Pre-imposition vs. in situ Negotiation of Group and Individual Identities: Spanish and English in US Service Encounters

Pre-imposition vs. In Situ Negotiation
of Group and Individual Identities:
Spanish and English in US Service Encounters

Abstract:
This paper examines the supposed opposition between essentialist or positivist approaches to identity—which categorize group and individual members by a priori properties of sex, race, ethnicity, and native speakerhood—and constructionism, which views such properties as relational and negotiable. Even when categories such as sex, race, ethnicity, and native speakerhood are considered to have been imposed a priori, there nonetheless persists a general recognition that these are at least to some extent social constructs—if not the categories themselves, the ideas we have about them. Using data from previous empirical work in Spanish and English code-choice in US service encounters, this paper argues that 1) the social constructs of ethnicity and native speakerhood can be either preimposed on a given interaction or formed in situ, that 2) that these two processes are not incompatible, and that 3) these practices perhaps even necessitate one another.

Two positions often arise in reviews and criticism of sociolinguistic research. These positions each go by various names, and are often presented grosso modo as opposing sides. Under the first position, groups and their members are categorized a priori according to properties such as ethnicity, race, and gender. This position is referred to as essentialist. This term for some time now has carried negative connotations, as the following example—one of many—will serve to illustrate. In the 2006 collection The Sociolinguistics of Identity, editors Omoniyi and White are at pains to distance the volume from any taint of essentialism, as statements from their introduction show. After outlining the major approaches espoused in the volume, the foremost of which is “that identity is not fixed”, the editors acknowledge that some of the contributors may not be in total agreement with all points in the editorial platform, but affirm that each nevertheless recognizes “that [these approaches] offer richer ways of researching between language and identity than earlier structuralist/essentialist approaches” (2). The editors praise a contributor who manages to describe “the influence of social structures on individuals, not in terms of essentialized social constructs such as ethnicity and gender but as participation in communities of practice” (3), and mention another contributor’s finding “that much of the existing literature on language loss and endangerment takes a rather essentialist view of identity and tends to assume that if a language dies, a particular identity disappears” (5-6). Finally, after wondering whether one of the papers might have broader applications, the editors caution that “we must ask if it is reasonable to refer to universal essentialist categories such as ‘immigrants,’ ‘refugees,’ and ‘borderlanders’” (7).

Under the second position—which is now most often put forward as the one researchers should follow, despite not-so-occasional criticisms that the methodology of such studies lacks rigor (1)—properties such as ethnicity, race, and gender cannot be imposed prior to an encounter, because they are considered to be contextual, negotiable, subject to change within interactions between individual speakers. This position has been referred to as social constructionist or constructivist. But even when categories such as ethnicity, race, and gender are considered to have been imposed a priori, in an essentialist sense, there is recognition that they are still at least to some extent social constructs. Warnke (2007) demonstrates convincingly that even what we understand as male and female sex, divisions for which there are some biological arguments, is highly dependent on social and situational interpretations. (2)

Scientific evidence has shown that there is little biological basis for the precise racial classifications that have been made over the past few centuries, and that such subdivisions reflect instead historical and social tides. Arguments for biological criteria continue to provoke debate. (3) What is certain is that criteria for various racial/ethnic classifications shift, and this shifting supports a constructionist view. In synchronic terms, however, categories of sex, gender, and race have enough stability to bear at least a modicum of psychological reality for individuals as individuals and, especially, for individuals as members of groups. In addition, racial/ethnic classifications are institutionalized in legal definitions, with very real consequences for the parties classified. Drawing on the work of Omi and Winant (1986) and Hacking (1999), Warnke describes the interplay of macro- and micro-levels of racial construction, which, I would argue, ultimately results in their essentialization. At the macro-level, “social, historical, and political practices, events, and actions” in combination contribute to the establishment of a nation’s “particular racial typology” (Warnke 2007, 62). Then, “[i]ndividuals take up their external racial identification as part of who they are” (64), and “this micro-level of racial identification means that individuals adopt and demand recognition for the identities they have internalized” (65).

