Code-Switching in Spanish/English Bilingual Speech: The Case of ...

Retrievable at http://www.tc.columbia.edu/tesolalwebjournal
Code-Switching in Spanish/English Bilingual Speech: The Case of Two Recent
Immigrants of Mexican Descent
Antonieta Cal y Mayor Turnbull1
INTRODUCTION
This study presents the analysis of a home-based interaction between two
Spanish/English bilingual sisters. The purpose of the study is to investigate the functions code-
switching plays in this conversation and the role of these functions in the construction of their
identity as elite bilinguals. In addition, this study aims to contribute to the body of research that has
been conducted to identify the different functions of code-switching in bilingual interaction,
particularly with regards to the creation of an ethnic or linguistic identity.
Research on code-switching has been extensive. However, the majority of the studies have been
conducted on stable bilingual communities whose members are early bilinguals or sequential
bilinguals who have resided in the community for an extended period of time (Blom & Gumperz,
1972; Myers-Scotton, 1993b; Poplack 1980, 1981), or else they have been conducted in
educational settings where one or the two languages are used as main medium of instruction
(Cromdal & Aronsson, 2000; Rampton, 1999; Zentella, 1981). Furthermore, studies of code-
switching in Spanish/English bilinguals of Mexican origin have analyzed primarily the interaction
of second generation Mexican-Americans or that of first generation immigrants who become
circumstantial bilinguals once in the United States (García, 1981; Valdés, 1988; Valdés-Fallis,
1977). Little attention has been paid to another type of population: that of the recent, educated
1 Antonieta Cal y Mayor Turnbull completed her M.A. and Ed.M. in Applied Linguistics at Teachers College,
Columbia University. She is currently a full time lecturer in the Bachelor of Education in English Language
Teaching at the Universidad Autónoma de Chiapas in Southern Mexico. Her research interests include face-saving
and relations of power and dominance, developmental pragmatics and code-switching. She is also interested in
education and training of second and foreign language teachers, particularly teachers of English, French and
Spanish. Correspondence should be sent to: Antonieta Cal y Mayor Turnbull. Escuela de Lenguas C-I, Universidad
Autónoma de Chiapas. Boulevard Belisario Domínguez Km. 1081, Calzada a Rectoría s/n. Tuxtla Gutiérrez,
Chiapas. CP. 29050. E-mail: acymt@hotmail.com
1

---

Retrievable at http://www.tc.columbia.edu/tesolalwebjournal
immigrant who is already an elite bilingual. The current study thus attempts to fill this gap in the
literature by analyzing the functions of code-switching in the interaction of two recent
Spanish/English bilingual immigrants of Mexican descent.
REVIEW OF THE LITERATURE
Bilingualism and types of bilingualism
One of the greatest challenges when writing about bilingualism is finding a theoretical
framework that clearly defines what bilingualism is and who is bilingual. In fact, researchers and
theorists diverge on their definitions of who is purely monolingual and who is bilingual. The
definitions of bilingualism in the twentieth and twenty-first centuries have evolved from a purely
dichotomous approach (monolingual vs. bilingual) towards seeing bilingualism as a complex
phenomenon that includes key features.
May, Hill, and Tiakiwai (2005) present the historical evolution of the definition of bilingualism.
They begin with Bloomfield (1935) who defines bilingualism as having native-like control of two
languages. Then, they point out that later approaches, such as the ones posited by Haugen (1953)
and Macnamara (1966), claim that a bilingual is a speaker of one language who is also able to
communicate and understand another language, even if that capacity to communicate and
understand is minimal. Finally, they introduce Weinreich’s (1968) position, which considers
bilingualism as the ability to alternatively use two languages. The above definitions are
problematic because they tend to be minimalist or maximalist in terms of whom they see as
bilingual, and because they do not take into account the enormous variability across individuals.
In order to overcome the pitfalls of these early definitions, contemporary research has
approached bilingualism from both a linguistic and a social perspective. Romaine (1995) and
Valdés and Figueroa (1994) consider the degree of proficiency in each language as a key attribute
2

