Talking Spanish English

‘Talking Both Languages’: 20 Perspectives on the Use of Spanish and English Inside and Outside the Workplace Laura Callahan Department of Foreign Languages and Literatures, The City College of New York, New York, NY, USA Strategies for deciding which language to use when forms part of the sociolinguistic competence of speakers who can speak more than one language. Language choices are shaped by a number of factors, including linguistic proficiency of both speaker and interlocutor, the ingroup or outgroup status of each, and the setting in which an exchange takes place. Language choices in the workplace are further influenced by company policies and by the asymmetrical power dynamic in worker customer interactions. This paper reports on interviews with 20 Latino workers who each use Spanish and English in a different workplace in New York City. Participants were asked about their use of the two languages with customers, coworkers and supervisors in the workplace, and with friends, family members and strangers outside the workplace. The major themes that emerged from the interviews are presented, including some commonalities found with other studies of language pairs in the commercial setting.




Keywords: ingroup versus outgroup, language choice, language ideology, sociolinguistic competence, Spanish and English in the USA, workplace

Introduction This paper reports on the reflections of 20 individuals with regard to their use of Spanish and English in the workplace and elsewhere. The research was undertaken as part of a larger project entitled Language Choice in Interethnic Communication: Spanish and English in Urban Service Encounters. Data collection for the parent project includes anonymous encounters between service workers and researchers posing as customers, and passive or participant observation in selected workplaces. It will thus produce quantitative data based on observed behaviour. The present study offers triangulation of those results with qualitative data based on reported behaviour. The purpose of the interviews was to discover what factors dictate language choice in the workplace, and what differences there are between choosing Spanish or English at work as opposed to outside of work. Sociolinguistic competence Having communicative, or sociolinguistic, competence involves having knowledge of the social rules governing the appropriate use of a language in a variety of contexts (Hymes, 1972). A speaker with such competence controls 0143-4632/05/04 275-21 $20.00/0 J. OF MULTILINGUAL AND MULTICULTURAL DEVELOPMENT

275

– 2005 L. Callahan Vol. 26, No. 4, 2005

276

Journal of Multilingual and Multicultural Development

not only basic grammatical structures, but also the address system, locutions for greeting and leave-taking and other formulas, both verbal and nonverbal. He or she must have an awareness of which topics may be discussed when, by whom and with whom. These variables are influenced by several factors, such as the speech situation and the age, gender and socioeconomic status of both speaker and addressee (e.g. Tsitsipis, 1989). A speaker with a reasonable degree of communicative competence also knows which register or degree of formality to use for most situations. Having the ability to speak more than one language adds an important dimension to communicative competence. A speaker who knows more than one language has to decide which language to use when. Such speakers meet this challenge in different ways, but, as we will see, there are some common rules and strategies. This paper focuses on ingroup members: all 20 of the interviewees were members of a Latino ethnic group.1 An outgroup member using a community language may encounter a different reception, and must therefore follow different rules than a speaker who is judged to be a member of the ingroup (Bernsten, 1994; Callahan, 2004; SanAntonio, 1988: 36; Weyers, 1999; Woolard, 1989). Outgroup members have to be able to determine when reciprocal rather than anticipatory codeswitching is more appropriate. In other words, when not to initiate conversation in a minority language, but to wait for ingroup members to do so. Some of the rules that are in force in other situations may be suspended for outgroup members engaged in a service encounter, at which time their categorisation as outsiders may be overridden by their customer status.

Method Semi-structured interviews were conducted in March and April of 2004. There were eight men and twelve women, ranging in age from 18 to 48, with a median age of 21.2 Sixteen of the interviews were conducted in English and four in Spanish.3 This was a convenience rather than random sample; participants were recruited with fliers and by word of mouth at a four-year commuter college in New York City. Prior to the on-campus recruitment, unsuccessful attempts were made to recruit workers directly from commercial establishments. Hence, all of the interviewees were students who also held or had recently held positions off-campus. Although all of the interviewees were students, the college from which students were recruited has a very workingclass student population. Various types of workplaces were represented, from beauty salon to supermarket to car rental agency to law office. At each informant’s worksite, Spanish and English were used on a regular basis with customers and coworkers. The structure common to service encounters across business types is considered to be sufficiently uniform, notwithstanding minor variations between one type of store and another. Hence, the heterogeneity of the workplace type is not considered to be an obstacle to the generalisations that will be made about worker
customer and worker
worker language selection. Interviewees were asked closed and open-ended questions to elicit information about their use of Spanish and English. In addition, they were shown photographs of eight people, each one of whom represented an age:

‘Talking Both Languages’

277

under 30 and over 30, a gender, and an ethnic group: Latino or non-Latino. The interviewees were asked to comment on what their language choices would be if they were to participate in an exchange with each one of the eight individuals pictured.

Results In this section the major themes that emerged from the interviews are presented. The terms ‘interviewee’, ‘informant’ and ‘worker’ are used interchangeably, and Spanish and English are referred to as ‘opposite languages’. Workers follow different language choice rules in the workplace The workplace represents a setting distinct from many others, and workers follow a different set of parameters when deciding which language to use there. A worker may have to modify his or her personal language choices
whether these be dictated by ideologies or intuition
to comply with workplace policies or to please the customer. For example, a worker who might use knowledge of the community and of the physical appearance and speech characteristics to identify Spanish speakers may suspend these techniques at work in favour of automatic English use, at least at the initiation of an encounter. Answering in the opposite language is a dispreferred response in the workplace. Workers recognise the negative affective consequences of not following a customer’s language choice once the latter makes this known.4 Asked to comment on a situation in which the customer addressed the worker first, 18 of the 20 interviewees stated that they would answer in Spanish if addressed in Spanish, regardless of the customer’s ethnicity. One interviewee, when asked to confirm that she would answer in Spanish if addressed in Spanish by hypothetical customers of various ethnic appearances, seemed incredulous at the question. ‘Of course!’ she said. ‘You know, they’re customers, so you’ve got to be nice to the customer’ (103). Another interviewee echoed this, stating that he would respond in Spanish to any customer who addressed him in that language ‘because they’re the customer’ (006). This parameter imposes itself in exchanges between strangers; in conversations with co-workers, friends and family members less significance is attached to answering in English when addressed in Spanish, or vice versa: (1) Researcher:

