ETHNICITY AND THE STUDY OF IMMIGRANT COMMUNITIES

The investigation of the construction and negotiation of ethnicity from a sociolinguistic perspective is particularly relevant in the light of the heated debate that this concept has generated and still generates among social scientists. Scholars in sociology, psychology and anthropology, among other disciplines, have struggled for years and still fundamentally disagree not only on the criteria that may be invoked in the definition of ethnicity, but even on the possibility of an objective definition. In a survey on the concept of ethnic identity in communication research, Leets, Giles & Clément 1996 underline that there is no comprehensive and unified theory of ethnicity in the social sciences, and that different ways of operationalizing this construct have produced divergent empirical results in applied research. Definitions of ethnicity traditionally invoke psychological, cultural, economic, and biologic criteria, or a mix of them. For example, Edwards 1985 stresses the centrality of the psychological dimensions because he regards ethnicity as defined by an individual's loyalty to a group that has an observable common past. Buriel & Cardoza 1993 also equate ethnicity with psychological affiliation as they argue that, regardless of variations in the biological, cultural, and social domains, if a person identifies with a particular ethnic group, then she or he will be willing to be perceived and treated as a member of that group. Farley 1988 gives preeminence to social and cultural factors such as nationality, language, and religion as defining factors in ethnicity. Levine 1997 invokes situational variables such as migration conditions and patterns. The different emphases in these definitions reflect a fundamental divide between “primordialists” and “circumstancialists” (Glazer & Moynihan 1975, Scott 1990). The former conceive of ethnicity as created around primordial and affective symbols and stemming from objective biological attributes (van den Berghe 1987). The latter stress the influence of different types of circumstances in the emergence of specific ethnicities.

The divide between competing theories cannot be resolved if ethnicity is seen, as in many of the definitions reviewed above, as a choice that can be attributed to individuals, rather than as a process enacted, produced, and negotiated in specific social contexts. Although it may be true that ethnicity is often related to language, race, or cultural tradition, the specific ways in which these symbolic connections are made radically change from one context to the other. In addition, the circumstances that draw groups and individuals toward or away from ethnic affiliations can be understood only based on in-depth analyses of when and how ethnic loyalties are drawn in concrete social arenas. Here, I argue for a vision of ethnicity as a social and interactional construct influenced by macro and local social circumstances and negotiated in concrete social contexts. As theorized by Barth 1969, ethnicity is used to define boundaries within and between groups, and therefore any aspect of social reality, from food to accent, can be used to symbolically index ethnic affiliation. Thus, there are no unified criteria that can universally define ethnic boundaries; rather, these are creatively invoked and negotiated by individuals and groups in response to their evolving social roles and circumstances.

If ethnicity is seen as a process, then the close analysis of the negotiation of ethnicity within and between different groups becomes central to our understanding of its role in the construction of identities. Sociolinguistic studies of the management of ethnic categories in discourse and interaction (Bucholtz 1999; De Fina 2000, 2006; Bailey 2001; Maryns & Blommaert 2001; Rampton 1995) have shown that ethnic loyalties are not given but negotiated, that they are indexed in subtle ways rather than openly declared, and that they often contradict expectations and stereotypes about received ethnic boundaries. In the social constructionist perspective that these studies support, ethnicity should not be regarded as an abstract attribute of the individual, but rather as an interactional achievement grounded in concrete social contexts and evolving with them.

