622   Spanish in the USA, Heritage Language, L2 Spanish
1
Villa, D. 1996. Choosing a "standard" variety of Spanish for the instruction of native Spanish speakers in the US. Foreign Language Annals 29.191-200.
Villa, D. 2000. Languages have armies, and economies, too: the presence of US Spanish in the Spanish-speaking world. Southwest Journal of Linguistics 19 (2). 143-154.
Villa, D. 2002a. The sanitizing of US Spanish in academia. Foreign Language Annals 35 (2). 222-230.
Villa, D. 2002b. Integrating technology into minority language preservation and teaching efforts: an inside job. Language, Learning and Technology 6 (2). 92-101.
Villarreal, H. 1981. Reading and Spanish for native speakers. In G. Valdes, A.G. Lozano, R. Garcia-Moya (eds), Teaching Spanish to the Hispanic bilingual: issues, aims, and methods, 157-168. New York: Teachers College Press.
Wiley, T.G. and G. Valdes. 2000. Special issue on heritage languages. Bilingual Research Journal 24 (4), i-v.
Winke, P., C. Stafford, and ERIC Clearinghouse on Languages and Linguistics. 2002. Selecting materials to
FURTHER READING
teach Spanish to Spanish speakers. ERIC Digest (No. EDO-FL-02-03). Washington, DC: ERIC Clearinghouse on Languages and Linguistics. Available at: http: // www.cal.org/ericcll/ DIGEST (last accessed March 29,2010).
Zapata, G., L. Sanchez, and, A. Toribio. 2005. Contact and contracting Spanish. International Journal ofBilingualism 9 (3/4). 377.
Ziegler, J. (1981). Guidelines for the construction of a Spanish placement examination for the Spanish-dominant Spanish-English bilingual. In G. Valdes, A.G. Lozano, and R. Garcia-Moya (eds), Teaching Spanish to the Hispanic bilingual: issues, aims, and methods, 211-214. New York: Teachers College Press.
Zentella, A. (ed.). 2003. Latin® responses to racist discourse. New York: Palgrave.
Zentella, A. 2003. Latin® responses to racist discourse. In D. Sommers (ed.), Bilingual games: some literary investigations, 51-66. New York: Palgrave.
Zentella, A. 2005. Building on strength: language and literacy in Latino families and communities. New York: Teachers College Press.
Migration Policy Institute. 2009. US
Historical Immigration Trends. Available at www.migrationinformation.org/ DataHub/historicaltrends.cfm (last accessed March 29, 2010).
National Center for Education Statistics. 2007. Status and trends in the education of racial and ethnic minorities. Available at: http://nces.ed.gov/pubsearch/ pubsinfo.asp?pubid=2007039 (last accessed March 29,2010).
29  Language Maintenance and Language Shift among US Latinos1
JORGE PORCEL
Language maintenance and language shift (LMLS) are outcomes of societal bilin-gualism. However, in western societies, language shift (LS) is by far the most frequent resolution. In these societies, the language of the group in power usually displaces minority languages. This phenomenon is far from universal. In fact, scholars have traced it back to the constitution and consolidation of modern European nation states and their success in promoting the ideology of monolin-gualism. This ideology considers that monolingualism is the natural state of human beings, and that monolingual societies are superior to multilingual ones (Gal 2007: 149). President Theodore Roosevelt summarized this ideology in the following statement:
We must have but one flag. We must have but one language [...] The greatness of this nation depends on the swift assimilation of the aliens she welcomes to her shores. Any force which attempts to retard that assimilative process is a force hostile to the highest interests of our country ...(Roosevelt cited by Crawford 1992 [1917]: 85)
President Roosevelt could not be more emphatic in declaring the importance of monolingualism for the well-being of the state. However, the previous statements constitute more than a personal stance. This overt declaration of war against languages other than English has been the unstated language policy of the United States toward language minorities since the beginning of the twentieth century. In effect, in the United States, the ideology of monolingualism has been a powerful force in the legitimization of English, and in the disenfranchisement of language minorities. Although widespread in modern Western thought, this language ideology is far from being a universal belief (see Dorian 2006: 443-449). This is something we should keep in mind as we study the phenomena of LMLS. Also, we
The Handbook of Hispanic Sociolinguistics, edited by Manuel Di'az-Campos © 2011 Blackwell Publishing Ltd.
624   Spanish in the USA, Heritage Language, L2 Spanish
Language Maintenance 625
should be aware that the statements and generalizations made in this article refer to tendencies observed in western societies, and more specifically in the Latino communities of the United States.
1 The dynamics of LMLS at the societal and individual levels
The terms LMLS do not designate static phenomena but ongoing processes whose directionality may change at any time. These processes take root at individual practices; it is the individual who makes the conscious or subconscious decision of keeping, limiting, or abandoning altogether the use of a language in certain or every situation(s). However, LMLS research is interested in these processes as societal phenomena. In effect, the difference between the societal and the individual level is not only quantitative but also qualitative. It is evident that individuals' decisions have an effect upon other individuals' decisions. Theories such as Adam Smith's invisible hand - used to explain economic inflationary processes, stock market panics, or diachronic language change - are meant to deal with such influences. Furthermore, behaviors such as cooperation, competition, and solidarity become meaningful at the aggregative level, in which individuals are not isolated but framed by social relations. Therefore, although social factors conditioning LMLS operate directly upon individuals, they do not react idiosyncratically to these variables. Since the aggregative level mediates individual responses, we perceive individuals reacting within a range of recurrent responses vis-a-vis said social factors.