And how about the reverse, have you ever found yourself answering in English when someone speaks to you in Spanish, someone addresses you in Spanish. Informant: To my mom. Researcher: Any situations with strangers that you’ve done that? Informant: Strangers? Researcher: Or with, you know, with non-family members in other situations? Informant: That they speak in Spanish and I answer them in English? Researcher: Right. Informant: No, not with strangers. (104)

Journal of Multilingual and Multicultural Development

278

All participants either implicitly or explicitly acknowledged the assumption that accommodation to an interlocutor’s language choice is the preferred response. More language accommodation takes place in the workplace than outside the workplace. In (2), for example, a worker who reported switching to Spanish as needed for her own customers’ comprehension recounted an incident in which she herself was the customer and she declined to accommodate to the dominant language of her interlocutors, even when doing so would have facilitated communication: (2) Researcher:

Do you, um, if somebody speaks to you in English that is hard to understand, do you ever switch to Spanish? Informant: Work situations, definitely. But when I’m outside, and I don’t know why I do this, I, I guess um, I don’t know, in a sense I’m resentful because of the fact that, they, I don’t know, I judge them. I’m, I’m, it’s horrible. But it doesn’t, I, I don’t speak to them in Spanish right away. Researcher: So you’re talking about in what type of situation? Informant: In a situation where I’m on the street or I’m in a grocery store. This happened the other day. I’m in a grocery store and I’m asking for cayenne pepper. They have no clue what cayenne pepper is. I was like, OK, where are your condiments? They have no clue what it is that I mean by condiments. So, I’m frustrated because they don’t understand, and I should speak Spanish right away, but I’m like [sighs], and I get really mad and I just walk away, and just look for the condiments myself. I tend to do that in the street. I get annoyed that, I don’t know, I mean, I know that it’s wrong because I shouldn’t judge a person, you know perhaps they’ve been here one year or two, but I just, I’m just to the belief that you should try to move outside of your neighborhood. (104)

Workers are sensitive to and will accommodate the language proficiency or language preferences of interlocutors in the workplace more than they may be willing to do in other situations. Seventeen (85%) of the interviewees stated that they would switch to Spanish with a customer or coworker who spoke to them in nonfluent English. Language proficiency and preference do not always coincide, and workers are aware of this. Several mentioned that they would ask first whether the person preferred to speak Spanish, and that they would continue the conversation in English if their interlocutor wished to practice English. In (3), an informant described the language display technique he used to offer assistance to second language English speakers in a non-facethreatening manner: (3) . . . if I see them [having] difficulty, I don’t know if they’re trying to use the language, because they studied the language, they came to the US, they want to really practice it, or they don’t know I know Spanish. So I jump in with a Spanish word so they could hear the Spanish word and they can understand it. So I use it every time I see the person having difficulty with

‘Talking Both Languages’

279

the language. Because, like I said before, people are very sensitive with that, because they study it and then they want to use it. (001) When approached by persons speaking nonfluent Spanish, some workers reported that they do not switch to English, because they assign a voluntary motive to the person’s use of Spanish, and assume that the person wishes to practice using the language. Note this differential treatment in (4): Si alguien te habla en ingle´s, y es difı´cil de entender, su ingle´s, ¿cambias tu´ al espan˜ol para contestarle? Informant: Sı´, por supuesto, sı´. Si yo veo que esa persona luce como hispana, entonces sı´, inmediatamente. Researcher: Y al reve´s, si alguien te habla en un espan˜ol que es difı´cil de entender, ¿cambiarı´as tu´ al ingle´s? Informant: No, realmente, cuando las personas hablan espan˜ol, es porque quieren practicar su, tal vez esta´n estudiando espan˜ol y quieren practicarlo, entonces por lo general tengo mucha paciencia con ellos.. . . Si le quiero decir algo que quiero que entiendan bien, se lo digo en ingle´s y entonces otra vez al espan˜ol otra vez. (101)

(4) Researcher:

Researcher: If someone speaks to you in English, and it’s hard to understand, their English, do you switch to Spanish to answer them? Informant: Yes, of course, yes. If I see that that person looks Hispanic, then yes, immediately. Researcher: And the other way around, if someone speaks to you in Spanish that’s hard to understand, would you change to English? Informant: No, really, when people speak Spanish, it’s because they want to practice their, maybe they’re studying Spanish and they want to practice it, so I generally have a lot of patience with them.. . . If I want to tell them something that I want them to understand well, I tell them in English and then [switch] back to Spanish again. (my translation) In contrast, a person speaking English as a second language, perhaps nonfluently, is more likely to be doing so because of its status as the default language for public encounters. Nevertheless, other workers stated that they would automatically switch to English if addressed in nonfluent Spanish. Fewer of the workers had had this experience; being addressed in nonfluent English was more common. When workers will address a customer in Spanish Ethnicity of interlocutor, as determined by physical appearance, has been observed to determine which language is selected by adults and by children as young as two (e.g. Genessee, 2003; Schiffman, 2002; Villa, 2002; Zentella, 1997). In many cases language choice is a form of accommodation, based on what knowledge the speaker has about the linguistic competence of the person he or

Journal of Multilingual and Multicultural Development

280

she is talking to. 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 one is supposed to be the addressee’s ‘native’ language, or the one which is most associated with the addressee’s ethnic group. The corpus was almost evenly split between workers who stated that their default language to initiate any encounter was English, and those who said they would address someone in Spanish if the person appeared to be Hispanic, as in (5) and (6): (5) Researcher: . . . you might address them in Spanish? Informant: Yeah, sometimes, yeah, like if they look, like really Hispanic or something, and I know they would speak Spanish. (106) (6) Researcher:

Informant:

When you, when you decide what language, if you ever have to talk to somebody before they’ve spoken to you yet, you’re the one who’s initiating it, how do you decide . . . By the way they look. If they look, I don’t know how, but if they look like, more Hispanic, I will, like I, I don’t know, I guess the way they look, if they, I don’t know, I kind of know, like, who will speak Spanish. (111)