ITALIAN AMERICANS IN THE ETHNICITY DEBATE

Most sociological and psychological research on immigrants takes for granted their homogeneity under ethnic labels, and describes their characteristics and attitudes solely based on quantitative surveys. The tendency to discuss ethnicity without the support of naturally occurring data is clear in the debate over Italian Americans and other groups of European ancestry that has been developing in the past decade among social scientists. Italian Americans are seen by migration scholars as emblematic for the study of immigrants' trajectories in the United States because of their history and characteristics. They are among the oldest and most numerous immigrant groups, as shown by the presence of around 12 million citizens of Italian ancestry in the country (Rosoli 1989). They were the protagonists of the most massive immigration movement in the history of the United States, a wave that brought about 6 million individuals from Italy between 1890 and 1920. Because of their low social position and poverty, they were subjected to relentless racism and discrimination based on their dark skin color and on their association with images of ignorance, violence, and organized crime (Richards 1999). Nonetheless, even in the face of continuous negative stereotyping and discrimination in public discourse and the media, Italian Americans steadily moved up the social ladder. This upward mobility was marked, in the 1950s, by their exodus from the urban “Little Italies” into the suburbs and by their greater integration with other groups, shown in the constant increase in their intermarriage rates (Nelli 1983). Today Italian Americans are in line with national averages with respect to education and professional development (Rosoli 1989), and this is why they are regarded by many as the typical example of a successful immigrant community.

Given these characteristics, they have also been at the center of the debate over the role of ethnicity in immigrant communities. The sociologist Herbert Gans 1979 used the example of Italian Americans to argue that European immigrants to the United States have substantially assimilated, that the type of ethnicity that they cultivate is private and “symbolic,” and that it does not represent a manifestation of identification with the ethnic group of origin. Taking his argument as a starting point, scholars of Italian migration have argued for or against the presence of a strong ethnic sentiment in the community. Alba argues, for example, that the only significant identity for Italian Americans, as for all the other immigrant groups of European descent, is a kind of new pan-European ethnicity, and that we are witnessing the “twilight” of ethnicity, a stage in which ethnic differences “remain visible, but only faintly so” (Alba 2000:60). Others oppose this theory, stressing that Italians are not “just white folks” (Vecoli 2000) and that ethnicity has, in fact, shown an unsuspected vitality in the history of this group (Conzen et al. 1992).

These arguments are based on answers to questionnaires and surveys, but little has been said about ethnicity on the basis of the analysis of naturally occurring data.¹

BACKGROUND DATA AND SUBJECTS

The data analyzed here derive from an ethnographic study carried out between 2002 and 2003 with an-all male card playing club, Il Circolo della Briscola, operating in the Washington area. During the period indicated above I participated in Briscola nights as an observer, took notes, interviewed club members, and made video and audio recordings of games and other types of interactions. The data presented here come from one videotaped and five audiotaped sessions of about 2 hours each.

IL CIRCOLO DELLA BRISCOLA

Il Circolo della Briscola is an all-men club funded in 1991 by an Italian American doctor (called the President), an American of Calabrese origin, born in New York. The President started to play with friends as a recreational activity, but then decided to initiate a club because of the interest that his initiative sparked among friends and acquaintances. Below I describe some of the main aspects of the group's life in order to provide a frame for the analysis of the symbolic practices in which and through which ethnicity is built and negotiated.

Since 1991 the Circolo has steadily grown and has become an organization with members, regulations, and activities throughout the year. At the moment in which I conducted my research, the Circolo had 48 members who met monthly to play briscola or tressette. Briscola is a popular Italian card game. It is usually played by four participants in teams of two competing against each other. The game is played with a deck of 40 cards, which have specific numerical values. It is a trick-taking game – that is, its objective is to take cards that will give the team a high score. Every time the game starts, one suit of cards is randomly assigned as the trump (briscola), and players can use trumps to beat all other suits and assemble points by picking up the cards on the table. Briscola players are allowed both to communicate with each other through secret signals and openly to advise each other. Tressette is similar to briscola as it is played with 40 cards that have specific point values and its object is to take tricks containing valuable cards. There are no trump cards, and so the players score points when they have the highest card of the suit established by the first player. As in briscola, Tressette players play in couples and can negotiate moves with each other.

Once a year the Circolo plays at a parish in Washington, DC (which is recognized as an important center of activities related to Italy and Italians) in order to raise funds for the church itself. Until November 2001, games used to take place at an Italian restaurant and, before that, at members' homes. At the end of 2002, however, the club started meeting at a local language school and cultural center very close to the church. Results of the games and general information are published in the Briscola News, distributed by the President.