2 Language maintenance (LM), language shift (LS) and language loss (LL)
At the phenomenal level, LMLS result from the perception that a change - or lack thereof - in the pattern of habitual language use has occurred for a given speech community (Fishman 1972a: 76). Providing evidence of one of these two processes constitutes the initial step of any research on LMLS. We can label this initial research step as language assessment. Language assessments should be supported by at least two measures taken at two different times (real time technique) or two different generations (apparent time technique) for the same language Lx and speech community Cx. Thus, LM corresponds to the situation in which the measures remain constant. When we have a discrepancy between the measures, we are facing LS. LS can be regressive (Lx is losing ground to another language LY) or progressive (Lx is (re)gaining ground). Finally, LL designates the special case in which a regressive shift has reached completion and the language is no longer used in the community. Still the definition of LM requires a criterion to decide how long a habitual pattern needs to last to count as maintenance. The rule of thumb is
Figure 29.1   The two basic stages in the study of LMLS. Source: Author.
that maintenance refers to a situation "that has persisted without dramatic change for more than three or four generations and that shows no sign of incipient change" (Thomason 2001: 23).
2.1   Stages in the study of LMLS
There are two basic stages in the study of LMLS (see Figure 29.1). The first, language assessment, focuses on measuring language practices to demonstrate a change - or lack thereof - in the habitual patterns of language usage. Three types of measures are commonly used at this stage: direct, indirect, and surrogate (see below for discussion). The second stage, the sociolinguistic assessment, isolates the external variables that may provide an explanation for the directionality of the observed process. In Figure 1 we have divided these into three categories: (i) variables related to the structure of the speech community (who uses the language with whom and for what purposes); (ii) structural variables, which locate individuals (or an aggregate thereof) in the broader society; (iii) attitudinal-behavioral variables, which account for in-/out-group dispositions and reactions toward the competing languages. Since the variables in the sociolinguistic assessment stage are "mutually modifying dimensions [...] interrelated and often interdependent" (Mackey 2006: 617), it is almost impossible to disregard their interconnection,
626   Spanish in the USA, Heritage Language, L2 Spanish
Language Maintenance 627
which will become evident through this presentation. The rest of the chapter will develop the components presented in Figure 29.1.
2.2  Assessment of language practices
To determine the occurrence of LM or LS, we must select ways of measuring language practices, by using one or more of the following techniques: direct, indirect, and surrogate measures. Once the selection is accomplished, we need to obtain, for the same community, two or more of the said measure taken at two or more time points (real time technique), and/or taken for two different age cohorts or generation groups (apparent time technique).
Direct measures: these consist of the direct observation of participants' performance. Zentella (1997) followed this approach in her study of a Puerto Rican community in East Harlem, New York City. She studied the displacement of Spanish by English comparing bilinguals' competences in real time: in 1980 and 1993. Using a scale of proficiency based on the verb tenses participants produced when answering to a number of questions, she placed respondents in a bilingual continuum ranging from Spanish to English monolingualism. Balanced bilinguals occupied the middle point of the scale. One indicator of LS was that none of the subjects fit into the category Spanish monolingual. Another indicator was that English monolingualism had tripled from 1980 to 1990 (Zentella 1997, chapter 9).
Indirect measures: these rely on participants' behavioral self-assessment. Porcel (2006) used this approach to study the language situation of Miami Cuban Americans. Participants completed a sociolinguistic questionnaire, which contained questions such as: (1) What was/were the first language(s) (L,) you spoke? (2) What language(s) do you feel most comfortable using? (3) What language(s) did you used at home when you were a child? (4) What language(s) do you use at home now? (v) On a daily basis, what percentage of Spanish do you normally use? For questions one through four, the response categories were "English, Spanish, or Both." For question five, respondents could mark any percentage from zero to 100. Several indicators demonstrated the ongoing LS. First, although all participants claimed Spanish was their Lv English had become the preferred language for 37% of the sample. In addition, the use of Spanish at home had dropped from 91% to 53% from childhood to the moment of the study, and 28% estimated that on daily bases they used Spanish less than 50% of the time.. Therefore, we may conclude that some participants are undergoing LS - either because they never fully developed their Spanish competence, or because they started using it less. Brezinger (1997: 283-284) conceptualizes this process in terms of a receding spiral: the less the minority language is used, the less the exposure to it, which results in decreasing proficiency and confidence in the use of the language, along with increasing reliance on the dominant language. Porcel (2006) also used the apparent time by controlling for the generations to which the respondents belonged. Since there were significant statistical correlations between the second generation - that is US born or foreign born arriving before the age of five - and the previous indicators of LS, it was possible to conclude that the second generation was leading the shift
and that the higher the generation number, the lower the Spanish proficiency, the use of Spanish at home, and the use of Spanish on daily bases would be.