Tokens (7) and (8) show that workers also take cues from customers’ conversation among themselves to decide which language to use: (7) Also, yeah, if I, also if I hear them speak to someone they’re with speaking Spanish then I will approach them in Spanish. (111) (8) If I hear that they’re talking in a certain language, then I’ll talk to them in that language. (003) These workers’ ingroup status makes such an approach fairly unproblematic. Nevertheless, as discussed in the next section, workers’ language choices occasionally encounter obstacles. When workers make a mistake in predicting a customer’s language preference This is most relevant when the worker decides to initiate an exchange in Spanish, as English-initiated exchanges that meet with incomprehension are easily repaired. All workers stated that they would simply switch to Spanish if a customer indicated incomprehension or responded in Spanish to the worker’s English. The negative consequences are greater when a worker chooses Spanish to address a customer who does not speak Spanish
or does not wish to speak it
because the selection of Spanish entails a greater social risk than the selection of the default language, English. The worker’s selection of Spanish carries several possible implications. One is that the worker considers him or herself to share membership in a group with the customer. Other implications are that the customer prefers to be attended to in Spanish, is fluent in Spanish, and perhaps does not speak English. The worker’s selection

‘Talking Both Languages’

281

of English carries fewer implications, because it is the default language which can be used even in predominantly non-English-speaking communities in the USA as a starting point in a commercial encounter (Weyers, 1999). When the worker’s prediction of the customer’s language preference is accurate, the rewards for speaking Spanish can be considerable: improved comprehension, a sense of solidarity and overall increased customer satisfaction. But the costs, in the case of an inaccurate prediction, can also be higher. The use of Spanish may carry the implication, noted above, that one’s listener cannot speak English. This was expressed by an interviewee who described her experiences being addressed in Spanish (see also example 12, below): (9) Informant: Well, when I was young, people noticed that I was Spanish, you know, and they would automatically talk to me in English, I mean in Spanish. Yeah. They would, yeah, like they’d look at me and then automatically talk in Spanish, they’d automatically assume . . . Researcher: When you were a child? Informant: No, even in high school. Yeah, they would look at me and say that I’m Spanish, you know, so they’d start speaking to me in Spanish you know, thinking that I don’t know any English. Researcher: So what situations, like what kind of people, friends, people in, like store clerks? Informant: Like staff from school, you know, people who work in the school, um jobs, past jobs, friends, you know, people that I know from other, through other people, might speak to me in Spanish automatically or just people in the street who want to ask me something, you know. Researcher: And so how would you react? Informant: I would react, well, automatically, I would think, ‘Oh, he thinks I don’t know English’. But yes, I would just reply back in Spanish. Or sometimes in English just to show them that I know English. (108) An incorrect assumption that a customer is capable of speaking Spanish can cause embarrassment, threatening the face of both the worker and the customer. Both parties apologise: the worker for placing the customer in a position in which he or she must admit to the lack of a skill, and the customer for forcing the worker to change languages: (10) . . . they don’t speak Spanish so then they’re like, ‘Ah’, they’ll just be like really, ‘Oh, I don’t speak Spanish’, and then I’ll be like, ‘Oh, I’m sorry’ . . . I started out in Spanish but then they’re like, ‘Oh, no, I don’t speak Spanish, sorry.’ (005) The three workers who mentioned this situation said it arose when they categorised a customer as Hispanic based on physical appearance. Two would take the customer’s profession of a lack of Spanish-speaking ability at face value, one ascribing it to the person’s having been raised in the USA, and accepting that, in the words of one worker, ‘there’s a lot of Hispanic people

282

Journal of Multilingual and Multicultural Development

that they, even though they look Hispanic, they don’t speak Spanish’ (005). But the third worker expressed disbelief in one customer’s profession of ignorance: (11) . . . una vez le hable´ a una sen˜ora en espan˜ol, y ella no hablaba espan˜ol . . . ella dijo que no hablaba espan˜ol, y yo me quede´ sorprendida. Ella parecı´a que hablara espan˜ol. Y ella [dijo]
No, yo no hablo espan˜ol. Pero yo me quede´ pensando que ella hablaba espan˜ol y tal vez no le gustaba hablarlo . . . Porque ella . . . parecı´a dominicana. (1010) . . . one time I spoke to a lady in Spanish, and she didn’t speak Spanish . . . she said that she didn’t speak Spanish, and I was surprised. She looked like she would speak Spanish. And she [said], ‘No, I don’t speak Spanish.’ But I got the feeling that she did speak Spanish and maybe she didn’t like to speak it. Because she . . . looked Dominican. (my translation) If this worker’s intuition were correct, it could mean that her customer wished to distance herself from shared membership in a group with the worker.

When workers answer in the opposite language Lo (1999: 472) shows how English is used to rebuff outgroup use of an ingroup language. Hers was a study of two young Asian American men in Los Angeles, one of whom was a Chinese American learner of Korean, and the other was Korean American. Pedraza, in his 1987 study of the Puerto Rican community of East Harlem, noted that in age asymmetrical intraethnic interactions, the younger person would respond in the language in which he or she was addressed. In peer exchanges, adolescents ‘would often answer each other in English even if addressed in Spanish, assuming that the interlocutor was part of the group, or at least familiar enough so that this would not be taken as an insult’ (Pedraza, 1987: 38).5 In other words, to use English after being addressed in Spanish could be considered offensive, unless the degree of intimacy between addressor and addressee allowed for such liberties. Workers are much more apt to report answering in the language opposite from the one in which they were addressed when the addressee is someone known to them, such as a co-worker, friend, or relative. Seventeen (85%) of the workers reported differential behaviour in this regard when the addressor was a customer. In other words, while they acknowledged switching languages between turns with co-workers, friends and family members, they stated that they would always follow a customer’s language choice, responding in whatever language the customer used first. The three workers who reported that they had or would answer a customer in the opposite language were all referring to an exchange in which a non-Latino customer were to address them in Spanish. Each alluded to the customer’s proficiency in Spanish as the factor that would determine whether or not they would respond in the same language. Two stated that they would continue the conversation in English if the person’s Spanish were not fluent, whereas one worker said she would do this only to

‘Talking Both Languages’