Given the above, I am not arguing against constructionism, but rather asking when the construction is supposed to have taken place: pre-interaction, at a macro-level, or during each interaction, at a micro-level. Before I go any further, I must acknowledge what some readers may be thinking: that this is a non-issue. It might seem that what I represent as an either/or situation is actually more nuanced, and that in fact no one is asking researchers to choose between one or the other position. This is because only in the most relativistic interpretation of social constructionism is absolutely everything considered to emerge in situ. A version of constructionism influenced by realism is compatible with the assumption that individuals do bring histories and prejudices to their interactions, and hence would seem to acknowledge that some parts of their interlocutors’ identities are in fact pre-constructed rather than co-constructed. (4) For example, as I note in a review of Block (2009), there are “authors who have objected to an over-reliance on the construct of individual agency, and [Block] ultimately demonstrates that the progression from essentialist to poststructuralist and social constructivist views can be seen as a building onto rather than a full scale replacement of one school by another” (Callahan 2009b; see also Joseph 2004, 89-90).

These nuances often seem to be lost, however, if one is to judge from the emphasis on the construction and negotiation of identity that predominates in current work. Here, it is difficult to discern much acknowledgement of the possibility that someone might arrive at an encounter with a set of categories into which he or she is already prepared to place the other. Hacking (1999, 1) offers a sample listing of titles in which the phrase “social construction” appears; my own search of the Modern Language Association International Bibliography database yielded 543 titles in which the words “construction” and “identity” co-occur, and sixty-seven with “negotiation” and “identity” (May 31, 2012). Bucholtz and Hall (2005) “argue for a view of identity that is intersubjectively rather than individually produced and interactionally emergent rather than assigned in an a priori fashion” (587).For these authors identity does not precede discourse (607). Eelen (2001, 235) likewise maintains that “[s]ocial reality is the outcome of social practice” and that “[…] just as in Bourdieu’s notion of ‘ethnic identity’, ‘culture’ is also primarily a discursive notion, a tool for the (subjective) representation of reality rather than itself an objective reality. Culture should not be treated as a given entity, but rather as an argumentative practice” (238, emphasis in the original).

Despite these progressive academic analyses, however, it appears that essentialist tendencies do inform individual thought and behavior, if not without exception, a substantial amount of the time for a substantial number of people. (5) Bucholtz and Hall (2005) indirectly acknowledge the salience that a priori categories have for everyday people, in their observation that

it is perhaps easiest to recognize identity as emergent in cases where speakers’ language use does not conform with the social category to which they are normatively assigned. […] Such cases are striking only because they sever the ideologically expected mapping between language and biology and culture; that is, they subvert essentialist preconceptions of linguistic ownership. (588)

In order for said subversion to be noticed, or even to occur, essentialist preconceptions have to exist in the first place. An example can be seen in the remarks of an interviewee from my own research (to be discussed further below) who recounted his great surprise at finding Russian and Chinese service workers speaking fluent Spanish in New York City (Callahan 2009a). The fact that notions may be in place prior to an encounter does not preclude their reevaluation and revision. Witness the person who professes to begin an encounter as an essentialist only to become a constructionist during the exchange. (6) However, this can only happen if an encounter lasts long enough to give interlocutors a chance to build on what they bring to the situation. Certain types of encounters, for their brevity, are less conducive to this.

What follows is a brief overview of some of my own sociolinguistic research, some reactions to it, and a discussion of how both might illuminate the issue of pre-imposition vs. in situ construction of group and individual identities. My project on Spanish and English in US service encounters had three stages: (1) interviews with service workers on their use of these two languages with customers, (2) participant observation of service encounters in which fieldworkers entered business establishments in the role of customer and addressed service workers in Spanish, noting the language of response, and (3) participant observation of telephone service encounters, in which fieldworkers called some of these businesses and initiated conversations in Spanish. The interviews were conducted in New York City and the service encounters took place in both New York City and the San Francisco Bay Area. (7) In the second stage, participant observation of in-person service encounters—which, with more than 700 encounters, contributed the largest amount of data—a variationist analysis was performed on factors hypothesized to have the potential to influence workers’ language choice when addressed in Spanish by customers. The customer’s Latino or non-Latino ethnicity was found to be an important factor in whether or not the worker reciprocated the customer’s use of Spanish. The percentage of English language responses received to questions asked in Spanish—that is, non-reciprocal language choice—was more than twice as high for the non-Latino fieldworkers.

