indeed,
> perhaps only an minority) see code-switching as involving
> recontextualization "by definition." There has been a good deal of
> work, for example, by Minimalist syntacticians or by phonologists
> which is labled "code-switching" research, but has very little to do
> context, or indeed meaning.
> That said, within the areas of linguistics which I personally
> study (pragmatics, linguistic anthropology, etc.) the distinction
> you draw between language alternation and "code-switching" per se is
> a useful one. A problem arises, perhaps, with the terminology used.
> You are using "code-switching" in a different, more
> specialized way than, say, formal linguists who use the same term to
> describe any language alternation. I would be interested to hear
> what others on this list think about the distinction Celso makes,
> and whether your own definitions of code-switching are similarly
> precise. Does it seem useful (or even possible) to reframe the
> notion of code-switching as Celso suggests, or is the term too
> broadly used by linguists in various subfields?
> Chad D. Nilep
> University of Colorado at Boulder
> To Post a message: email@example.com
> To Unsubscribe, send a blank message to:
> Web page: http//groups.yahoo.com/group/code-switching
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
Adam Blaxter Paliwala MA
University of Sydney
Department of Linguistics
Australia NSW 2006
+61 4 1489 3136
This mail sent through IMP at ArtsIT: http://admin.arts.usyd.edu.au/horde/imp/
There have been a couple of very useful messages in reply to my own on CS
and language alternation. Thank you.
Chad Nilep wrote:
>First, you assert, "[C]ode-switching theoretically involves theYou're absolutely right, of course, I forgot "necessarily", that is, I
>alternation of systems which are not distinct 'languages', but
>dialects, registers, etc." It is not my understanding that code-
>switching necessarily refers to dialects or registers of a single
meant to say:
the most widespread view of code-switching theoretically involves the
alternation of systems which are not NECESSARILY distinct 'languages', as
everyone knows... I apologize if this created confusion in my message.
Adam Blaxter Paliwala wrote:
>'Language Alternation' seems an ideal title for a particularWell, I thought I was clear that LA stands for alternation between any type
>category of 'code-switching' research, just as 'dialect alternation'
>or 'language insertion' might be.
of speech varieties. What I meant to say is that LA is the surface
phenomenon by which signals from (seemingly distinct) varieties A and B
appear sequentially (thus being amenable to grammatical inspection),
whereas CS is the underlying recontextualizing phenomenon which may or may
not be signalled by LA (language-variety alternation, 'insertion' or the like).
In _A Theory of Semiotics_, Eco goes at length to exemplify two types of
'codes' that roughly correspond to what I'm talking about (Eco, Umberto.
(1976): A theory of Semiotics. Bloomington: Indiana University Press,
particularly pp. 32ff). I'll try to oversimplify Eco's dense arguments
with a linguistic example.
Let us take classical LA examples of the type:
"Y en Puerto Rico he would say que cortaba caña, even though tenía su
negocio, you know"
(From Sankoff, David, and Shana Poplack (1981) A formal grammar for
code-switching', Papers in Linguistics: International Journal of Human
"It's creating hard feelings avec les Anglais"
(Heller, Monica S. (1992) The politics of codeswitching and language
choice', in Carol M. Eastman (ed.) Codeswitching, Clevedon: Multilingual
I'll focus only on
"He would say que cortaba caña"
Let's represent it as Eng-Sp (English=>Spanish).
In this utterance, I propose to distinguish between two possible processes:
(a) The processes by which linguistic meanings (about 'third person
masculine', 'say', 'cut' 'cane', etc.) are coupled with surface signals
belonging to two apparently distinct systems (Eng and Sp), such as ['hi],
['wUd], ['seI], [ke], [kor'taBa], ['kaÑa], etc., and ordered that way. (b)
The process by which (or not) the ordered coupling of meanings with surface
signals in Eng and Sp encodes or not communicative meanings that would be
different IF the speaker had uttered instead:
-- Él decía que cortaba caña [Sp-Sp] OR
-- Él decía that he used to cut cane [Sp-Eng], OR
-- He would say that he used to cut cane [Eng-Eng]
The Eng and Sp systems working at the first (a) level can be understood as
Eco's S-codes (standing for "Syntactic/Semantic codes"), whereas the
signification mechanism underlying process (b) would be, in my mind, what
Eco simply calls "the code". Code-switching, to me, occurs in the second
process. [Now, in my opinion the ordering of processes is the opposite:
first comes the encoding of overall pragmatic, interactional contents and
intentions through (b), then comes linguistic and other types of signalling
Of course, whether a given Sp-Eng or Eng-Sp sequence does or does not carry
interactional meanings going beyond semantic coding is an empirical
question. Very often, these alternations are automatic, and are then
comparable to phonological "coding".
So, let me adventure more precise definitions about the interactions
between CS and LA:
- CS with LA is the phenomenon by which something X is communicated by
means of two linguistic signalling systems A and B on a given occasion when
a different combination of A and/or B signals with roughly equivalent
propositional content by the same speaker would mean something different at
socio-interactional levels, that is:
The AB sequence =/= (not equals) BA OR =/= AA OR =/= BB
- LA without CS is the phenomenon by which the juxtaposition of A and B
signals on a given occasion is communicatively roughly equivalent to other
possible A and/or B combinations with equivalent propositional contents,
AB = AA = BB = BA
Now, it may very well be the case that where social norms for frequent
alternation are flouted by, precisely, a speaker's sticking to
*monolingual* discourse. That is, where an A=>B alternation should be
expected, a monolingual A=>A sequence is produced instead. In that case,
how to make sense of and explain the *recontextualizing* effect that the
*monolingual* sequence may have in regards to expectations about bilingual
behavior? Simply: there we probably have communicative code-switching
between the two parts (A, and A), WITHOUT language alternation. I couldn't
be more structuralist here!: where two paradigmatic alternates can occur,
and the occurrence of one is umarked (i.e. alternating languages), its
non-occurrence and the occurrence of the other is marked (i.e.
I admit that I am not very up-to-date in recent grammatical work on LA, so
please do not flack me. I very much agree with Adam on the necessary
convergence of viewpoints on the study of 'mixed language behavior', and it
was not my intention at all to point at redrawing disciplinary boundaries
(for which I am in no position at all!). Simply, my point is to emphasize
the "switching", communicative part of this mixed (or not!) behavior, and
to re-center the CS notion precisely on that communicatively active part.
Therefore, research on syntactic constraints on language alternation is of
course very relevant to CS the way I've proposed. However, it is not
directly relevant as to the occurrence of CS, but as to the grammatical
conditions of possibility by which speech alternation between a given pair
of languages/varieties (LA) MAY on a particular occasion signal (manifest,
carry, indicate) recontextualization ("otherness", as Auer would say)
Celso Alvarez Cáccamo
Linguística Geral, Fac. de Filologia
Universidade da Corunha
15071 A Corunha, Galiza (Espanha)
Tel. +34 981 167000 ext. 1888
FAX +34 981 167151