283

give a brief display of her English-speaking ability for the customer, before returning to Spanish: (12) If it’s like good, good Spanish, you know, if they’re fluent, fluent in Spanish, otherwise if I see that they’re probably stereotyping me, then, you know, to speak to me in Spanish, but they don’t really speak [Spanish], I’ll continue in English so I could, so they can know that I speak English, yeah. Like if they speak broken up Spanish just to try to help me, then I’ll answer them in English, [and] whichever language they feel comfortable [in], I’ll continue. (111) Nonreciprocal language choice was reported to take place most often in nonwork settings. In some cases, informants would answer a friend or family member in Spanish after being addressed in English, even if their interlocutor did not understand Spanish. However, there were more reports of answering in Spanish when addressed in English by a Spanish speaker. Referring to fellow members of an organisation for young Dominicans, an interviewee reports: (13) . . . muchas veces ellos hablan ingle´s, porque les es ma´s fa´cil expresarse en ingle´s. Y yo contesto en espan˜ol porque yo se´ que ellos entienden. (109) . . . a lot of times they speak English, because it’s easier for them to express themselves in English. And I answer in Spanish because I know that they understand. (my translation) Workers mentioned two main causes for nonreciprocal response: the ability to access vocabulary faster in one language, or what they termed confusion. The vocabulary motive is illustrated in (14) and (15), and the confusion as a cause is illustrated in (16). Specific concepts were mentioned to illustrate the first type of scenario: (14) Researcher: Informant:

Do you ever answer in Spanish when somebody speaks to you in English? Math. Like if it’s a, like a quantity, I still, I can’t do math in English. (104)

Again, a non-work situation and familiarity with interlocutors makes a language change more likely to occur. Both factors are noted by the informant in (15): (15) Researcher:

. . . if somebody speaks to you in English, do you ever answer in Spanish? Informant: No. Researcher: Not necessarily with customers, but with co-workers? Informant: Yeah, I’ll speak Spanglish, yeah definitely, everybody that knows both languages and they’re amongst themselves, sometime you forget a word in one language. You go into the Spanish or vice versa. Researcher: How about the reverse, if somebody, do you ever answer in English . . .

Journal of Multilingual and Multicultural Development

284

Informant:

Usually, that happens more. Usually my parents talk to me; oh, you want to know about at work? (001)

The instances that workers ascribed to confusion generally included an immediately previous conversation in which they were speaking the opposite language. For example: (16) Oh, that happens to me, when, for example, after a meeting or when I’m doing a lot of things, or when I, when I’ve been speaking English all day, and then somebody speaks Spanish to me, then I answer in English because I’ve, I’ve been speaking English or listening to English and doing everything in the English language, and so . . . when a person asks me something in Spanish then I will answer in English. It’s happened to me many times. Or when I pick up the phone and somebody asks me something in Spanish, then I answer in English. (102) English as a neutral language As noted above, English can be used as a starting point when the speaker is either uncertain of the addressee’s language preference, or is prevented from using it by other factors. In the USA and in many other parts of the world, English has become a neutral language, a code that can be used to avoid whatever implications may result from choosing a more identity-laden language (Bernsten, 1994). Eight of the interviewees reported that they would use English to initiate any first-time encounter with customers, and a ninth said he would do so unless he heard the customer speaking Spanish to companions.6 English as the language of work Informants reported switching from Spanish to English within the workplace in response to a topic or task change. As illustrated in (17)
(20), a worker might joke and discuss personal matters with co-workers in Spanish, but switch to English to discuss business, participate in a meeting or handle a cooperative task: (17) Informant: I don’t know, like, for me English is more professional, it’s like, yeah, like this gets the job done, this language. I use my Spanish like, for my down time, to relax, and have fun, and goof around, that’s how I feel. So it depends on what I have to do. If we have a job, we have a task, we’re talking English. Researcher: Even with people who speak Spanish. Informant: Right. But if we have some down time, and we relax, there’s no customers, and we’re just fixing the store, we start joking around, we talk Spanish. More relaxed. (18) Informant: The attorney, he’s my supervisor, trial counsel, and he speaks exclusively to me in Spanish and I answer him in English. Researcher: Do you know why you answer him in English?

‘Talking Both Languages’

285

No. It just happens. It just naturally happens. Now when we’re talking, if it’s business. If we’re talking on a personal basis, you know, something personal about myself, or even gossiping, then we both speak Spanish. Researcher: Oh, OK. So maybe topic. Informant: Yes. Depending, I guess, on the topic and maybe the seriousness of the matter, but when I was gossiping with him recently, it was Spanish. All the way. All the way. But then when the conversation shifted onto something, it went right back. Him Spanish and me English. (112) Informant:

(19) . . . with my supervisors, I speak more English. Most of the time, like, more than with my co-workers... when I speak to my co-workers it’s mostly things that I did or we did and, you know. And when I speak to my supervisor it’s business or something so I will have to, I speak to them in English. I don’t know why. (111) (20) I speak sometimes Spanish, sometime English, but when it’s a meeting, when I have a meeting, or when I pick up the phone or when somebody else is coming, I speak English. Or if there is somebody else besides my boss and the super . . . the coordinator. Because they, they both speak Spanish. Because we are, three of us who speak Spanish. But if there is somebody else, we, we speak English, or if there is a meeting or, you know, when I pick up the phone, it’s English. (102) While all of the workers interviewed were expected to attend to customers in Spanish whenever the need arose, half had been told not to speak Spanish with coworkers in the presence of non-Spanish-speaking coworkers or in the presence of customers. These were not official company policies, but were made clear to the workers nonetheless7: (21) Researcher: Informant:

Researcher: Informant:

Researcher: Informant:

Researcher: Informant: Researcher:

At this job, or at another job in your life, can you remember anyone ever telling you to speak only English? Yes. This job does it a lot. [Company name] does that a lot. And I used to work at my other jobs, yeah, and [company name] does that a lot. A lot of places do it. So, is, is it like a policy, a supervisor says that? Yeah, supervisors. Like, I read the rule books, they make you read the rule books before you start working at these places. And nowhere in the rulebook do they say: ‘You must speak English’. It’s not written down. Yeah. Like at this job, due to the fact that we’re international, we got all international people. They prefer you talk in two languages, but they don’t like you talking, when it’s, when you have to be professional about it. So they only want you to use it . . . When it’s necessary. When it’s necessary. So there’s sort of, it sounds like there’s kind of an unwritten policy.