Surrogate measures: surrogate measures are "often based on a logical relationship between the measure and the language variables of interest" (Macias 2000:11). This technique is frequently used when linguists resort to nationwide databases such as the US Census Bureau. For example, Veltman (2000) analyzed census data to establish the state of LM among US language minorities. One problem he faced was that US census questionnaires do not elicit information about the mother tongues of the US population. To obtain this data, Veltman (2000: 82) used the categorization by origin into specific ancestry groups - such as Mexican, Puerto Rican, Cuban, etc.- and used this indirect measure as surrogate for Spanish mother tongue. The logical connection is that if a person claims a specific Hispanic origin, s/he "is ancestrally connected to the Spanish language," (Bills, Hernandez Chavez, and Hudson 1990: 16). Also, based on census data, Hudson, Hernandez Chavez, and Bills (1995: 166-167,178-180) designed two constructs to measure language loyalty and retention. The index of language loyalty was considered to be the ratio of Hispanics claiming they use the language at home to the total of individuals claiming Hispanic origin. To calculate retention, they divided the index of language loyalty for the younger cohort (ages 5 to 17) by the adult cohort (18+). (Notice both indexes are surrogate measures, and also that retention uses the apparent time technique.)
Indexes of language loyalty and retention
Total of Hispanics claiming use of Spanish at home anguage oya y     Total Hispanic origin claimants
^ _ Index of language loyalty for the younger cohort
Index of language loyalty for the adult cohort
So far, we have illustrated the first step any study of LMLS needs to accomplish, and the types of approaches used to assess ongoing processes of LMLS in a bilingual community. The following paragraphs will be devoted to the second step.
2.3   The sociolinguistic assessment
A language assessment will be incomplete if we do not correlate linguistic behaviors (i.e. LM or LS) to social factors. I will call this second research stage the sociolinguistic assessment. Sociolinguists have struggled, with little success, to find hierarchies of variables or implicational relations which could help to make safe predictions about LMLS. A good proof is that the presentation of these variables usually adopts the form of an open-ended list (e.g. Edwards 1992; Paulston 1994). In general, it is possible to indicate the catalytic effect of certain variables for particular case studies. However, not many wide-ranging generalizations can be drawn, because it is frequent that variables either " 'cut both ways' in different contexts or have no general significance when viewed in broader perspective" (Fishman 1972a: 95; 96-100). The sociolinguistic assessment represented in Figure 1 groups
628   Spanish in the USA, Heritage Language, L2 Spanish
these variables into: (i) variables which affect the structure of the speech community, that is how language is used and takes on meaning in the minority groups' lives; (ii) socio-structural variables, which locate individuals - or groups - in the broader society; and (iii) attitudinal-behavioral/subjective variables. In the following sections we will review the impact of a set of independent variables, grouped in this threefold division. Consequently, the first section will discuss the factors related to the structure of the speech community.
2.4   Variables structuring the speech community
Language contact and societal bilingualism: language contact is a precondition for the emergence of either LM or LS (Brezinger 1997: 282; Fasold 1987: 213). But what do linguists refer to when they talk about language contact or societal bilingualism? At the most obvious level, two or more languages must coexist within a single speech community. Notwithstanding, a bilingual speech community does not need to be one in which everyone - or not even the majority - is bilingual. In Western societies the opposite is often the case: members of minority groups become bilinguals by adding the language of the dominant group(s) (DomL) to their home language. Contrarily, the members of the dominant group do not rush to learn the language of subordinated groups (SubL), and if they learn it, the SubL does not become their preferred code of communication nor does it displace their mother tongue. Therefore, LMLS usually involves a SubL. Also, minorities do not select the DomL that they add to their repertoire. They acquire the DomL of the community where they live, work, and socialize, because this language represents a way of avoiding marginalization and improving their social conditions (Siguan 2001:19). In this section, I have delineated a basic description of language contact and societal bilingualism. The following paragraphs will be devoted to present how the various situations of Spanish and English in contact affect the dynamics of LMLS of Hispanics in the United States.
Onset of contact and minority status: the specific group that occupies the land at the onset of contact defines two situations: If the original occupants end up absorbed into the power structure of the newcomers, this group becomes a national minority. If the newcomers integrate into the existing social structure, they emerge as an immigrant minority. In either case, the minority language suffers devaluation, as one of many consequences of becoming minoritized (Safran 1999: 84). Minority is essentially not a quantitative concept. It refers to groups "singled out for unequal treatment and who regard themselves as objects of collective discrimination," and their unequal treatment can surface as based on physical (racial) or cultural (ethnic) differences (Henslin 2001: 321).
Therefore, minoritization is subordination to another group, which is larger, more powerful, efficacious or capable of providing benefits to minoritized groups. However, in terms of LM and other cultural institutions, national minorities have material and psychological advantages over immigrants. On the one hand, their institutions are already in place and at work; they do not have to transplant them and build them from scratch, as immigrants must do. More importantly, national
Language Maintenance 629
minorities feel a sense of legitimacy and entitlement, which plays an important role in LM, and which immigrant groups lack. Mexican Americans, who settled in the southwest at the time of the signing of Guadalupe-Hidalgo Treaty (1848), constitute a national minority. Also, Puerto Ricans are in a similar situation since the US invasion and annexation in 1898. However, one wonders why this status is rarely recognized. Contrarily, the most common public representation of US Latinos is as an illegal "alien: an immigrant, a recent arrival, a foreigner not really belonging to, or in, America," even though 60% are native born (Ramirez 2005: 8). This historical misrepresentation has been reinforced since the nineteenth century. In 1878, an amendment to California's State Constitution was proposed. It prohibited the laws and official documents of the State from being published in Spanish, as agreed upon after the Mexican cession. During the debate, Mr. Tinnin, an amendment supporter, argued that such a practice was "for the benefit of foreigners," and another congressman, Mr. Rolfe, asked him: "Do you call the native population of this State foreigners?" "They had ample time to learn the language" was Mr. Tinnin's answer (Crawford 1992: 53). This disingenuous characterization represents US Latinos as foreigners because this way, their demands can be easily manipulated in the public eyes by means of nativist and xenophobic discourses. On the contrary, recognizing their status as a national minority would change the terms of the debate about the legitimacy of Spanish and the right to its maintenance in the United States.