The club meets once a month, except when there are special tournaments or special occasions such as picnics or parties. There is a regular agenda. The evening starts with hors d'oeuvre and dinner at 6:30, followed by games. Before and during dinner the President and Treasurer collect money and organize a draw in order to form tables of players. During this period the President, or occasionally other members, also makes announcements, discusses or takes votes on issues related to the club, and introduces new members. Games start at around 8 p.m. Usually players are assigned to tables through the draw, but the President can decide to form tables according to other criteria and always sets up a table for new players who want to become familiar with the game. There are around 30 people each night, and most players play briscola, but there are also one or two tressette tables. At the end of all the games the score is updated, and when the season is over the winner (or winners, if there is a draw) gets a trophy.

As mentioned, the members of the Circolo are all men. The admission of women has been voted on and rejected numerous times, the official explanation for this being that men want to spend the evening by themselves, or, as one member said, “with the boys.” Most of the members are between the ages of 55 and 65, so many of them are retired. There are very few younger members. The men's social background is middle and upper middle class. Some of them are professionals such as architects, engineers, or medical doctors, and others include public employees, schoolteachers, travel agents, and others. As it emerged both in the interviews and in comments made during the games, there are differences in their social perceptions of each other, with some of them being perceived as “rich” or “powerful” and others as more modest; in general terms, however, all the members belong to the middle or upper middle class.

In terms of their origin, members can be divided into two groups. The first group is composed of 25 second- or third-generation immigrants who often come from traditional areas or cities of migration in the Northeast, such as New York state or city, Baltimore, or Pittsburgh. This reflects the migration history of Washington, since the metropolitan area was not an important center of Italian migration but rather received immigrants from other regions. The second group is composed of 23 first-generation immigrants who were born in Italy. Of these, 17 came to the United States as adults mainly to study or work, while 6 came as very young boys when their families migrated there. This second group also reflects the migration patterns of the city, since Washington attracts a continuous flux of new immigrants working in federal and international organizations and in education. Thus, all the club's members are well assimilated into American society and follow the national trend in their belonging to a fairly educated middle or middle upper class. There are, however, also many internal differences in terms of profession, age, and migration history that are reflected in differences in language repertoires and usage, since most Italian-born members speak both Italian and English fluently, while most U.S.-born members do not speak Italian. I will come back to this point when I discuss code-switching and choice.

COMMUNITY PRACTICES AND THE CONSTRUCTION OF ETHNIC IDENTITY

Circolo members engage in different practices in which they enact, project, and negotiate identities of various kinds and at different levels: from situational identities (Zimmerman 1998:95) such as that of cardplayer, treasurer, or president, to wider social identities such as those related to ethnic or gender categories. For example, given the nature of the club as an all-male organization, in informal conversation members often make relevant their gender identity. Identities can also combine with each other so that a member performing, claiming, or discussing an identity as a man may, at the same time, claim an identity as an Italian man, or a member who is presenting himself as a professional may also emphasize his role as a male.

However, while the inventory of identities that can become relevant for individuals in specific interactional moments is wide and relatively open, the aspects that are promoted as relevant for a collective identity are much more subtly regulated because they are closely linked to the functions and objectives perceived as fundamental to the community in general. Collective identities are less contingent upon local factors because they represent the image that a community has and projects about itself. They have to do with what people think characterizes them as a group that is different from others. Thus, the shaping of a collective image for a social organization reflects implicit conceptions about the qualities and traits that are perceived as defining the community. However, collective identity is not a mental image, but rather an emergent construct that takes shape within significant practices. There is an important connection between being and doing, between building and projecting an image of the group and carrying out certain activities. This relationship between social action and identity in groups is well captured in the sociolinguistic construct of community of practice (Eckert & McConnell-Ginet 1992), which conceptualizes a community as a group defined by the existence of a joint enterprise, mutual engagement, and a shared repertoire of resources (Wenger 1998:73). Looking at a group as a community of practice helps analyze its identity as emerging through specific intragroup practices (Del Negro & Berger 2004), and specifically through the association between socially relevant traits and activities constructed as central to the life of the community.