Journal of Multilingual and Multicultural Development

286

Informant: Yeah. Researcher: At this job and at other jobs you’ve had, too. Informant: Yeah. Stick to the English. (001) Some of the workers expressed their attitudes toward these restrictions, but for the most part they were reported without evaluation: (22) No, yeah it was the manager. It was on Upper West Side, in a store, . . . and she speaks Spanish, that’s her first language. And she was just like, no, we can’t speak Spanish here in the store, even if you want to talk to me in Spanish. So sometimes, you know, it happened that I would speak Spanish to her, so when I talked to her she would answer me in English, then she’d say, remember we aren’t speaking, we don’t speak Spanish here. And I was just like, OK. But, you know I, I felt uncomfortable because she speaks Spanish. She’s Spanish. (102) (23) Well, at my job, actually, I basically got hired because I speak Spanish, but when I’m with, like an employee break room, or at a meeting with other people that don’t speak Spanish, I’m not allowed to speak Spanish. Like if I’m with a friend and some other person, whoever it is, another employee, that doesn’t speak Spanish, I have that conflict, I can’t speak Spanish to my friend, because that other person is there, so I’m restricted. But if it’s to the customers or job-related, I can. (107) (24) Researcher: Informant:

Have you ever had the experience of, on a job, being told not to speak Spanish, to speak only English? Well, actually, yeah, I was, for example, because my other co-workers, they speak Spanish also, and sometimes, you know, I’ll be actually trying to talk Spanish with them. We’re from the same country. But the uh, employer, he actually tells us that when somebody who looks white, or you should be trying to have your conversation in English, that way they won’t feel weird about it, or that way, the customer doesn’t think that you’re talking about them, or something like that. So actually he said, ‘Oh, when somebody comes like that, you should actually talk English, that way they can actually know what you’re talking about. And they won’t feel like, OK, like, they’re talking bad about you.’ (006)

The interviewees who had never been subject to language restrictions worked for the most part in predominately Spanish-speaking environments. The language workers prefer to speak in the workplace Workplace language preferences seemed to be for the most part driven by relative proficiency and utility: the worker expressed a preference for the language in which he or she felt most comfortable, or the one used by most of the other workers.8 But there were exceptions. Two interviewees saw work as an opportunity to practice their English; one of these two workers also saw his

‘Talking Both Languages’

287

current job, at which there were many Spanish-speaking customers, as an opportunity to maintain his Spanish. Language ideologies were also expressed. One informant stated that she used English as the base language for attending to all customers, changing to Spanish only when necessary for comprehension. Whereas such a practice was mandated at some of the interviewees’ workplaces, in this case it was the informant’s personal choice: (25) . . . if they don’t understand something, then I’ll ask them if language is a problem, and I’ll explain to them in Spanish. But I choose to speak English because I think it’s very important that people from Latin America or anyone, just anyone that knows another language, that if they’re going to be living here that they should know English. (104) Twelve (60%) of the interviewees had been hired in part for their ability to speak Spanish, but none had ever received a salary differential. Although their ability did not translate into higher wages, some of the workers did express a feeling of satisfaction at being able to help customers and enhance the quality of the service encounter via the use of Spanish. (26) Well, I have a lot of positive things when I speak Spanish because I work in retail. So in retail, I work in Times Square, and there’s a lot of tourists. So my good experience is getting to know people from different places around the world. And because I can communicate with them in Spanish, they come back to me and they really appreciate me, because they don’t find anybody that can talk Spanish sometimes, so they feel uncomfortable. So they come back to me, and it’s a good experience, I really find, like they enjoy it when they find out I speak Spanish. (001) (27) I help them out because I speak Spanish. Like I get to, to help more customers, to help translate if anybody needs anything, or, you know, it’s more useful . . . Because I speak Spanish it’s more, I’m more able to do it. (107)

Spanish as the language of intergenerational communication For outside the workplace, workers reported the highest use of Spanish with their mothers. In most cases the mother’s lack of proficiency in English was cited, but one informant, a person who refused to speak Spanish with nonfamily members unless it were unavoidable, also refused to speak English with her mother, who was fluent in both languages. Fourteen (70%) of the workers reported speaking only or mostly Spanish with their parents, and either 50
50 Spanish and English or more English than Spanish with people of their own generation, whether these were siblings, friends or coworkers. Workers reported being more apt to address younger customers in English, regardless of ethnic appearance. When shown the photographs of the eight hypothetical customers, workers who said that they would address an older Latino man or woman in Spanish said that they would address all of the younger people in English. The excerpts in (28) and (29) illustrate this association between youth and the use of English:

288

Journal of Multilingual and Multicultural Development

(28) Most of the time it’s more English than Spanish, it’s more like youthoriented . . . Probably because they’re younger, you’ll probably catch what they’re saying more easily. Probably they’ll speak English and Spanish. (002) (29) . . . it depends on the situation. For example, if I see a young person, Hispanic, for example, my age, I usually talk to them in English because sometimes I feel more comfortable, and they could feel more comfortable, and we could communicate better. But if it’s like, for example, an old, an older person, like usually most of the time we speak, we talk in English, I mean, sorry, in Spanish. (006)

Race, ethnicity and language A California high school student of Vietnamese ethnicity who learned Spanish reports, ‘. . . people are like, ‘‘Why is this Asian girl speaking Spanish?’’’ (San Jose Mercury News , 2004). As we have seen, race and ethnicity, as judged by physical appearance, is often used by workers to determine the linguistic proficiency of an individual. When they were shown the photographs of hypothetical customers, those interviewees who did not report using English as the categorical language in which to initiate all encounters with customers stated that they would address in Spanish the people who, in the words of several informants, ‘looked Hispanic’, and in English those who did not. A person’s use of a language not popularly associated with their ethnicity
even when the language is their first language
continues to be cause for comment, as illustrated in (30): (30) Nunca me imagine´ . . . Rusos hablando un espan˜ol bien fluido, siendo rusos . . . . . . fui a un Dunkin’ Donut, y pedı´ en ingle´s una, algo para desayunar, y la mujer me contesto´ perfectamente en espan˜ol, y yo quede´ tan sorprendido que no podı´a creerlo, */Ud. habla tan buen espan˜ol; */ ¡Es que yo soy peruana! Ella no me dijo que */Yo soy china. Me dijo */Yo soy peruana. (011) I never imagined . . . Russians speaking very fluent Spanish, being Russian . . . . . . I went to a Dunkin’ Donut, and I ordered in English a, something for breakfast, and the woman answered me in perfect Spanish, and I was so surprised that I couldn’t believe it. ‘You speak Spanish so well’; ‘I’m Peruvian!’ She didn’t say, ‘I’m Chinese’. She said, ‘I’m Peruvian.’ (my translation)9 Just as it does outside of work, race plays a role in everyday interactions in the workplace. In one case, an interviewee felt that his coworker was the target of deprecatory treatment due to the colour of his skin. In other words, his race motivated his interlocutors’ lack of sympathy with his low proficiency in Spanish:

