Code Switching

Code switching (hereafter CS), a type of discourse that occurs as a natural

outcome of language contact and an inevitable consequence of bilingualism, has
attracted linguists’ attention and been studied from a variety of perspectives.
Scholars do not seem to share a single definition of the concept, and this is perhaps
inevitable, given the different concerns of formal linguists, psycholinguists,
sociolinguists, anthropo-linguists and so forth. Many scholars use a definition of CS
similar to Heller’s (1988a:1): “the use of more than one language in the course of a
single communicative episode”; Auer (1984:1), for example, pints out that CS is “the al-
ternating use of more than one language”; while Milroy and Muysken (1995:7) rec-
ognize that it is: “the alternative use by bilinguals of two or more languages in the same con-
versation” (Quoted inBoztepe, 2008 : 4).Whatever the definitions are, it is obvious that any-
one who speaks more than one code switches between them or mixes them according to certain
circumstances.

On the light of Trudgill' quotation (1992:16) CS is, however: “the process

whereby bilingual or bidialectal speakers switch back and forth between one
language or dialect and another within the same conversation”. Such a definition
clearly denotes that CS can occur in a monolingual community, or in a plurilingual
speech collectivity. In a monolingual context, CS relates to a diglossic situation
where speakers make use of two varieties for well-defined set of functions: a H
variety, generally the standard, for formal contexts, and a L variety typically for
everyday informal communicative acts. In addition to alternation between H and L
varieties, speakers may also switch between the dialects available to them in that
community via a process of CS. In such a case, i.e. monolingual context, CS is classified as
being ‘internal’, as the switch occurs between different varieties of the same language. In a
multilingual community, the switch is between two or more linguistic systems. This is referred
to as ‘external’ CS.
 Studies of CS can be divided into three broad approaches: structural, psycholinguistic
(is not included here) and sociolinguistic as it will be explained below.

The Sociolinguistic Approach to Code Switching

The role of sociolinguistic studies is to answer the broad general question:

“why do bilinguals switch languages?”. In other terms, sociolinguistic research
deals with CS as a process. By the way, it is wiser to return to Muller and Ball’s
distinction (2005:51) between CS as a product or a process who stated that a first
distinction is “whether our focus of analysis is going to be the language (talk, writing)
produced, and preserved in some medium […], or the process of producing language”.
More precisely, CS as a product attempts to solve the question ‘where
does CS occur, and how it is patterned’, i.e. it identifies syntactic and morpho-
syntactic constraints on CS; it also investigates the possible role of CS in textual
organization, stylistic features or levels of formality. CS as a process, on the other hand, deals
primarily with CS as “behaviour”, i.e., an aspect of a speaker's linguistic ‘performance’, influ-
enced by different factors such as: topic, attitudes, competence and so forth.

In dealing with CS as a process, sociolinguistic studies have been conducted

from two levels: macro and micro levels. The macro level was adopted by Fishman
(1965) in his referential work ‘Domain Analysis’ where he focuses on “the correlation between
code choice and types of activity” (Boztepe, 2008:12). This differs considerably from Blom
and Gumperz (1972) micro approach which,
[…] regard the motivations for code switching not deriving from overall
societal norms but from the interlocuters themselves. With micro-level
approaches, code switching is explored at an interactional level” (Nguen,
2008:6)

This approach identified two types of code choice: situational switching and
metaphoricalswitching. Situational CS, as its name implies, depends on the situation, i.e. the
language used in formal situation is different from the one used in informal one. It is very clear
that, for many parts, the social context defines the linguistic choice, and such a choice is
controlled by social rules that have been become integrated part of the daily linguistic
behaviour of individuals as a result of experience. Metaphorical CS, on the other hand, occurs
according to changes in topic rather than the social situation. Here, it is “the choice of language
that determines the situation” (Hudson, 1996:53). Metaphorical switching is then topic-
related.Unlike the two preceding types, at which switching corresponds to a point
where the situation or topic changes, Conversational CS was added to CS
terminology to describe functions. This type of switching takes place in random
way and does not consider the context in which it may occur but rather the structure
of utterances.

Later on, Myers Scotton (1993b), as an effort to incorporate the macro and micro per-
spectives, introduced her “Markedness Model” since she employs “the conversation between
bilinguals as unit of analysis but also considers social norms and expectations as influencing
factors” (Nguen, 2008:7). This model provides the theoretical framework for examining the
bilinguals’ motivations for code switching. It is, therefore, regarded as a complementary device to
“account for CS by proposing that speakers have unmarked and marked choices
available to them when they speak” (Wardhaugh, 2006:109-110). These choices are consid-
ered by Scotton (1980:360) as “individually motivated negotiations” whose success only
depends on the degree of awareness and adequate use of“the communally recognized
norms” (1983a:123) which establish the meanings of the choices in different types of talk sit-
uations.