It is from this perspective that the “card-playing” identity appears as an essential trait of the collective image of the Circolo della Briscola. The club's main official objective as a community is to play briscola and, as the club website puts it, to “enjoy an unparalleled competitive camaraderie, in our monthly quest for an annually-awarded trophy.”²

Organizational practices have to do with the concrete functioning of the club: They include the choice and decoration of places, food preparation and consumption,³

Within these kinds of practices, identity work is done at different levels and in different ways. For example, the use of symbols of Italian identity transforms practices related to the functioning and organization of the club into practices that index Italianness. The choice and decoration of meeting places, food preparation and consumption, the distribution and use of gadgets and utensils necessary for the games are all transformed into activities embodying an Italian identity when they are invested with specific symbolic elements. For instance, the choice of meeting places is not casual: The Circolo gathers either at Italian restaurants or, more recently, at a center for the teaching and learning of Italian run within a Washington parish traditionally associated with the Italian community. The room where players meet to have dinner is arranged in a way that recalls the big family dinners traditionally associated with Italian culture: A very long table is placed in the center, and all the members sit together. The choice of places also enforces a symbolic association with Italy through names (the restaurants always have Italian names), activities (such as cooking Italian food, or teaching Italian), and institutions (such as the Catholic Church and religion). People belonging to the Circolo della Briscola do not simply meet to play, they meet to play in places that underline their Italian origins.

Food preparation and consumption are particularly central for the indexing of Italian identity within the Circolo. Food and food practices are more or less implicitly presented and lived as “traditional,” as shown by the choice of menus that always offer Italian or Italian American dishes, the practice of bringing food made at home by the players or their wives, and also the preparation of specific Italian foods related to traditional religious or mundane occasions. Food is also at the center of discourse in the Circolo, since members make comments and discuss it at every meeting, talking about the quality of the dishes offered at the table, their origins and authenticity, the way they are prepared, and so on. They tell stories about how similar dishes were made in their homes and talk about food or drinks that they have prepared according to traditional recipes. Thus, food preparation and consumption are important arenas for individual identity displays, which range from Italian American heritage to regional Italian identities, while also contributing to the perception of Italian identity as a collective point of reference that defines membership into the club.

Another set of organizational practices that becomes an arena for the construction of a collective ethnicity is the production of printed materials such as badges and T-shirts for the players to wear, or of the newsletter containing the latest game scores. These practices index Italian identity through the use of colors and other symbols. In fact, the selection of green, white, and red used for the Briscola News, and the T-shirts and the badges that players wear at card games, symbolizes Italianness by association with the colors of the flag. The newsletter also presents the image of an Italian card with its name in Italian: Fante di coppe sardo ‘Sardinian jack of cups’. Finally, the exclusive use of Italian cards in the games and the President's assembling and display of a collection of Italian regional briscola cards on special occasions indexes, in the same way, the club's interest in and allegiance to Italian traditions.

But among the strategies that have the greatest role in indexing ethnicity, language use appears to be the most important. It is for this reason that I focus here on the strategic use of code choice and switching in public discourse and in socialization practices to illustrate how Italianness is constructed as a central feature of the club's identity.

