Auer’s Conversation Analytic Framework
Auer (1995) presents a very different development of Gumperz’s interactional paradigm. Critical of Gumperz’s characterization of speakers’ language choices as realizations of a pre-established set of functions (e.g., quotation, interjection, reiteration), Auer argued that not only was such a list theoretically problematic, but it could also never be complete in principle. He suggested that the problems posed by a functional analysis of code-switching could be solved by adopting the framework of Conversation Analysis (CA) (see Atkinson and Heritage, 1984; Levinson, 1983). In Auer’s view, the CA approach has at least two advantages. First, it recognizes “the fact that whatever language a participant chooses for the organization of his or her turn . . . exerts an influence on subsequent language choices by the same or other speakers” (Auer, 1984: 5). Second, it “limits the external analysts’ interpretational leeway because it relates his or her interpretations back to the members’ mutual understanding of their utterances as manifest in their behavior” (Auer, 1984: 6).
Auer (1995) draws a useful distinction between “participant-related code-switching” and “discourse-related code-switching.” Participant-related code-switching is used to negotiate the proper language for the interaction— ideally, one that is both socially appropriate and accommodates the language competences and preferences of everyone in the conversation. Discourse- related code-switching, on the other hand, sets up a contrast that structures some part of the conversation (e.g., to mark out an off-topic sequence).
Below, I present a conversational sequence from my study of bilingual Korean-English-speaking children, where examples of both participant-related and discourse-related code-switching are illustrated (Shin, 2005). Note that this conversation takes place in a mainstream American first-grade classroom. Mrs. Kim, the teacher, immigrated to the U.S. at the age of seven with her family from Korea. Though she is bilingual in English and Korean, Mrs. Kim had specifically instructed her Korean students at the beginning of the school year not to speak to her in Korean out of consideration for the non-Korean students in her class. But she did not interfere with her students speaking Korean among themselves.
In Extract (5), from Shin (2005: 86-88), Mrs. Kim gave Jae and Abel the task of studying a snail. She asked them to measure the length of the snail, identify the different body parts, and determine how far it travels in a given amount of time. After Mrs. Kim walks away from them in line 21, Jae and Abel begin talking about cooking and eating snails. In line 33, Abel switches to Korean and says that one can eat the snail shell because it is hard, which Jae corrects by saying that one cannot eat it. Abel then says that one can die from swallowing a hard shell. In line 38, Jae explicitly tells Abel to speak English. This ends the use of Korean for the rest of the activity and the conversation continues solely in English. In Auer’s framework, the switch to English in lines 38 and 39 would constitute a participant-related code-switch since Jae’s preference for English was the motivation for the switch.
In contrast, Jae’s initial switch from English to Korean in line 24 is a discourse-related code-switch, which marks out the beginning of a side sequence where the boys talk about whether they could eat snail shells. Note that when Abel says that he eats snail shells, Jae’s disagreement is accompanied by a code-switch into Korean despite Jae’s usual preference for English. In line 27, Abel then switches into Korean and agrees with Jae. Li & Milroy (1995) report a similar pattern in their study, where Chinese-English bilinguals used code-switching to contextualize disagreement. In both Li & Milroy (1995) and Shin (2005), preferred responses (e.g., agreement) are accompanied by language alignment while dispreferred responses (e.g., disagreement) are accompanied by code-switching. In this example, Jae disagrees with Abel with a different language choice from the previous turn while Abel agrees with Jae in the same language Jae used in the previous turn.
Abel and Jae follow the movements of a snail assigned to them. They measure the length of the body, how long it travels, etc. Abel and Jae have been speaking exclusively in English for ten minutes. Mrs. Kim approaches their desk and checks on their progress. Korean is in italics.
shell TOP cannot eat-right
this-TOP discard-must (You should throw this out)
26 ike man mek-ko/
this only eat-and
and put in ice and we could eat it right (and if you put ice in it you could eat it, right?)
this too hard-because eat
Cannot eat (You can’t eat it.)
35 Abel: e ttakttakhay/
Um hard (Yeah, it’s hard.)
do-if die (You die if (unintelligible).)
By comparing the language choice in one utterance against the language choice in the previous turn, Auer’s (1995) sequential analysis shows ways in which bilinguals use code-switching to structure their conversation and to negotiate the language for the interaction. As the conversation proceeds, individuals carefully monitor other participants’ speech production and adapt their own language choice to their assessment of the bilingual abilities and preferences of the other participants. In sum, bilinguals use code-switching as an additional means to communicate their intentions and preferences to other participants in the conversation. This conclusion directly contradicts the view that code-switching is evidence of some sort of communicative deficit in bilinguals.
Next, we will see how bilinguals use code-switching to wield power over others.