A reader of this research criticized my characterization of language preference as being “based on an external ideology that dictates which language is to be spoken in a given speech situation,” stating that this
points to a rather underdeveloped notion of ideology as somehow being ‘external’ to the interaction. A more nuanced and reflexive position (consistent with much research into language ideologies) might be that ideologies may be instantiated, reconstructed, and even contested in any given service encounter, such that ideologies are not external but central to and contextually shaped in talk.

I agree that language ideology may in some cases be created at the interactional level; this does not mean that external ideologies cease to exist or to exercise a powerful influence. In my interviews with workers, repeated allusion was made to workplace language ideologies. These ideologies situated English as the language of work, and dictated its use in the presence of customers. These ideologies at the same time called for accommodation to a customer’s non-English language choice, even in cases in which this was counterintuitive to the worker, for example when the customer’s choice did not match his or her physical appearance and, following from this, did not correspond to the worker’s stereotypical association of one or the other language with that particular customer. In one instance, a person described how she would bring her own ideology to an interaction, according to which she would accommodate to a Spanish speaker who were a customer at work, but not to strangers in other situations, such as stores in which she herself were the customer. She admitted to passing negative judgments on people whom she heard speaking a non-English language in public (Callahan 2009a, 14-15). From this I would conclude that an ideology can be individual or societal, but in either case it can be in existence prior to an interaction. And if it is in existence prior to an interaction, it is logical to assume that it will be instantiated during the interaction. Thus these positions are not mutually exclusive.

The reader of my work also criticized its ratification of  “the concept of ingroup/outgroup […] that assume[s] sharp distinctions, non-porous boundaries between supposedly separate and discrete social and linguistic groups, each with an essentialized and often racialized native language and native speaker identity.” (8) Nevertheless, I believe that the concept of ingroup/outgroup is an extremely useful one to explain what takes place when members of different groups encounter one another in public. Use of the social psychological model of ingroup and outgroup is, again, not mutually exclusive with an acknowledgment of the occasional porousness of boundaries.

In regard to the racialization of a language and its speakers, Barrett (2006), Hill (1998), Leeman (2005), Urciuoli (1996, 2011), and Zentella (1997, 2003) have documented the racialization of Spanish in the United States and of US Latinos. As Urciuoli (1996, 17) states, “racializing discourses speak of unindividuated populations” where the emphasis is on natural, ascribed attributes. The comments of my interviewees also reflect this racialization. Some of the respondents racialized, and some were racialized. At its most basic level, racialization manifested itself in their reports of judging others by physical appearance and in their accounts of being the subject of such judgments. For example, one young woman told how she was sometimes mistaken for African American (see Bailey 2000), and was assumed not to know Spanish by customers who would then use that language to speak about her in her presence. Other interviewees described the reverse, i.e. strangers assuming that they were unable to speak English. Most common, however, were reports from service workers that they were and would be guided by physical appearance to decide which customers to address in Spanish rather than English. In fact, fourteen out of twenty interviewees made explicit mention of appearance as one of their main criteria. A young man stated:

If it’s like, if it’s on the street like in the Hispanic community, I would most probably talk Spanish, depending on like the way they look, not to be kind of racial, but if they were dark-skinned in the Hispanic community, I most probably would talk to them in English, but if they were like Indian-skinned like me, you know, my color, I would talk Spanish. (unpublished data, Callahan) Socioeconomic factors were also mentioned. (9) For example, one interviewee stated that she would address most strangers in English, except for cab drivers who “look Hispanic.” Another interviewee cited her supervisor’s admonitions against addressing her in Spanish, which was both women’s native language. This took place in a clothing store located in an affluent Manhattan neighborhood, and the interviewee felt that the manager’s prohibition was based on a sense of elitism: “there was that feeling of being better […] because the clothes were expensive [imitates snobbish tone] and […] the ladies that come over there, they have money” (unpublished data, Callahan). I have written elsewhere about the influence an interlocutor’s appearance has on speakers’ language selection: (10)

Assumptions about linguistic competence—similar to assumptions about ethnicity—are often based on appearance. In cases where the addressee is known to have proficiency in more than one language, the speaker is likely to select whichever language is supposed to be the addressee’s ‘native’ one, or the one which is most associated with his or her ethnic group (which is not always synonymous, depending on one’s definition of native language —a point to which I will return).