‘Talking Both Languages’

(31)

289

. . . I’m going to tell you something about Spanish people from Spain. They’re, I don’t know, I feel like they’re a little racist, should I say. Because every time they come into the store, they come in and they’re talking Spanish, and remember that guy I told you, half Black, half Dominican? He really tries with them, to talk Spanish with them. But they just make fun of him. I feel so bad, I feel so, because the guy’s really trying to learn the language. But they just make fun of him. I’ll be like, and then I step into it, step to it, and I’ll be like, I come in and I’m like, ‘Hey, why you treating him so bad?’ in Spanish; I’m like ‘¿Por que´ lo esta´ tratando tan mal? No lo trate mal. Esta´ tratando bien duro con usted.’ And they feel bad. So I put them in their place. Researcher: So do you get like, um, that doesn’t happen with him and people from other Spanish-speaking places? Informant: It, basically Caribbean Spanish [people], no, South American people, no.. . . I guess it’s because, it’s his skin tone. That’s what I think. (001)

In another case race was a factor in customers’ assessment of a worker as incapable of understanding Spanish: (32) Informant: Well, sometimes, people don’t know, they, sometimes they judge you by how you look. Sometimes they think I’m Black and not Hispanic. And they say something in Spanish, thinking that I don’t know what they’re saying. Researcher: So do you do anything when that happens? Informant: Yeah. I speak Spanish. ‘You think I don’t know Spanish? I’m Dominican’, I tell them. ‘Oh, you don’t look it.’ ‘Yeah, I speak Spanish.’ (1010)

Nonprestige varieties of Spanish A few of the workers expressed their own or others’ beliefs about which dialects of Spanish are less desirable. Spanish with single word borrowings from English, or with frequent intrasentential codeswitches, was the variety most often noted in pejorative terms, as in (33)
(35): (33) . . . lo que no me gusta es que me vengan a hablar en espan˜ol y en ingle´s . . . Yo me tiro para uno de los dos . . ..Por ejemplo, si algu´n hispano me quiere hablar con algunas palabras metidas de ingle´s, hablando espan˜ol, entonces me voy al ingle´s solo . . . (011) . . . what I don’t like is for them to come speak to me in Spanish and in English . . . I’ll go with one of the two. For example, if a Hispanic person tries to talk to me with some English words thrown in, while speaking Spanish, then I’ll switch to English alone . . . (my translation) (34) We didn’t substitute words because we didn’t know the word for something. We didn’t interject Spanish for the word; if we didn’t know

290

Journal of Multilingual and Multicultural Development

what dog was in English, we didn’t say perro [‘dog’], we just, we knew what both of them were. It just happened naturally. Now, you see people speaking what they think is bilingual, but they can’t speak either one. And that’s a problem, and I find that very insulting. Sometimes I tell them, let’s just stick to one. You know, one language. (112) Referring to a Spanish professor, one interviewee noted that this person’s negative view of using English and Spanish together was not restricted to the language classroom, and therefore did not stem solely from pedagogical concerns: (35) . . . [a] ella no le gustaba que mezcla´ramos los dos idiomas. Ella dice que es una falta de respeto para el espan˜ol, si mezclamos palabras en ingle´s. Ella no esta´ de acuerdo con el Spanglish, porque no esta´ bien para ella . . . en su vida personal, no le gustaba. (109) . . . she didn’t like for us to mix the two language. She says that it’s disrespectful of Spanish, if we mix in words in English. She doesn’t approve of Spanglish, because it’s not good in her view . . . in her personal life, she doesn’t like it. (my translation) In (36), in addition to some morphological features of a contact dialect, a stigmatised pronunciation characteristic of Caribbean Spanish was mentioned. (36) Informant: . . . I’m proud to speak Spanish, and I am proud to be able to read and write it. What makes me upset is the Latin Americans who, or whomever, using the Spanish language, well, I would say native Latin Americans, or those who have grown up here with Latin American parents, who don’t speak the language properly, who massacre the language. I cannot stand that. Researcher: And, and you, what you would consider to be not speaking it properly, can you give any examples? Informant: Um, just not making a proper sentence or just not, um, like saying: ‘Mira para alla´’ or ‘Subiendo para arriba ’ which means ‘going up’; it’s like a way of saying you’re going up the stairs up, of course you’re going up the stairs up. You know, things like that. Researcher: So, redundancies? Informant: Yeah, redundancies. Or just the way they massacre the language not pronouncing it correctly: ‘Arroz’ [with uvular R]. It’s not ‘arroz’ [with uvular R], it’s ‘arroz’ [with trill]. Um, so I can’t stand the way they mas-, I would understand more if you’re learning English and you have a problem pronunciating [sic] the English language. But when you’re Spanish, and you have problems pronunciating [sic] it, just don’t speak it at all, just don’t massacre the language. (104) The requirement of a higher standard of proficiency
however defined by the informant
for speaking Spanish, is a theme that has emerged in earlier investigations (Callahan, 2004). There is overall a much greater tolerance for

‘Talking Both Languages’

291

imperfect English. Nevertheless, workers do show some tolerance for speakers of nonfluent Spanish, as illustrated by some of the informants’ willingness to maintain a conversation with second language learners.