This Markedness modelparticipates in clarifying the socio-psychological motivations for

code switching. Under her Markedness model, Myers Scotton lists three maxims of code
choice: ‘the unmarked choices’ are expected and do not produce any special effect whereas
‘marked choices’ are unusual, un-expected and encode the speaker's social disapproval. The
third maxim is ‘the exploratory choice’ which is assigned to ‘explore’ or to ‘negotiate’ the
unmarked choice between interlocutors when the choice of code is not clearly apparent.

Structural Aspects of Code Switching

Studies on this aspect are looking for what constitutes CS. In other terms, this approach
attempts to answer questions beginning with the word ‘what’ taking CS as a product. It en-
deavours to explore the grammatical constraints which restrict switching between two codes
and describe the grammatical aspects of one's speech.
 In studying linguistic restriction on CS, some scholars have tried to present models or
approaches. The most common approaches are those of Poplack and her associatesand My-
ers Scotton Structural model as we shall see below.

1. Poplack’s ‘Linear Order Constraint’

Poplack’s study is an early but, indeed, an influential contribution to the linguistic aspects
of CS.Poplack (1980) investigates CS phenomenon among Spanish-English bilingual speakers
from Puerto Rico in New York. She tries to identify structural constraints where CS can occur in a
sentence, while noting, however, that “there is little doubt that functional factors are the
strongest constraints on the occurrence of ode switching” (ibid: 585). She proposes two major
constraints: ‘the free morpheme constraint’ and ‘the equivalence constraint’.

A. The Free Morpheme Constraint

The free morpheme constraint prohibits switching between a lexical item and
abound morpheme.In support of the free morpheme constraint, the following example can
be mentioned: / eat-iendo/. This word contains the English root ‘eat’ and the Spanish bound
morpheme ‘-iendo’, meaning (-ing). According to Poplack, this switching is not
permissible and cannot occur in bilingual speech because of the existence of a bound
morpheme.

B. The Equivalence Constraint

Switches, in the equivalence constraint, from one code to

another do not violate a syntactic rule of either language. It tends to occur at “points
around which the surface structure of the two languages map on to each other”(Poplack,
1980: 586). This constraint emphasizes that CS is almost likely to
occur where the two codes share the same word order. Accordingly, in the case of
Spanish/English CS, English attributive adjectives ordinarily precede the nouns whereas in
Spanish those adjectives follow it. This is why, switches may not occur between nouns and
adjectives in the noun phrase. If it is to happen, it will violate a syntactic rule of either language
as it can be appeared in the following examples: / the casa white / or / the blanca house /.
Both examples are, in fact, ruled out by the equivalence constraint since the former violate
English syntactic rules while the latter violate the Spanish ones. The sentences: / I like
youporqueeres simpatico / or / I told him that pa’que la trajeraligero /, for instance, which
mean ‘I like you because you are nice’ and ‘ I told him so that he would bring it fast’
respectively are, however, allowed as they obey the syntactic rules of both English and Spanish.

2. Myers’ Matrix Language Frame Model

In 1993, Carol Myers-Scotton developed a theoretical mode, perhaps the most detailed
one, non-linear, which is constructed on a more psycholinguistic speech production theory. She
called it the Matrix Language Frame Model (or MLF for short). It is currently one of the most
influential models “to account for the structures in intrasentential CS” (1993a:5). It
endeavours at extending the earlier constraints and identifying where the switch can take place
within a sentence.

Myers Scotton worked on a Swahili/English corpus in Nairobi, the capital of Kenya. She
takes her insights from Joshi’s (1985) asymmetry model. Such a feature is disregarded in
Poplack’s (1980, 1981) syntactic considerations of CS when he claimed that “speakers and
hearers generally agree on which language the mixed sentence is coming from” (1985:190).
Therefore, her MLF model is based on the notion that there is a hierarchical and asymmetrical
relation between language pairs in CS. The model, in fact, includes two major hierarchies. The
first one is the Matrix language ( ML) Vs the Embedded language( EL).

She describes the ML as “the language providing relatively more morphemes for the
relevant interaction type than the other language (s) used in the same conversation”
(1992:205). That is, ML is the dominant language and is“is responsible for constructing the
morpho-syntactic order of the CS sentences”. The EL, on the other hand, is “less active and
plays a restricted role in CS”.

What is striking is that researchers of the structural approach proposed many typolo-
gies and identified three main types of CS which are classified according to the switch points
and the way different codes are articulated. They are named: intra-sentential, inter-senten-
tial, and extra-sentential CS. Extra-sentential CS refers to the insertion of tags, exclamations,
and ready-made expressions such as: / I mean, you know/, from the donor language into the
recipient language whereas inter-sentential CS signifies the switch occurrence outside the
sentence and/or the clause level, ‘between sentences’ (Myers Scotton, 1995:4). Intra-senten-
tial switching, however, takes place within a sentence and/or clause or even word boundaries,
with no apparent change in topics, interlocutor, or setting as appeared in this switch: /rani tala3
la mairinxaajles papiers/: ‘I am going to the town hall to get some papers’. It may be a process
of inserting a noun, a verb, or even a clause, in a complex sentence (Poplack, 1980); it is often
referred to as ‘codemixes’ or ‘code mixing’.