Anecdotes about initiating an exchange in Spanish or in another language-other-than-English in the US and being answered to in English often involve speculation about the role of physical appearance. Latinos as well as non-Latinos can be of any race. Nevertheless, popular stereotypes persist, a fact which was borne out when participants were shown photographs of hypothetical customers (during the interviews conducted for this study), and when they described how they decided which language to use for addressing customers in first-time encounters (Callahan 2005). As mentioned above, several individuals cited physical appearance as their main criterion. Moreover, four native speakers of Spanish living in New York City reported to me encounters in which their use of Spanish had been received with surprise, and in which interlocutors (service workers) had continued to address them in English. These four individuals were from Spain, Venezuela, Uruguay, and the Dominican Republic. Each had light skin and facial features characteristic of a Western European phenotype (Callahan 2009a, 1-2). This gives us yet another glance at the racialization of Spanish in the U.S. that has been so well documented by Barrett, Hill, Leeman, Urciuoli, and Zentella, among others, and is an example of social construction at the macro-level. (11)

I mentioned that one’s native language and the language associated with one’s ethnic group are not always synonymous, which will not be news to anyone reading this essay. So far I have been using the terms native speaker and non-native speaker as based on language competence and order of acquisition, however problematic those criteria may be. However, many of the interviewees in my study, as well as some researchers themselves in other studies, use the terms native speaker and non-native speaker as more or less synonymous with ingroup and outgroup member. Ingroup member in this case means a person who acquires the label native speaker by way of inheritance, in the sense that Rampton (1995, 342) uses: “Inheritance occurs within social boundaries, while affiliation takes place across them.” Non-Latinos who learn Spanish have a connection to the language by affiliation—across social boundaries, whereas Latinos with various degrees of proficiency in Spanish—including those who have none—all have a connection by inheritance—inside social boundaries (Callahan 2004, 8; 2009a, 2-3.) Although Rampton’s is an excellent example of work that has been concerned with the emergence and negotiation of identities, his statements here support my point about the existence of preconstructed identities. If one can cross a boundary, there has to be one there in the first place. These boundaries are the product of social practice, which become societal structures, which means that they do not emerge anew in every encounter between individuals, although they may be enacted in each one.

A review of the literature on language choice in service encounters is outside the scope of this paper (see Callahan 2009a), but one case in particular bears mentioning here. Bernsten (1994), reporting on her experience as a Westerner speaking Shona in Harare, the capital city of Zimbabwe, states:

After a month of going to the same market, using Shona with the clerks, and being answered in English, I gave up. On the thirty-first day, I walked in and said ‘Good morning’. The clerk frowned and said, ‘But you are the lady who always speaks to us in Shona.’ And I said, ‘Yes, and you always answer me in English.’ And he said, ‘We do?’ Thus, I discovered another reason for the difficulties that learners experience in trying to speak Shona with bilinguals. The clerks in the store had not been consciously choosing English, but it had automatically been chosen as the appropriate language for a non-Shona conversation partner. (Bernsten 1994, 415-416)

As I have noted elsewhere, this automatic choice is consonant with Bourdieu’s view of habitus: ‘a set of dispositions which incline agents to act and react in certain ways’ (Thompson 1991, 12). Joseph (2004, 75) highlights the shift from identity production to identity reception, stating that “the identities others interpret onto us […] will be shaped by their own habitus” (Callahan 2009a, 8). These frameworks, though they do not totally exclude the possibility of an in situ, discursive construction, support a preconstructed one, imposed from without. Butler (1997), in her critique of Bourdieu’s “account of performative acts”, points to the existence of collective social forces, which, though they do not always involve official structures or institutions, constitute “prevailing forms of social power. […] For example, the racialization of the subject or its gendering or, indeed, its social abjection more generally is performatively induced from various and diffuse quarters that do not always operate as ‘official’ discourse’” (156-157). The fact that these determinative forces are present prior to the individual situations does not, of course, preclude challenges to situational norms, nor to their ultimate de-, re-, or co-construction.