Discussion This is a qualitative study, using a nonrandom sample, which is in any case too small to permit statistically reliable generalisations to a larger population. Nevertheless, some tendencies do emerge. In addition, there is a high degree of congruence between the behaviour reported here and the behaviour observed in the course of the anonymous encounters carried out in the parent project, Language Choice in Interethnic Communication: Spanish and English in Urban Service Encounters. To wit, in 91% (n /146) of the 160 total encounters conducted by the researcher, informants answered at least partially in Spanish when addressed by an outgroup member in Spanish.10 Anecdotal evidence suggests that, in the absence of the desire to please the customer, there might be more nonreciprocal language choice, and less accommodation to the outgroup member trying to initiate an exchange in Spanish. On two occasions, both in a neighbourhood in which Spanish is heard more than English in public, the researcher was answered in English after she had spoken in Spanish. In both cases, the person she had addressed in Spanish had been speaking Spanish to another person on the scene. Workers who have linguistic competence in Spanish and English demonstrate their sociolinguistic competence inside and outside the workplace by using a variety of strategies. Their knowledge of the community and of workplace norms affords them a high degree of accuracy in matching speakers’ characteristics with the appropriate language choice. Similar patterns have been documented elsewhere, between other language pairs. For example, Gardner-Chloros (1997: 364) found that: . . . an understanding of the situational norms is relevant. These norms are, on the one hand, common to many different social groups, such as the rule that ‘the customer is always right’, which makes the customer’s variety dominate in customer-salesperson interactions (Genesee & Bourhis, 1982), and, on the other hand, they are enmeshed with the particular diglossic configuration in question: thus in Strasbourg an Alsatian-speaking customer might well refrain from imposing her language because there is a conflicting norm tending to prefer French as the language of public conversations with interlocutors one does not know personally. (emphasis in the original) Many of the workers in the present study gave preference to English as a public language, in as much as the workplace and many of the situations found therein are defined as public. These included working in the presence of customers, meeting with supervisors and performing work-related tasks with coworkers. Nonpublic situations within the workplace were limited to conversations with Spanish-speaking coworkers when no non Spanishspeaking person was present. The majority of the informants would accommodate the customer’s preference; the customer’s preference thus had

292

Journal of Multilingual and Multicultural Development

the power to make Spanish the public language in the workplace. Outside of work, the interlocutor’s language continued to exercise influence, but less rigidly so than in a service encounter. In Gardner-Chloros’ study, which was quantitative rather than qualitative, it was correctly predicted that ‘less Alsatian and more French would be spoken between customers and salespersons than within groups of customers or within groups of salespersons: the independent variable was therefore the ingroup/outgroup distinction’ (Gardner-Chloros, 1997: 365; emphasis in the original). This also finds a parallel in my data, although not exact. It would be most accurate to say that more monolingual English and more monolingual Spanish is spoken between customers and workers in my study, and more Spanish and English together is spoken by the workers in ingroup situations. Workers base their language choices on both linguistic competence and episode-external ideological factors (Torras & Gafaranga, 2002). Linguistic competence has to do with what language(s) the speakers in an exchange have proficiency in, and episode-external ideological factors have to do with questions of allegiance to and ownership of a language. As discussed above, English in the USA is available for use by all parties in most public situations. Spanish may be used by ingroup members between themselves and to address others they categorise as Spanish speakers. Outgroup members may attempt to use Spanish in commercial encounters, and their status as customers assures them a higher rate of success than they may enjoy in other situations. Linguistic competence prevails over ideological concerns when communication would otherwise be impossible, for example, when a customer is monolingual in Spanish or English. When competence is not an overriding factor, when both parties are capable of speaking both languages, a combination of situational norms and each interlocutor’s ideologies determine language choice. Acknowledgements I am grateful to the interview participants for sharing their experiences and insights with me, and to the two anonymous reviewers for their comments on an earlier version of this paper. This project was partially funded by the Research Institute for the Study of Language in an Urban Society, at the Graduate Center of the City University of New York. Correspondence Any correspondence should be directed to Professor Laura Callahan, Department of Foreign Languages and Literatures, The City College of New York, Convent Ave. at 138th St, New York, NY 10031-9198, USA (Lcallahan@ ccny.cuny.edu). Notes 1. Thirteen of the participants were of Dominican descent, three Puerto Rican, three Colombian and one Venezuelan. In this paper the terms ‘ingroup’ and ‘outgroup’ refer to Latino and non-Latino ethnicity, respectively. 2. See Appendix 1 for demographic information on the interviewees. 3. The language of the interview was left up to the informants; however, the researcher’s outgroup status and the semi-formal situation of the interview may

‘Talking Both Languages’

4. 5. 6.

7. 8.

9. 10.

293

have influenced their choices. For the same reasons, although several informants mentioned codeswitching, none did so during the interview. For seminal work on speech accommodation in monolingual exchanges, see Giles and Powesland (1975) and Giles and Smith (1979); for accommodation in service encounters, see Labov (1966). For another seminal work on language choice between Spanish and English in New York City, see Poplack (1998). Another worker used Spanish in the same fashion; in her workplace Spanish was the default language. This notwithstanding, non-Spanish-speaking customers were frequent in this informant’s workplace and she knew how to identify them. She addressed them in Spanish also, to avoid being accused of judging people by their appearance. For an overview of language policies in the workplace, see Aguirre, 2003; Fishman, 1988; Imahara, 1993; Macı´as, 1997; Teboul, 2002; Ugalde, 1990; Valde´s, 1997; Wyld, 1997. The worker’s language preference was not discussed in all of the interviews. In the ones in which it was, workers indicated the following preferences (E /English; S/Spanish; B /both): (001) B; (002) B; (003) B; (004) S; (005) E; (006) B; (007) E; (101) S; (102) E; (103) E; (104) E; (105) B; (106) E; (107) S; (108) S; (1010) S; (111) E. Note that in this exchange, the informant, whose dominant language is Spanish, was judged by the worker to be a Spanish speaker based on his accented English. Seventy per cent (n/ 112) responded in monolingual Spanish, and 21% (n/ 34) with a Spanish/English codeswitched utterance.