---

Retrievable at http://www.tc.columbia.edu/tesolalwebjournal
of bilingualism. They argue that bilingualism can be determined by how fluent the speaker is in
each language and to what degree the proficiency in both languages can be compared. That is, to
them, one can be a balanced bilingual with equal proficiency in both languages, or have a
dominant language and be bilingual nonetheless. May, Hill, and Tiakiwai (2005) consider age as
an important factor. They draw the distinction between a simultaneous bilingual, a child who
acquires two languages simultaneously as a result of family or societal bilingualism, and a
sequential or consecutive bilingual, a person who acquires a second language after the first
language, almost always in puberty or adulthood and often as the result of education. May, Hill,
and Tiakiwai. (2005) also maintain that the social status of the two languages and the personal
circumstances of the bilingual individuals will have an impact on how they see themselves vis-à-
vis the two languages, and whether they will consider themselves bilingual or not. Consequently,
they argue the importance of distinguishing between the elective or elite bilingual and the
circumstantial bilingual.
García (1997) and May (2002a, 2002b) point out that a person becomes an elite bilingual when
the acquisition of the second language (L2) is actively promoted and supported, and it is clearly
seen as a social and cultural asset. This individual usually becomes bilingual by means of the
acquisition of an elite language in an educational setting. In this case, the learners/users of the
second language (L2) are elective bilinguals, for they decide of their own free will to acquire the
second language at school or in a study abroad context. Their native language (L1) and culture are
not endangered by the acquisition of the L2. On the other hand, circumstantial bilinguals are those
individuals who have to learn/acquire an L2 in order to survive in a given society without having a
choice. This is the case of language minority speakers such as immigrants, refugees, and colonized
indigenous people who have had to learn the language of their new country or of their colonizers in
order to find a place in the mainstream society at the risk of losing their native language.
3

---

Retrievable at http://www.tc.columbia.edu/tesolalwebjournal
Defining code-switching
As with any aspect of the social sciences, particularly linguistics, research on code-switching
(CS) is full of terminological confusion. Researchers use the same terms in different ways, or give
different names to the same phenomena. In fact, there is some overlap and confusion on the uses of
the terms code-switching, code-mixing, code-alternation, and borrowing.
Milroy and Muysken (1995) define code-switching as “the alternative use by bilinguals of two
or more languages in the same conversation” (p. 7). They use CS as an overarching term under
which different forms of language change take place. They distinguish between intra-sentential CS,
which is used to refer to switches within the sentence, and inter-sentential CS, which is used for
switches between sentences. Beebe (1977, 1981) and Gardner-Chloros (1991) emphasize that CS
occurs not only between languages, but between dialects of the same language (e.g., French and
Alsatian). Gumperz (1982) states that CS refers to “the juxtaposition within the same speech
exchange of passages of speech belonging to two different grammatical systems or subsystems” (p.
59), thus making it clear that CS can occur on varieties of the same language.
Kachru (1983) and Singh (1985) use the term code-switching for inter-sentential switches and
use code-mixing for intra-sentential switches. They claim that this is a necessary distinction since
only code-mixing requires the integration of the syntactic rules of both languages, whereas plain
code-switching does not. However, as Boztepe (2003) argues, the difference between code-
switching and code-mixing is also explained by the classification of intra-sentential and inter-
sentential CS, making the use of either terminology a matter of personal preference, but one that
can still create unnecessary confusion.
The term code-alternation has been used to refer to instances of one language being replaced by
the other in the middle of a sentence, and it is usually related to longer stretches of CS. Code-
4

---

Retrievable at http://www.tc.columbia.edu/tesolalwebjournal
alternation thus differs from insertion, in that the latter phenomenon refers to occurrences of single
lexical items in the base or recipient language, that is, the main language in the code-switched
utterances, or to put it plainly, the language in which the greatest part of the utterance is being
delivered (Muysken, 2000).
Borrowing, on the other hand, has been defined as “the introduction of single words or short,
frozen, idiomatic phrases from one variety into another” (Gumperz, 1982, p. 66). Poplack,
Wheeler, and Westwood (1987) proposed a continuum for borrowing in CS utterances. At one end
of the continuum are the nonce borrowings, that is, the lexical items that might lack phonological
integration and refer only to content words. At the other end of the continuum are established loan
words, those that are fully integrated into the base language and that are also recurrent, accepted,
and widespread in the community.
Myers-Scotton (1992, 1993a) considers the distinction between code-switching and borrowing
unnecessary. However, she still distinguishes between cultural borrowing and core borrowing.
Myers-Scotton (1993a) argues that cultural borrowing refers to those lexical items that are new to
the recipient language culture (e.g., whisky), while core borrowings refer to those lexical items that
have equivalents or near equivalents in the recipient language, and therefore, do not really fulfill
any lexical gap in the recipient language (e.g., beer, cerveza, bière). Gumperz (1982) posits that
when the language of the core borrowed item has a higher symbolic value (cf. Bourdieu, 1991), the
social prestige of the donor language produces the non-integration of the borrowed word. In fact,
Myers-Scotton (1993a) suggests that educated bilingual speakers will try to pronounce borrowed
items as closely to the originals as possible in order to show they belong to the elite group that
knows the prestigious language. This is a common practice when using certain French expressions
such as ‘grande dame’ and ‘savoir vivre’ (Gumperz, 1982, p. 68).
5