To recapitulate, my work on service encounters was criticized as being based on essentialist notions of race, ethnicity and gender—which are variables in my study—and it was proposed that discursive, social constructionist theories of identity have more validity. I certainly do not disparage such theories. (12) But I am not convinced that essentialism and social constructionism are mutually exclusive. More to the point, I argue that if one’s interlocutor is guided by a so-called essentialist definition (though this be in itself socially constructed) of who or what one is, then that has a certain force and social consequences that have to be acknowledged. And I think we can distinguish between on the one hand believing in the inherent, absolute, existence of categories such as race, ethnicity, native language, and on the other hand in recognizing that others do in fact believe in their existence.

Edwards (2009), citing the work of Barth (1969), Tajfel (1978, 1982), and Turner and Giles (1981), notes that, “[ethnonational] borders can have a permanence that contrasts with the almost infinite mutability of the culture contained within them” (26). Laypersons accept a tendency toward essentialism as part of the human condition. According to Cronin (2000), “[t]he desire for simpler worlds is visibly articulated in national stereotype. At a conference in the Czech Republic attended by [travel writer Eva] Hoffman, the Irishman is predictably drunk, the Englishman ironic and the Frenchman obscure. Busy minds hunger for the economy of labels that make reflection on cultural difference a mindless subroutine” (88). On a slightly more academic note, Edwards (2009) maintains:

Our personal characteristics derive from our socialisation within the group (or rather, groups) to which we belong […] Thus, individual identities will be both components and reflections of particular social (or cultural) ones, and the latter will always be, to some extent at least, stereotypic in nature because of their necessary generality across the individual components. (20, emphasis in the original)

To conclude, I wonder how much joint construction of identity there can be in a brief one-time interaction. And I think this is an important question, because it is just such a type of interaction that can have very negative consequences if participants approach it with preconceived notions of the other. As valid as quibbles with the assumption of sharp distinctions, non-porous boundaries between supposedly separate and discrete social and linguistic groups may be, a simple glance at news headlines both past and present should suffice to convince us that such demurrals mean little outside of academia. As May (2012) cautions, “[…] we need to explain more adequately why ethnicity does seem to continue to mean something to so many people” (46). Edwards (2009) observes that “[f]or most societies throughout history, ethnocentrism, hostility and prejudice towards ‘out-groups’ have been the norm” (48). Hence, the rationale for focusing on the societal level is that

the most life-altering—to say nothing of the most egregious—consequences of the possession of one identity or another have almost always occurred at this level of generalisation. It may be that no strongly logical dividing line can be drawn between the individual and the social, but the most cursory attention reveals that the course of human history, and its implications for every individual, is by and large fuelled by perceptions of groups. (Edwards 2009, 22; emphasis in the original)

I submit that intergroup conflicts would be far less common if boundaries and identities were constructed anew in each interaction.

References

Bailey, Benjamin. 2000. “Language and Negotiation of Ethnic/Racial Identity among Dominican Americans.” Language in Society 29: 555-582.

Barrett, Rusty. 2006. “Language Ideology and Racial Inequality: Competing Functions of Spanish in an Anglo-Owned Mexican Restaurant.” Language in Society 35: 163-204.

Barth, Fredrik, ed. 1969. Ethnic Groups and Boundaries. Boston: Little, Brown.

Bernsten, Jan. 1994. “English and Shona in Zimbabwe.” World Englishes 13, no. 3: 411-418.

Block, David. 2009. Second Language Identities. London: Continuum.

Bourdieu, Pierre. 1991. Language and Symbolic Power. Cambridge: Polity Press.

Brubaker, Rogers, and Frederick Cooper. 2000. “Beyond ‘Identity’.” Theory and Society 29: 1-47.

Bucholtz, Mary, and Kira Hall. 2005. “Identity and Interaction: A Sociocultural Linguistic Approach.” Discourse Studies 7, nos. 4-5: 585-614.

Butler, Judith. 1997. Excitable Speech. New York: Routledge.

Callahan, Laura. 2004. “Native Speakers’ Attitudes Toward the Public Use of Spanish by Non-Native Speakers: From George W. to J. Lo.” Southwest Journal of Linguistics 23, no. 1: 1-28.