References Aguirre, A. Jr. (2003) Linguistic diversity in the workforce: Understanding social relations in the workplace. Sociological Focus 36 (1), 65
80. Bernsten, J. (1994) English and Shona in Zimbabwe. World Englishes 13 (3), 411
418. Callahan, L. (2004) Native speakers’ attitudes toward the public use of Spanish by nonnative speakers: From George W. to J. Lo. Southwest Journal of Linguistics 23 (1), 1
28. Fishman, J.A. (1988) ‘English only’: Its ghosts, myths, and dangers. International Journal of the Sociology of Language 74, 125
140. Gardner-Chloros, P. (1997) Code-switching: Language selection in three Strasbourg department stores. In N. Coupland and A. Jaworski (eds) Sociolinguistics. A Reader (pp. 361
375). New York: St. Martin’s Press. Genesee, F. (2003) Bilingual acquisition: Exploring the limits of the language faculty. Keynote speech. 4th International Symposium on Bilingualism, Arizona State University, 30 April. Genesee, F. and Bourhis, R.Y. (1982) The social psychological significance of codeswitching in cross-cultural communication. Journal of Language and Social Psychology 1 (1), 1
27. Giles, H. and Powesland, P.F. (1975) Speech Style and Social Evaluation . London: Academic Press. Giles, H. and Smith, P.M. (1979) Accommodation theory: Optimal levels of convergence. In H. Giles and R.N. St. Clair (eds) Language and Social Psychology (pp. 45
65). Baltimore: University Park Press. Hymes, D. (1972) Models of the interaction of language and social life. In J.J. Gumperz and D. Hymes (eds, 1986) Directions in Sociolinguistics: The Ethnography of Communication (pp. 35
71). Oxford: Basil Blackwell. Imahara, K.K. (1993) Language rights issues to the year 2020 and beyond: Language rights policy. State of Asian Pacific America: Policy issues to the year 2020 (pp. 233
251). [CD-ROM]. Available: Current Issues SourceFile. Record: A152
2. Labov, W. (1966) The Social Stratification of English in New York City. Washington DC: Center for Applied Linguistics. Lo, A. (1999) Codeswitching, speech community membership, and the construction of ethnic identity. Journal of Sociolinguistics 4, 461
479.

294

Journal of Multilingual and Multicultural Development

Macı´as, R.F. (1997) Bilingual workers and language use rules in the workplace: A case study of a nondiscriminatory language policy. International Journal of the Sociology of Language 127, 53
70. Pedraza, P. (1987) An ethnographic analysis of language use in the Puerto Rican community of East Harlem. Centro de Estudios Puertorriquen˜os Working Papers . New York: Centro de Estudios Puertorriquen˜os. Poplack, S. (1998) Contrasting patterns of code-switching in two communities. In P. Trudgill and J. Cheshire (eds) The Sociolinguistics Reader, Vol. 1: Multilingualism and Variation (pp. 44
65). London: Arnold. SanAntonio, P.M. (1988) Social mobility and language use in an American company in Japan. In W.B. Gudykunst (ed.) Language and Ethnic Identity (pp. 35
44). Clevedon: Multilingual Matters. San Jose Mercury News (2004) Mastering multiple languages earns graduates’ school honors. 6 June, B1
B2. Schiffman, H.F. (2002) Online posting. Re: [code-switching] re: ’Biracial’. The Codeswitching Forum . On WWW at http://www.egroups.com/message/code-switching/ 734. Accessed 15.2.2002. Teboul, J.C.B. (2002) Case study: The language dilemma. Management Communication Quarterly 15 (4), 603
608. Torras, M.C. and Gafaranga, J. (2002) Social identities and language alternation in nonformal institutional bilingual talk: Trilingual service encounters in Barcelona. Language in Society 31, 527
548. Tsitsipis, L.D. (1989) Skewed performance and full performance in language obsolescence: The case of an Albanian variety. In N.C. Dorian (ed.) Investigating Obsolescence: Studies in Language Contraction and Death (pp. 117
137). Cambridge: University Press. Ugalde, A.M. (1990) ‘No Se Habla Espanol’: English-only rules in the workplace. University of Miami Law Review 44 (5), 1209
1241. Valde´s, G. (1997) Bilinguals and bilingualism: Language policy in an anti-immigrant age. International Journal of the Sociology of Language 127, 25
52. Villa, D. (2002) Online posting. Re: [code-switching] re: ’Biracial’. The Code-switching Forum . On WWW at http://www.egroups.com/message/code-switching/736. Accessed 15.2. 2002. Weyers, J.R. (1999) Spanish as an ethnic marker in El Paso, Texas. Southwest Journal of Linguistics 18 (1), 103
116. Woolard, K.A. (1989) Double Talk: Bilingualism and the Politics of Ethnicity in Catalonia. Stanford, CA: Stanford University Press. Wyld, D.C. (1997) Managing in a Nuevo Mundo: Accommodating linguistic and cultural diversity under Title VII. Employee Responsibilities and Rights Journal 10 (1), 1
19. Zentella, A.C. (1997) Growing Up Bilingual . Oxford: Blackwell.

‘Talking Both Languages’

295

Appendix: Demographic Characteristics of the Interviewees Code

Sex

Age

Selfreported dominant language

Place of birth

Parents’ place of birth

Age of arrival in USA

Years in USA

001

M

24

English

USA

DR

0

24 (life)

002

M

20

English

USA

DR

0

20 (life)

003

M

18

Both

DR

DR

7

11

004

M

18

Spanish

DR

DR

14

4

005

M

20

Both

USA

DR

0

20 (life)

006

M

20

Both

DR

DR

6

14

007

M

18

Both

USA

DR

0

18 (life)

011

M

48

Spanish

Col

Col

36

12

101

F

22

Spanish

Ven

Ven

18

4

102

F

23

Spanish

Col

Col

18

5

103

F

29

Both

PR

PR

18

11

104

F

29

Both

USA*

Col

10

20

105

F

19

Both

DR

DR

5

14

106

F

21

Both

USA

DR

0

21 (life)

107

F

18

Both

USA**

DR

10

108

F

20

Both

USA

DR

0

109

F

22

Spanish

DR

DR

15

7

1010

F

19

Spanish

DR

DR

8

11

111

F

32

English

USA

PR

0

32 (life)

112

F

44

Spanish

USA

PR

0

44 (life)

8 20 (life)

*Moved to Colombia at age 4. Returned to USA at age 10. Finished high school in Colombia. **Moved to DR at age 1.