Frequency of contact with monolinguals: for a language minority, contact with monolinguals of both languages, DomL and SubL, is essential. Lacking access to DomL creates a situation which resembles LM but it is only social violence and linguistic marginalization. On the other hand, contact with monolinguals of the minority language creates meaningful contexts for the use of the SubL. In recent years, transnationalism has been identified as a positive force of LM. Transnationalism refers to diasporic groups, notably mainland Puerto Ricans and immigrant Dominicans, leading dual lives across national spaces: "engaging in double consciousness, cultivating dual loyalties, living serially between their islands and the mainland" (Suarez-Orozco and Paez 2002: 6). Language is a central component of transnational identities (Zentella 2004: 189). Furthermore, the Mexican government has taken steps to foster transnationalism among Mexican Americans. The sanction of dual citizenship and absentee-ballot laws promote meaningful participation in Mexican political affairs among Mexican Americans as well as fluid contact with the Mexican culture and most particularly fluency in Spanish (Morris 2003: 149). Dual citizenship will encourage Chicanos to continue revisiting their homeland and cultivating ties there, fostering Spanish proficiency. Thus, transnationalism, as a practice of continuous contact with ethnic ties, has emerged as a positive trend of Spanish maintenance in the United States.
Duration of contact: if voluntary migrants have access to public schools (therefore to the DomL), and economic incentives, making upward mobility possible, LS and LL could be likely (Paulston 1992: 70-71). Many studies have reported this trend: the length of contact with English and attendance in public school has a significant impact on Spanish displacement (Fradd 1996; Porcel 2006; Rusinovich 1990; Veltman 1988; Zurer and McGee 1993). Furthermore, it has been claimed that
630   Spanish in the USA, Heritage Language, L2 Spanish
Language Maintenance 631
Spanish maintenance in the United States is a statistical illusion created by the constant influx of new monolinguals, which in the lapse of 15 years become bilin-guals, and by the third generation, English monolinguals (Fradd 1996; Veltman 1988). In other words, statistics reflect merely a constant renovation of monolinguals that may eventually go through the process of LS.
2.5   Spatial distribution of the speech communities
This is the second group of variables related to the structure of the speech community. Here we will focus on concentration, adjoining minorities, and urbanization.
Concentration: group size is essential for LM, although without concentration, it is of little significance. However, Rusinovich (1990) has shown that group cohesion enhances the effects of concentration. Analyzing census data, she found that Spanish retention was higher where concentration of national groups was high. Thus, Mexican-Americans were more retentive in the southwest (69%) than in Florida (57%) or New York (44%). Similarly, Cubans and Puerto Ricans showed higher retention in Florida (93%) and New York (82%) respectively than in other regions (Rusinovich 1990: 39-41). Also, concentration favors institutional support from democratic governments, which are becoming more sensitive to minority needs (Mackey 2006: 617-618). For example, after the 1975 amendment, Titles II and III of the Voting Rights Act require election material to be provided in an appropriate language "if more than five percent of the voters in an election district [spoke] the same non-English language" (Schmidt 1990: 20-21).
Adjoining minorities: geographical adjacency with monolinguals of the same minority language in different states is a factor that favors LM (Edwards 1992: 38-39). Hudson, Hernandez Chavez, and Bills (1995) analyzed this variable for the 22 largest cities and the 401 counties nearest to the US-Mexican border. Using, among other factors, the measures of language loyalty and retention (see above), they concluded that for cities and counties, proximity to the border favored retention, while greater distance favored shift. Proximity, they explain, increases contact with Spanish monolinguals, relatives, businesses on the Mexican side, and greater interaction with Mexican citizens crossing the border into the United States for business or pleasure. They also found a more positive attitude toward Spanish deriving from community and institutional support, and from the presence of foreign-born Mexicans, more densely concentrated near the border. Similarly, the South Florida area is considered the border with Latin America and the Caribbean. The distribution pattern is similar to that just reported: the high concentration in the south disseminates as we move northward. Latin Americans arrive in Miami all the time for business or leisure. South Florida is called "Mall of the Americas" because it ranks at the top in the sales of apparel and accessories in the United States due to retail sellers from Latin America who come to make their purchases in this area (Fradd 1996: 7; Fradd and Boswell 1996). As a result of a higher frequency of contact with Spanish monolinguals, the value of the language increases, and this potentiates higher proficiency and more positive language attitudes.
Urbanization, isolation, and self-sufficiency: urbanization and industrialization are usually found to balance against minority languages (Thomason 2001:21-22). Cities are the type of setting exerting strong pressures toward cultural assimilation due to demands from the labor market, as well as insertion into more open and diffused social networks in which people have to engage in their daily lives. In contrast, rural dwellers usually enjoy greater isolation and self-sufficiency. Their social networks are dense and limited, and the agricultural labor market poses less stress on English as a job requirement (Fishman 1972a: 97-98). In the past, the urban/rural opposition played a key role among Mexican Americans settled in the southwest due to segregation (Macnab and Christian 1966). However, in the United States, the importance of this factor has weakened as society has moved from early to late capitalism. In the year 2000, only 5% of the US Hispanic labor force held occupations related to farming, fishing and forestry (Ramirez 2005:13). In fact, the recent Hispanic immigration to the United States is a metropolitan phenomenon, concentrated in the largest cities: in 1993 less than 5% of legal immigrants went to live in non-urban areas, whereas more than half settled in just 10 metropolitan locations (Portes and Rumbaut 1996: 43—14). In summary, whereas concentration of Hispanics favors LM, the fact they tend to settle in metropolitan locations weakens the effects of the first.