Callahan, Laura. 2005. “‘Talking Both Languages’: 20 Perspectives on the Use of Spanish and English Inside and Outside the Workplace.” Journal of Multilingual and Multicultural Development 26, no. 4: 275-295.

Callahan, Laura. 2009a. Spanish and English in U.S. Service Encounters. New York: Palgrave.

Callahan, Laura. 2009b. Review of Second Language Identities, by David Block. Linguist List Vol. 20, no. 2672.

Cronin, Michael. 2000. Across the Lines. Travel, Language and Translation. Cork, Ireland: Cork University Press.

Edwards, John. 2009. Language and Identity. Cambridge: Cambridge University Press.

Eelen, Gino. 2001. A Critique of Politeness Theories. Manchester: St. Jerome.

Genesee, Fred. 2005. “The Capacity of the Language Faculty: Contributions from Studies of Simultaneous Bilingual Acquisition.” In ISB4: Proceedings of the 4th International Symposium on Bilingualism, edited by James Cohen, Kara T. McAlister, Kellie Rolstad, and Jeff MacSwan, 890-901. Somerville, Massachusetts: Cascadilla.

Joseph, John E. 2004. Language and Identity: National, Ethnic, Religious. New York: Palgrave.

Hacking, Ian. 1999. The Social Construction of What? Cambridge, Massachusetts: Harvard University Press.

Hill, Jane. 1998. “Language, Race, and White Public Space.” American Anthropologist 100: 680-689.

Leeman, Jennifer. 2004. “Racializing Language: A History of Linguistic Ideologies in the U.S. Census.” Journal of Language and Politics 3, no. 3: 507-534.

May, Stephen. 2012. Language and Minority Rights. Ethnicity, Nationalism and the Politics of Language. 2nd ed. London: Routledge.

Milroy, Lesley. 2001. “The Social Categories of Race and Class: Language Ideology and Sociolinguistics.” In Sociolinguistics and Social Theory, edited by Nikolas Coupland, Srikant Sarangi, and Christopher Candlin, 235-260. London: Longman.

Morning, Ann. 2009. “Toward a Sociology of Racial Conceptualization for the 21st Century.” Social Forces 87, no. 3: 1167-1192.

Omi, Michael, and Howard Winant. 1986. Racial Formation in the United States from the 1960s to the 1980s. New York: Routledge.

Omoniyi, Tope, and Goodith White. 2006. Introduction to The Sociolinguistics of Identity, edited by Tope Omoniyi and Goodith White, 1-10. London: Continuum.

Rampton, Ben. 1995. Crossing: Language and ethnicity among adolescents. London: Longman.

Schiffman, Harold F. 2002. “Re: [code-switching] re: ‘Biracial’.” The Code-switching Forum, February 15 (3:57 pm). Accessed June 7, 2012. Link

Social Science Research Council. 2005. Is Race Real? Accessed June 9, 2012. Link

Tajfel, Henri, ed. 1978. Differentiation Between Social Groups. London: Academic Press.

Tajfel, Henri, ed. 1982. Social Identity and Intergroup Relations. Cambridge: Cambridge University Press.

Thompson, John B. 1991. Introduction to Language and Symbolic Power, by Pierre Bourdieu, edited by John B.

Thompson, translated by Gino Raymond and Matthew Adamson, 1-31. Cambridge, Massachusetts: Harvard University Press.

Turner, John C., and Howard Giles, eds. 1981. Intergroup Behaviour. Oxford: Basil Blackwell.

Urciuoli, Bonnie. 1996. Exposing Prejudice: Puerto Rican Experiences of Language, Race, and Class. Oxford: Westview Press.

Urciuoli, Bonnie. 2011. “Discussion Essay: Semiotic Properties of Racializing Discourses.” Journal of Linguistic Anthropology 21, no. s1: E113-E122.

Villa, Daniel. 2002. Online posting. “Re: [code-switching] re: ‘Biracial’.” The Code-switching Forum, February 15 (5:34 pm). Accessed June 13, 2012. Link

Warnke, Georgia. 2007. After Identity: Rethinking Race, Sex, and Gender. Cambridge: Cambridge University Press.

Zentella, Ana Celia. 1997. Growing Up Bilingual. Oxford: Blackwell.