LANGUAGE REPERTOIRE AND CODE-SWITCHING IN THE CIRCOLO

The centrality of language use in the expression of ethnicity among multilingual communities is an established fact in sociolinguistics. The Circolo is no exception, as language choice and code-switching figure as the most prominent linguistic strategies in the negotiation of Italian ethnicity. In this article I use the term “code-switching” to indicate “the juxtaposition within the same speech exchange of passages of speech belonging to two different grammatical systems or sub-systems” (Gumperz 1982:59). Thus, I use the term as an umbrella expression for different phenomena including word insertion and intrasentential and intersentential switching. The symbolic functions of code-switching in verbal interaction within the Circolo cannot be understood without reference to the linguistic repertoire and linguistic competence of its members, because the languages spoken by different members include English, Italian in its regional varieties,⁴

Competence in all these different codes varies widely in the Circolo. While all members speak English fluently, only some of them are fluent in Italian. The group of fluent Italian speakers includes all of those who were born in Italy, but only few of those who were born in the United States. In addition, there are American-born players who speak some Italian and others who understand it but cannot speak it. Also, competence in a dialect varies because there are American-born members who still remember some dialect, Italian-born members who speak their dialect well, and Italian- and American-born members who do not speak any regional dialect.

Given these differences in linguistic competence, the potential meaning and functions of code-switching in relation to ethnicity are extremely complex. Many classic sociolinguistic studies of code-switching presuppose a stable relationship between the use of the language variety spoken by the ethnic group and the expression of in-group ethnic solidarity. Thus Blom & Gumperz 1972, Gal 1979, and Milroy 1987 regard the different codes spoken by bilinguals as associated in a rather fixed way to a “we” or a “they” representing respectively the in-group and the out-group. But all these studies dealt with homogeneous communities with a high degree of bilingualism, while here I am dealing with a community that is linguistically diverse. Second, a direct association of codes with symbolic rights and obligations does not recognize the fact that speakers creatively use symbolic resources to construct, not merely reflect, social meanings and categories. The analysis of the use of code-switching within the Circolo is then best understood if code-switching is seen, as in many interactionally inspired models of bilingual speech (Alfonzetti 1998; Auer 1998, 2005; Meeuwis & Blommaert 1994; Woolard 1999) as a strategy whose meaning needs to be investigated within specific interactional contexts.

An analysis of language use in the Circolo confirms the hypothesis that there is no necessary association between the use of Italian and the expression of in-group solidarity. There are, as happens in many other plurilingual communities (Fishman 1972), general associations of language varieties with activity domains, which are partly determined by the fact that English is the lingua franca. Thus, in public oral and written communication addressed to all members by the President or other individuals about topics of general interest, English is mostly used. During these occasions switches into Italian occur and are expected, but Italian does not constitute the base language. On the other hand, in informal conversation and during the games, Italian can be the predominant language in an exchange or a whole game if all players are Italian-dominant. The use of dialect, on the other hand, is much less frequent and more marked (see De Fina forthcoming 2007 on this point) because most club members either do not speak any dialect or do not speak the same one. Whichever the base language, code-switching occurs frequently, but it does not always index ethnic identification or disengagement.

In many situations code-switching appears to function as a contextualization cue (Gumperz 1982) to convey changes and moves that reflect local interactional dynamics. In addition, even language choice does not necessarily depend on situational factors such as linguistic competence, since there are many cases in which men who are not particularly fluent in Italian try to speak in this language, or men who are very comfortable in Italian and are born in Italy speak English. The following example reflects both a non-situationally determined choice of language and the metaphorical use of code-switching. The fragment is taken from a mock Tressette game in which Enzo and Carlo are teaching Peter⁶