The last group of variables related to the speech community is language functions, which will be the topic of the next section.
2.6  Language functions
Language functions refer to the effect that purpose, topic, audience, and place of interaction have on the selection of a language variety. In order to study such a relationship, Fishman adopted two concepts: domain and diglossia.
Domain and diglossia: domains are spheres of activity - family, friendship, religion, education, employment, etc. - in which topics of communication, roles, forms of interaction, and locales usually trigger the selection of certain varieties (Fishman 1972a: 82). Diglossia designates the specific situation of functional specialization: that is when language selection is rigidly compartmentalized vis-a-vis domains of interaction. In the first study which used these notions, Greenfield and Fishman (1968: 453-454) presented a series of hypothetical conversations to New York Puerto Ricans, who had to decide on how much Spanish and English they would be likely to use. Conversations were tied to five different domains: family, friendship, religion, education, and employment. They found Spanish was primarily selected for the family, secondarily for friendship and religion, and minimally for education and employment, whereas the reverse pattern emerged for English. This study promoted a line of research concerned with the connection between code selection and domains. Laosa (1975) studied a sample of 295 children in elementary school from three different Hispanic groups: 100 Mexicans residing in Central Texas, 100 Miami Cubans, and 95 New York Puerto Ricans. Mothers and teachers were the informants of children's behavior at the home and school. Three domains were selected: the home, the classroom, and school-recreational activities. The categories for language selection were: English, Spanish, Both, or Mixed Code.
632   Spanish in the USA, Heritage Language, L2 Spanish
Across domains, Mexicans showed no apparent diglossia, whereas Cubans and Puerto Ricans did between family and classroom. He also reported for Puerto Ricans and Cubans that if the use of both languages at home was the norm, English emerged as the primary language at school. Another significant correlation was found for Cubans: if Spanish was the predominant language at home, the use of both languages was selected in the classroom. He also studied if there was a connection between the language parents used at home and the language of the children. In general, children tended to adopt the same language pattern used by adults at home in various social contexts.
Castellanos (1990) studied language selection in public and private domains in Miami in two generations of Cuban Americans. She asked participants what language they normally select when interacting with physicians, tellers, police agents, post office employees, work supervisors, coworkers, college professors, boyfriends/girlfriends, etc. Although certain institutions - such as Education -and certain interlocutors - such as managers at work - favored the use of English, Castellanos did not find a pattern of diglossia but one of transitional bilingualism: the second generation used English, whereas the first used it as a default choice when Spanish was not an option. At home, she reported a typical generational slide pattern: that is, same code within the same generation and different codes across generations, with older generations preferring Spanish and younger cohorts preferring English. Hidalgo (1993) conducted a somewhat similar study in Chula Vista, near San Diego. The population consisted of 136, high school students and 81 parents. Contrary to Castellanos, Hidalgo encountered a divide between the public and the private spheres. As a general pattern, Spanish occupied the highest frequencies of usage in the home, but in public domains - business, health, and government agencies - the use of Spanish was minimal.
Stable diglossia and LM: the concept of diglossia was intended to be not merely descriptive, but predictive as well. Fishman (1972a, 1972b) argued that if a bilingual community attains stable diglossia, the functional compartmentalization of languages would protect the SubL and favor its retention. Contrarily, societal bilingualism without diglossia indicates transition towards LS/LL. In addition, Fishman (1985) considers family and religion - if the language is considered essential to the religious practice - the strongholds of LM. Furthermore, he claims, the reversal of LS is impossible without preserving the level of intimacy at the family and the local community levels (Fishman 1991: 93-95). According to these claims, the New York Puerto Ricans studied by Greenfield and Fishman (1968) and Hidalgo's (1993) Mexican American fit the previous pattern but not Miami Cubans. However, the association between stable diglossia and LM has been challenged. Zentella (1981) questioned the application/interpretation of diglossia, since second generation New York Puerto Ricans used English at home whereas recently arrived Puerto Ricans used Spanish at school. Also Boyd and Latomaa (1999: 304) observe the application of the notion is difficult since the very definition is not clear. Pedraza (1980: 38) and Zentella (1981: 236) share concerns about the rigorous use of the concept in bilingual communities where code-switching becomes a rightful variety (among others) in bilinguals' repertoire. Finally, Eckert (1980) presents another
Language Maintenance 633
point of view. She rejects the stance that diglossia can be a positive factor for LM. Diglossia is not simply a harmless allocation of codes by domains, but a demarcation of power (Meyer-Scotton 2006: 87). Therefore, it poses a serious threat to the community's (self)-image, and the value of its language (Eckert 1980).
2.7  Socio-structural variables
Socio-structural variables encompass social descriptors in which society is organized. Four factors are considered here: demography, socioeconomic stratification, cultural, and legal status.