---

Retrievable at http://www.tc.columbia.edu/tesolalwebjournal
Other researchers have followed Myer-Scotton’s argument that borrowing and CS are related
processes and part of a single phenomenon. Gysels (1992) and Backus (1996) stress that the line
between borrowing and CS is not easy to draw. They argue that a non-base language lexical item
can be either a switch or a borrowing, depending on the overall discourse structure or on the
individual speakers’ motivations.
Having analyzed the arguments and definitions provided above, it can be concluded that the
term code-switching encompasses both code-mixing and code-alternation (intra-sentential and
inter-sentential CS). It can also be concluded that a categorical distinction between CS and
borrowing is unnecessary since they are related processes. Therefore, in the present paper, the term
code-switching will be used to refer indistinctly to code-mixing, code-alternation, insertion, and
borrowing. The term code-switching will be used in its broadest sense, that is, to indicate “the use
of two or more linguistic varieties in the same conversation or interaction” (Scotton & Ury, 1977,
p. 5).
Functions of code-switching
In a seminal account of discourse strategies, Gumperz (1982) adopted the cover term
contextualization cues to refer to linguistic devices that index “what the activity is, how semantic
content is to be understood and how each sentence relates to what precedes or follows” (p. 131).
Gumperz (2001) also defines contextualization cues as “any verbal sign that, when processed in
co-ocurrence with symbolic grammatical and lexical signs, serves to construct the contextual
ground for situated interpretation and thereby affects how constituent messages are understood” (p.
221).
Gumperz (1997) identifies four different kinds of contextualization cues: (1) prosody, which
includes intonation, stress, accenting and pitch shifts; (2) paralinguistic signs, which comprises
6

---

Retrievable at http://www.tc.columbia.edu/tesolalwebjournal
tempo, pausing and hesitation, and conversational synchrony; (3) code choice, which consists of
code-switching, style switching, phonetic, phonological and morphosyntactic choices; and (4)
lexical forms and formulaic expressions, for instance, opening or closing routines.
Contextualization cues play an important role in the co-construction of conversation and
they are readily available to monolinguals, bilinguals, and multilinguals. Consequently, when
code-switching is used for contextualization purposes in bilingual or multilingual situations, it is
often accompanied by changes in intonation, pitch, and rhythm. In a study of the realization of
four discourse features (asides, reiterations, buts, and quotations) by six members of the
Punjabi/English bilingual community in London, Gardner-Chloros, Charles, and Cheshire (2000)
found that, when acting monolingual, participants used a series of linguistic devices such as stress
and intonation to accomplish the discourse features. On the other hand, when they acted bilingual,
they simply added code-switching to those linguistic devices, thus double-signaling the
conversational functions of the features. The researchers concluded that CS is simply an additional
linguistic device at the disposal of bilinguals, the effect of which “was almost always over and
above what could be achieved monolingually” (Gardner-Chloros, Charles, & Cheshire, 2000, p.
1335).
Gumperz (1982) identified six basic discourse functions that code-switching serves in
conversation. These are (a) quotations, (b) addressee specification, (c) interjections, (d) reiteration,
(e) message qualification, and (f) personalization versus objectivization. Quotations are simply
occurrences of CS where someone else’s utterance is reported as direct quotation or reported
speech. In addressee specification, the switch serves to direct the message to one particular person
among the several prospective addressees. Interjections simply serve to mark sentence fillers, such
as the Spanish expression ándale (i.e., swell) in an otherwise purely English utterance. Reiteration
occurs when the speaker repeats the message in the other code. It can be employed to clarify what
7