Notice that both Italian-dominant speakers talk to each other in English (lines 01–02 and 10–12), even if the others at the table are fluent in Italian. Notice also that they code-switch in a way that is unrelated to speaking ability or interlocutor's origin. The switches can rather be classified as metaphorical (Gumperz 1982) in the sense that they mark subtle changes at the interactional level. Enzo starts the fragment speaking in English when addressing Carlo (lines 01–02). Carlo, on the other hand, speaks to Peter in Italian when challenging his move (lines 04–05). Following Carlo's choice, Peter responds in Italian (line 06) and Enzo also switches into Italian (line 07), thus changing the language with respect to his previous turn. The switch may be due to the fact that his turn is a repetition of Peter's words addressed to Carlo as a clarification of Peter's intentions. At this point Carlo acknowledges the clarification in English in his first utterance (line 08), but then switches into Italian in a move that appears to be a shift from acknowledgment to explanation. Enzo also chooses Italian to address Peter and again, to explain the game to him (line 09), but then switches back into English when he is justifying his verbal actions with Carlo (line 10). After that, he goes back to Italian for a new explanation about the game addressed to Peter (lines 13–14). In this example, language choice and switching are not situationally motivated, and code-switching appears to be discourse-motivated (Auer 1998) because it indexes subtle interactional moves. However, code-switching into Italian can become a powerful strategy of ethnic alignment and of in-group cohesion. How can we recognize such marked instances? The connections between code-switching and ethnicity can be established only by looking at recurrent patterns, the type of practice in which interactants are engaged, and the sequential development of talk. Below I analyze instances of code-switching in public discourse and socialization practices as examples of the symbolic use of this linguistic strategy.

CODE-SWITCHING AS A STRATEGIC CHOICE IN PUBLIC ADDRESSES AND SOCIALIZATION PRACTICES

Code-switching from English into Italian acquires special prominence in language practices related to official business – that is, in oral public addresses by the President or other members to the rest of the club and in the Briscola Newsletter, also written and circulated by the President. In these domains of action English is expected because not all the addressees speak Italian, and thus code-switching appears as a marked choice. On the other hand, public communication constitutes a central arena for the construction of a collective image of the Circolo and the President, who has much greater influence than all the other members precisely because of his management of communications, uses many strategies to mold that image. One of them is code-switching or code-mixing, the insertion of entire utterances or single words in Italian within an utterance in English.

As an example, we can look at one of the messages addressed by the President to Circolo members through the Briscola Newsletter:

In the letter, besides the greetings (Ciao, cari soci), all the words in Italian either refer to Italian food (ziti al forno, dolci, lasagna avellinese) or to items related to the Briscola game (partita). The President regularly inserts Italian words referring to these two fields in his communications with the members. In fact in the Briscola Newsletter the most frequently inserted Italian words refer to these two fields: food and the games (for a more detailed analysis, see De Fina forthcoming). These insertions therefore are not casual but highly symbolic. They serve the purpose of indexing Italian ethnicity in connection with two vital areas of the Circolo's life: playing cards and cooking. By consistently using the Italian terms for Italian dishes and for expressions related to the game, the President is emphasizing the Italianness of these areas and therefore also the traditional character of the club.

A similar trend is found in the President's oral addresses to the members. These take place during dinner and are centered on topics of general interest, such as dates for games and other events, payments due, the players' standing in the annual tournament, and so on. The following fragment in taken from one of these addresses:

Here the President is discussing the choice of food for the evening and for the next game. On line 08 he inserts the term pasta con vongole [pasta with clams] in Italian, but notice that in line 09, where he wants to know whether the choice is acceptable to everybody, he translates the term vongole for his audience (clams or clam sauce) in a move that seems to be motivated by a practical reason: the need to make sure that everybody has understood. The recourse to a translation confirms the fact that the President does not take it for granted that members of the club understand Italian, but he uses that language anyway for strategic words or utterances as a symbol to stress ethnicity. The translation also confirms the status of the utterance as a switch and not a borrowing, as could be the case for the word pesto (line 03). Thus, code-switching here is neither casual nor related to discourse function, but it has a highly symbolic value in building the collective identity of the Circolo as Italian in its traditions through its association with the cards and the cuisine.

The emphasis on Italianness in the game domain is also central to the construction of a collective ethnic identity because of the nature of the Circolo as a community whose main objective is to play cards. For this reason, social practices related to the game are powerful sites for identity work. The socialization of new members into the game has a particularly important role because it is in this area that implicit rules and expectations about the behavior of members in the Circolo are revealed and enforced.