2.7.1 Demographic status Group size is key to LM. Without speakers there is no use for a language; the larger its size, the more likely the language would be used. Four variables affect the demographic status of a group: absolute number, birth/ death rate, immigration and emigration (Giles, Bourhis, and Taylor 1977: 309), and all of these seem to weigh positively on the side of Spanish maintenance. In 2007, Hispanics reached 45.5 million -15% of the total US population (Bernstein 2008). In 2006, the US Hispanics' fertility rate (2.9 children per family) surpassed all the other groups - allowing the United States to overcome the negative growth affecting industrialized nations (Stein 2007). Latinos are projected to triple, by 2050, and their population share to double (El Nasser 2008 - USA Today). Finally, US Hispanics are young. In 2000, their median age was 26, and about a third was 18 or younger (Ramirez 2005: 4-5).
Fertility: taking a closer look, the positive impact of fertility and age are questionable. There is evidence that LS increases with the number of generations, and that by the third generation LS is either completed or near completion (but see Villa and Rivera-Mills 2009: 2). Also, a positive correlation has been reported for LM and age: that is LM increases with age, and decreases among younger cohorts. These correlations bring three negative variables into the equation: schooling in the United States, length of contact, and age of contact: the higher the years of schooling in the United States and the years of US residence, as well as the lower the age of contact with English, the higher the likelihood of LS.
Immigration: the increasing levels of Hispanic immigration since 1965 to the present represents a positive factor in terms of LM. However, several scholars have questioned its impact in the long term (Fradd 1996; Garcia and Otheguy 1988:186; Silva-Corvalan 2004: 212-213; Sole 1982: 267). Veltman (1988: iv) claims that
any interruption in the immigrant stream would stabilize the size of the Spanish-speaking population for approximately 15 years, after which a progressively more rapid decline would set in.
Two issues are at stake in this debate: Does immigration create a statistical illusion of LM? Is the sole impact of immigration capable of reversing LS? Contrary to the pessimistic answer, optimists argue the impact of the immigration flow is
634   Spanish in the USA, Heritage Language, L2 Spanish
Language Maintenance 635
overwhelming, and its massive growth has multilevel repercussions (Martinez 2009; Roca 1991). Politically, Latinos have become essential to winning local and national elections. Economically, the Latino market has become considerably important to the United States and Latin American economies (Carreira 2002; Villa 2000). As a result, communal and governmental institutional support is growing. In summary, whereas the pessimistic assessment projects a recurrent pattern, optimists believe there is a critical point at which the quantitative impact becomes qualitative, and thereafter the previous pattern no longer applies.
2.7.2 Socioeconomic status Socioeconomic status "refers to the degree of control a language group has gained over the economic life of its nation" (Giles, Bourhis, and Taylor 1977:310). In 2000, compared to the total population, Hispanics had the lowest probability of completing high school or college and of being employed (Ramirez 2005: 11-12). Their average annual household income was $34,397 USD, in contrast to $50,046 USD for the total population. Also, 22.6% lived at or below the poverty line, compared to 12.4% of the total population. As these indicators show, the socioeconomic status of US Latinos is one of discouraging disadvantage. However, what is the connection between these indicators and LM? Does Spanish retention have a positive or negative effect on the socioeconomic status of the speakers?
Stratification of Spanish speakers: using data from the 1980 and 1990 decennial census for the southwest, Hudson, Hernandez, and Bills (1995: 179) found that retention correlated negatively with educational attainment: it was highest among the most poorly educated and lowest among the most highly educated. Also, LM was favored where a higher proportion of Hispanics lived at or below the poverty level. Finally, it negatively correlated to economic success: where the per capita income of the population of Spanish origin was highest, the rate of Spanish retention was lowest and vice-versa. Using recent data, McCullough and Jenkins (2005) and Jenkins (2009: 23) applied the methodology from the previous study and found that previously strong correlations between Spanish LM and socioeconomic factors had weakened. Therefore, new data might reveal a positive trend.
LS and economic success: the myth that English monolingualism is a key factor for economic success in America has also been refuted. Ofelia Garcia studied the issue and concluded that although the rate of LS among Mexicans and Puerto Ricans was higher than among Cubans, the latter had higher incomes (Garcia 1995: 147). She also found that English monolingualism did not account for economic success among Hispanics and, furthermore, bilingualism rather than monolingualism was a useful commodity in some communities. Also, Fradd and Boswell (1999) conducted a study of ten cities highly impacted by immigration: Miami, Los Angeles, San Diego, San Francisco, San Antonio, Houston, El Paso, New York City, Jersey City, and Chicago. At the state level, only Florida showed an economic advantage for Spanish bilinguals. At the metropolitan level, in seven cities English-monolingual Hispanics earned more than those who were highly proficient in English and also knew Spanish, but in three cities Hispanics bilinguals had
higher incomes than English monolinguals. In general, where bilingualism was important for conducting business, Hispanic bilinguals received higher salaries. Furthermore, in three metropolitan areas the percentage of Hispanics below the poverty level was smaller for those using English and Spanish than for the English-only cohort, and in three areas large cohorts of English monolingual Latinos only had incomes below the poverty level (Fradd and Boswell 1999: 10). Therefore, English monolingualism does not account for economic success, and bilingualism does where the language has instrumental value.