---

Retrievable at http://www.tc.columbia.edu/tesolalwebjournal
has been said or to increase the utterance’s perlocutionary effect. Message qualification is defined
as an elaboration of the preceding utterance in the other code. Finally, personalization versus
objectivization indicates the degree of speaker involvement in what is being said.
Saville-Troike (1982) identified eight different functions of CS, some of which overlap with
those proposed by Gumperz (1982). These functions are: (a) softening or strengthening of a
request or command (b) intensification/elimination of ambiguity (repetition), (c) humorous effect,
direct quotation and repetition, (d) ideological statement, (e) lexical need, (f) exclusion of other
people within hearing, (g) avoidance strategy, and (h) repair strategy. Gardner-Chloros (1991)
argues that functions of CS change according to the characteristics of the particular conversation.
This is why CS may occur as an effect of the topic or the roles of the participants.
Code-switching has many other functions that go beyond the discursive ones. Some of these
functions are signaling group identity and/or ethnic identity, and showing solidarity among
members of the same speech community or ethnic community (Crystal, 1987; Gal, 1978; Li, 1995;
Lo, 1999; Myers-Scotton, 1993b). The non-compliance to code-switch, that is, the linguistic
misalignments the speaker shows, can be seen as an attempt on the part of the speaker to distance
himself or herself from the other participants, be it because he or she does not want to be taken as a
member of that community, or because he or she does not want to accept the participant who is
making the switch as part of his or her community (Lo, 1999). Gumperz’s (1982) distinction of
we-code versus they-code is related to one of the basic functions just mentioned, namely solidarity.
While the we-code refers to in-group relations and a language or variety which is sometimes
undervalued (e.g., Chicano Spanish, AAVE), the they-code refers to the dominant language which
tends to serve as the means of communication for out-group relations with the mainstream society.
Goffman (1981) suggests that code-switching is the prototypical device used by bilinguals to
signal changes in frame and footing and to demonstrate whether they are aligned or not. Research
8

---

Retrievable at http://www.tc.columbia.edu/tesolalwebjournal
on CS has shown that this is the case. Analyzing standard and regional Italian code-switching
among adult speakers, Alfonzetti (1998) found that CS was used for story prefacing, frame
shifting, misalignments, topic changes, and setting quotations. Auer (1984) studied language
alternation in Italian-German peer talk and adult-child conversation and found that CS plays an
important role to determine whether the speaker is addressing one or many participants in a group,
thus signaling the listeners’ role as ratified or unratified participants. Based on Goffman’s (1974,
1979) work on footing, Cromdal and Aronsson (2000) analyzed the play interaction of
Swedish/English bilingual children from a production formats and participation framework
approach. They found that code-switching was used by the speakers to signal their roles as
animators (relayers), authors (creators) or principals (fully backing what is said). They also found
that code-switching was used by the speakers to signal the roles of the listeners as ratified or
unratified participants, thus producing instances of byplay, crossplay and sideplay among the
participants and bystanders.
Code-switching also functions as a device to establish an identity and negotiate relations of
power and dominance (Bolonyai 2005) and, in the case of developing bilinguals; it serves as a
device to fill in a lexical gap (Jisa, 2000). CS also serves as a device to accommodate the
linguistic needs of the listener. In interaction with a monolingual speaker, the bilingual speaker
will code-switch into the listener’s language, not only as a means to address him/her directly
(addressee specification) but as a means to make him or herself understood.
RESEARCH QUESTION
The research question addressed in this paper is the following:
What functions does code-switching perform in the informal interaction between two
Spanish/English bilinguals?
9

---

Retrievable at http://www.tc.columbia.edu/tesolalwebjournal
METHOD
Participants
The two participants in this study are sisters Martha and Sara Martínez. Martha and Sara were
born and raised in Mexico and have lived in the United States during different periods of their
lives. However, this is the first time they are living together in the United States. On previous
occasions, they had come as exchange students and lived with English monolingual American
families. During those occasions, they attended school alongside native English speakers in
mainstream classrooms and were never placed in an English as second language class.
Martha is a thirty-five year old physician who self identifies as a Spanish/English bilingual,
with Spanish as her dominant language, but English as her preferred language while in the United
States. She first came into contact with English at the age of ten. Martha lived in the United States
for two months at the age of ten, for one year at ages thirteen and eighteen, respectively. In
between these two years she studied English in Mexico. She has also traveled to the United States
for pleasure or work several times as an adult.
Sara is a thirty-one year old graduate student who self identifies as a Spanish/English bilingual
with Spanish as her dominant language. She stated no preference for either language, saying that
she likes to speak both and has no problem accommodating to whatever language her interlocutor
wants to use. Her first contact with English was at the age of twelve. Sara lived in the United
States for one year at ages twelve and eighteen, respectively. In between these two years she
studied English in Mexico. She has also traveled to the United States for pleasure several times as
an adult.
Martha and Sara lived together in their hometown until Martha finished High School and left
home to attend Medical School. They both claim they only used Spanish while living in their
10

---