Socialization practices in the club have to do with welcoming new members into the Circolo, introducing them to the rules and regulations of the club, and teaching them Briscola or Tressette, if they do not know how to play. Tables for newcomers are set up by the President, who usually participates in the teaching with two other members. Verbal exchanges between old and new players show how the Italian identity of the Circolo is emphasized and reinforced through language alternation. In fact, while, as we have seen, in general language choice is highly negotiable and depends partly on the player's ability to speak Italian, and partly on accommodation to the preference of specific participants in the game, in socialization practices a certain type of language alternation is enforced. Players establish a strict association between card playing and speaking some Italian, so that it is tacitly understood that a good cardplayer in the club needs to be able to speak at least the basic words of the game in the native language. As a result, when neophytes learn the game they are also taught some of the terms for the cards and moves that characterize the game, and are expected to learn them. In the following fragment the President and Rob are teaching Andy, a newcomer, how to play the game. All the players are American-born, but the President is the only one fluent in Italian. Andy is the grandson of an Italian-born member. He has recently joined the club. The players have just started the lesson:

At the beginning of the fragment the President and Rob are negotiating with Andy on whether he wants to be taught directly or through an “open hand,” a game in which the cards are on the floor and visible to everyone in order to make the rules of the game clearer (lines 01–03). Andy accepts the open hand (line 04), noting that everyone usually learns that way (line 06). Rob confirms that this is normal practice for instructors (line 07). At this point an instance of code-switching occurs (line 08), where the President says the term open hand in English and then translates the term into Italian. We can interpret this move as the first lesson for the neophyte on terms that refer to the briscola game. In fact, the President models the term for the new player as he and the other instructors will do later in the same interaction with the names for the cards. But also notice how, in line 10, the President explicitly asks Andy if he speaks any Italian, and how then, having received a negative answer (line 11), he makes a joking announcement to the whole group about Andy's inability to speak the language (line 18). Andy assumes a defensive stance when he confesses that he can only speak Spanish and offers to get help from his grandparents, who happen to be present and who are first-generation Italian immigrants (line 15). Questions about language ability in Italian are never asked in normal conversation, as it is understood and accepted that members of the club may be English monolinguals. We saw in example (3) that translations of Italian words referring to food in Italian are provided by the President himself precisely because it cannot be taken for granted that all the members speak and understand the language. However, example (4) above and the ones following below clearly indicate that the use of some Italian is required at the game table. When games are played among competent members, these are heard using Italian terms for cards and moves even if they speak no Italian at all. See the following example:

Here Lino, John, and Franco are Italian-born players and can speak Italian, but Dave is American-born and does not speak the language. Lino is teaming with Dave and addresses him in English on line 01, but notice that he uses the word briscola ‘trump’ in Italian, and that Dave does the same in his answer (line 02). Then Lino again uses the word carico ‘a high trump’ in his following question, using the same pattern (line 03). Notice that the other lines in Italian are cases of code choice that reflect the fact that both Franco and John can speak Italian fluently and therefore can address each other in that language (lines 04 and 06). A question that could be raised about words like briscola and carico designating card functions is their possible status as loan words and not as code-switches. Most researchers (e.g. Poplack, Wheeler & Westwood 1987) propose that loan words, contrary to code-switches, are phonologically and morphologically integrated into the base language. Phonological integration is difficult to establish in the case of these words because players have varying competences in Italian phonology, and therefore they may be unable to pronounce even a code-switch with a native-like phonology. However, the status of game words as code-switches is confirmed by their lack of morphological integration (see also example 8, below, on this point) and also by the presence in the language spoken by the players of variants such as briscoletta ‘small briscola’ that carry the Italian morpheme -etta for the diminutive. In addition, the fact that they are explicitly taught to new players indicates that they are treated as foreign terms.