Immigration and instrumental value of Spanish: Grin (1999: 171) points out an elementary market driven mechanism: if the demand for a particular language increases, this will tend to drive up the value of these skills. Once it impacts on personal income, more people will have an incentive to maintain and acquire the language. This instrumental trend is visible in various aspects of everyday life in the United States. The estimated buying power of the Hispanic market is near one trillion dollars. To attract this clientele mostly comprised of recent immigrants - but in general Hispanics who continue to regularly use Spanish - businesses have been increasingly offering bilingual customer services. This trend has driven up the demand for bilinguals, and it has had obvious effects in the education domain, where Spanish has become the most taught FL in the United States.
Class values and LM: how does socioeconomic stratum affect LM? One frequently finds cases of parents who (un)deliberately block intergenerational transmission. They are usually victims of social, educational, and /or occupational disadvantage and also victims of myths such as the one discussed above (Dorian 2006). Lambert and Taylor (1996) studied the issue of middle- and working-class orientation toward LM among Miami Cuban mothers. They observed that middle-class mothers fostered additive bilingualism, implementing intergenerational transmission at home and leaving the promotion of English to school and other American institutions. On the contrary, working-class mothers encouraged subtractive bilingualism, emphasizing the development of English to guarantee the academic and later economic success of their children. Based on oral interviews, Lambert and Taylor (1996: 496^498) traced a connection between class (dis)advantage and values, and LM. Middle-class mothers succeeded economically; therefore their home language did not affect their achievements, whereas working-class mothers could blame their language for their lack of success. Also, due to their social stratum, working class parents have fewer resources to foster the development of both languages and also spend more time at work than middle class parents (Tuominen 1999: 73). In summary, the studies commented on in this section have shown that English monolingualism does not account for structural integration in America; bilingualism cannot be related to socioeconomic disadvantage, but monolingualism can be; finally, class values and experiences can determine different orientations toward LM.
2.7.3 Cultural status The status of a minority culture depends on their institutional support (I-support), that is their capacity to organize themselves into pressure groups to safeguard their interests (Giles, Bourhis, and Taylor 1977: 316). I-support can be formal (media, education, government) or informal (industry,
636   Spanish in the USA, Heritage Language, L2 Spanish
Language Maintenance 637
religion, cultural institutions), and it is essential for minorities to be well represented in both (Giles, Bourhis, and Taylor 1977:316). Interestingly, the word represented has two readings:Itcanbe understood as being part of or playing a role in these institutions, or it can be read as being well depicted or portrayed by those institutions. Both meanings are important in the cultural sphere.
The Miami-Dade ethnic enclave: this is an example of being well represented in the first reading. An ethnic enclave is a "distinctive economic formation, characterized by the spatial concentration of immigrants that has organized a variety of enterprises to serve their own ethnic market and the general population" (Portes and Bach 1985: 203). A key feature of Miami's enclave is its institutional completeness; it allows Hispanics in the area to live their lives within the social networks of their ethnic community (Perez 1992: 95). In this milieu, recent immigrants have less pressure to learn English and the ubiquitous presence of their language and culture eases their incorporation into America. In Miami, Hispanics are not only busboys at upscale restaurants or carwash employees but also the patrons. This social diversification collapses the easy associations between ethnicity and social stratum, Spanish LM and school drop out, etc. But how does strong I-support enhance the cultural status of a community? Recently, antibilin-gual policies have been implemented in California and Arizona. In contradistinction, the Miami-Dade County School Board - with a Cuban American majority - has given strong support to bilingual education and Spanish has been forcefully promoted through elementary school for every child (Stepick, Grenier, Castro, and Dunn 2003). Also, the Florida Department of Education, the University of Miami, and the University of Florida are working on the creation of a global-multilingual workforce for the twenty-first century to fulfill the increasing demand of Spanish-English bilinguals capable of conducting business with Latin America (Boswell 2000). Also, the enclave protects Hispanics from discrimination. The Cuban American organization SALAD (Spanish American League Against Discrimination), backed by spontaneous citizen action, has been at the forefront of the fight against discrimination. After English was declared as the official language of the State, a cashier at Publix supermarket was terminated for speaking Spanish at work. The Hispanic reaction was so forceful that in less than 48 hours the employee was reinstated, the manager transferred out of Dade County, and the store issued a public apology to their Spanish-speaking customers. Therefore, a strong institutional support can foster LM and partially restore some of the value and self-esteem that the language, the culture, and the community have lost in the process of minoritization.
Hispanic media: I-support is also provided by the way the minority is portrayed by the media, government statistics, and the educational system. Zentella (1997: 3) has denounced how the ways Puerto Ricans live, love, and raise their children has been misrepresented by the national statistics. Perea (1998: 583) has deplored the Latino invisibility created by the lack of a positive public image, the absence of public recognition, the prohibitions on the use of Spanish, the attributions of foreignness, laziness, criminality, machismo, ignorance, etc. Many scholars have argued the importance of the Spanish media based on numbers: TV channels,
radio stations, newspapers, etc. These studies are important but only reflect the first sense of representation. However, fewer studies have looked into the content. Garcia et al. (1985: 347-349) carried out a comparative analysis of the French, German, Spanish and Yiddish press in the United States and concluded that the Hispanic press showed the least interest in the topic of the mother tongue, and more than 50% of the references to the language and the ethnic community were negative. Also, Zentella (2002: 181) has questioned to what extent
i the Spanish media has helped to preserve Hispanic traditions and values, and
facilitate social integration into the United States culture. She also wonders whether it is a big business that sells self-hate. She observed a linguistic and economic stratification in the Spanish electronic media the most accessible and economic media, the radio, incorporates the local Spanish. As the media becomes more expensive, such as TV and cable/satellite, the Spanish variety becomes more distant as well as the phenotype of news anchors and ads' characters, whites and blond, because "white sells." Therefore, Zentella (2002: 185-86) concluded, the Spanish media reproduces the racism endemic in the community
< and cannot be a positive factor in LM.