The examples above show that there is an implicit expectation about Circolo members that they need to use some Italian when playing cards. As we have seen in example (4), the existence of this implicit rule is confirmed when newcomers are introduced to the rules of the game. Another example of this association between playing Briscola and speaking Italian is found in the following fragment, where Paul and Al are trying to teach Carl how to play briscola. Both Paul and Al are American-born and know very little Italian. As a first step they are showing Carl how to recognize cards. Carl also is American-born and does not speak any Italian:

In this fragment Carl is having trouble recognizing Italian cards and is asking his companion to clarify what each card is (lines 01–09). In line 05 he makes a gesture of desperation when he puts his hands on his head, and finally asks disapprovingly why people do not use American cards (line 10). Paul responds explaining that it is not ‘traditional’ (line 13) Notice that he code-switches into Italian to utter the word tradizionale. This is an important utterance because Paul is making explicit the implicit connection between playing Briscola and reviving Italian traditions that constitutes the basis for the ethnic allegiance of the Circolo. This process has been characterized by Bauman (1992:128) as “traditionalization,” or “a symbolic construction by which people in the present establish connections with a meaningful past and endow particular cultural forms with value and authority.” In this case, playing with Italian cards means adhering to a traditional Italian identity. Also notice that Paul makes a point of translating the names of the cards in lines 15 and 19 even though his competence in Italian is limited (see discussion of line 03 in example 7, below). The result of this teaching is that Carl starts accepting not only the need to recognize Sicilian cards, but also the need to learn their names, as shown in the conversation below, which took place later in the same game:

In this series of exchanges, we see that competent players teach the cards and their names while at the same time projecting the importance of mastering the Italian language for the game. It is also relevant to notice that American-born players make an effort to utter the words in Italian even if their language competence is relatively low. In excerpt (7), for example, Carl tries to resolve his uncertainty about the name of a card by addressing the researcher (line 03). However, he does make an effort to pronounce the names of the cards in Italian even if his utterances are not completely native-like, as in the case of cavallo bastoni (line 01), which should be cavallo di bastoni. These examples demonstrate that the use of the Italian jargon for the cards is sternly practiced because it is associated with the projection of an identity as a competent card player. We can see in excerpt (7) that Carl is being socialized into this practice when he shows that he has accepted the need to speak Italian and seeks confirmation for the name of the card in Italian (line 01). Thus, during card games code-switching from Italian into English and from English into Italian is expected, but insertion of Italian words and expressions for cards and moves is enforced.

The examples discussed show that the situational identity of “cardplayer,” which is central to Circolo membership, is continuously associated with Italian identity through code-switching into Italian. This practice is important not only to encode a competent player identity but also to index a collective Italian identity for the Circolo members in that it helps shape the club as a locus for the maintenance of tradition by inviting players to act as traditional Italian players. The value attributed to Italian language use during the games is emphasized by the existence of episodes of language that could be classified as a kind of “language crossing” (Rampton 1995), since in them speakers who are not fully competent in Italian try, nonetheless, to produce utterances in that language. In fact, although these speakers are of Italian ancestry, Italian has become to them equivalent to a foreign language. The following is an exchange between three American-born players during a game of briscola. All the interactants, Anthony, Joseph, and Sean, understand Italian, but their productive ability is quite limited:

In line 01 Anthony inserts the Italian word divertimento ‘amusement’ in his English utterance. Anthony is of Sicilian origin and knows only a few words in Italian and in Sicilian, but he often practices them by inserting them into his utterances in English. It is interesting to notice that Joseph, who also speaks just a few words of Italian, tries to pair his utterance with Anthony's by producing an Italian translation of the word ‘serious’ in Anthony's utterance. What he produces, however, is a mixture of Italian serio and English serious: the newly coined term serioso (line 02), which Anthony immediately repeats in the following turn. Then Sean intervenes, going back to the game with a comment on the fact that he always has high trumps that he cannot use (line 04). Notice, again, that the word carico ‘high trump’ is produced in Italian, and its status as a code-switch is confirmed by the absence of morphological assimilation into English, since it does not carry any plural morphology. The insistence on the use of Italian at the card table in instances of both crossing and code-switching, as shown in the previous example, illustrates the point that the language is used symbolically to stress Italianness.