2.7.4   Legal status  This is the last structural factor to be discussed. Here, we will , concentrate on the official status of English and the US policy toward language
minorities.
Official language in the United States: at the federal level, the United States has no official language. However, most Americans believed English had official status until recently (Ferguson and Heath 1981), when the English Only Movement resur-
1 faced. It is not surprising that English enjoys a de facto official status, by being the
language privileged in government functions, and education. Contrary to their lack of success at the federal level, English Only has made successful inroads at the State level. Not counting Hawaii, officially bilingual, 27 states have made English
i official. In 1997 the Congressional Research Service prepared a report for the US
Congress in which they concluded the declaration of English as the official language of the United States "would be a largely symbolic act of negligible legal effect." (Aleman, Andorra, and Dale 1997: 3). However, symbolic should not be considered harmless, or its effects negligible; as any reassurance of power, this one has a constitutive symbolic dimension, affecting language attitudes and behaviors. The example of the Publix supermarket in Miami after the passage of Amendment 11, encouraged discrimination; and this case was not an isolated one around that time. At the same time, language empowerment paves the way for legal actions: in 1999 the US Supreme Court ruled that an Arizona law requiring public employees to speak English only at work was unconstitutional. Proposition 187 has been more harmful in limiting the benefits to undocumented migrants in California as well as the state's efforts to dismantle bilingual education (Morris 2003:153).
Promotion and tolerance rights: promotion rights require the state to be actively committed to the protection of a certain language or languages, and this is the type of protection national that minorities should receive (Kloss 1977). Since this
638   Spanish in the USA, Heritage Language, L2 Spanish
Language Maintenance 639
status is denied to any US Hispanic group, those rights are not even discussed. Tolerance rights prestippose individuals will use whichever language they want (in their homes, the institutions of the civil society, the workplace, etc.) without state interference (Patten and Kymlicka 2003). There is a general consensus that the United States has followed this laissez-passer approach with immigrant minorities. However, from the perspective of LM, this policy is far from neutral. On one hand, this policy works in a context where bilingualism is acceptable only as a transition from one monolingualism to another (Grosjean 1982: 62). On the other, laissez-passer favors the DomL, which gains control of high functions, and relegates the SubL to the private sphere, where it eventually fades away (Laponce 2003: 59).
2.8  Language attitudes
"Attitudes are enduring mental representations of a person, place, or thing that evokes an emotional response and related behavior" (Rathus 1999: 636). Since they are socially learned, they are susceptible to change. Typically, language attitudes surface as sentiments of loyalty or antipathy, and manifest through behaviors such as language interventions, that is acts of "control or regulation of habitual language use via reinforcement, planning, prohibition, etc" (Fishman 1972a: 104, 111). Language attitudes are mediated by the ideological system, and positively correlate with attitudes toward the speakers of the language along the favorable/ unfavorable continuum. Thus, there is a strong agreement that although the immediate object may be the language, the real target is the minority. Finally, linguistic attitudes are the cornerstone of LMLS, capable of changing a dynamic into its opposite in the face of positive or adverse conditions.
Hispanics' attitudes toward English: in spite of the counter-examples we provided above, usually, Hispanics associate socioeconomic mobility and full participation in the American society with high English proficiency (Silva-Corvalan 2004: 226). English is the language of the colonizer, the educated, and the powerful in the United States, and parents and students want more and better instruction in it (Zentella 1981: 221-222). Garcia, Morin, and Rivera (2001: 51, 56) point out an insightful perspective for New York Puerto Ricans; in addition to English being the language of the powerful, proficiency in it or even the ability to intersperse features of it, is a mark of in-group status; it legitimates its users in the eyes of recent arrivals as connoisseurs of the city and the American life. Also, English is the language of the stigmatized African Americans with whom they share life and community. Therefore, English enjoys both overt and covert prestige: as a language of socioeconomic success, as a mark of status, and as a neighborhood language, it pushes bilinguals toward English monolingualism.
Anglos' attitudes toward Spanish: US whites have long associated Spanish with racial impurity and negative stereotypes (Silva-Corvalan 2004: 226; Urciouli 1996). In 1898, the US office of the school system in Puerto Rico, reported to the US government: "Their language is a patois almost unintelligible to the native of Barcelona and Madrid. It possess no literature and has little value as an intellectual medium"
(Language Policy Task Force 1992: 64). In 1852, President Fillmore argued that the acquisition of Cuba would be welcomed if ceded by Spain; otherwise, he considered the purchase a hazardous measure, "it would bring a population of a different national stock, speaking a different language, and not likely to harmonize with other members" (Urciouli 1996:19). Even loanwords from Spanish are loaded with negative attitudes. The use of mock Spanish reproduces and reinforces negative stereotypes in a way that most Americans consider innocent and clever (Hill 1995).
i Finally, the use of a minority language in public when unscripted or not framed by
middle class legitimization is seen as dangerously out of order (Urciouli 1996). It is commonplace in sociolinguistic literature to find narratives of language interventions to stop the use of Spanish in public.
Hispanic attitudes toward Spanish: they are mixed. In general, studies report positive attitudes (Lopez Morales 2000; Porcel 2006; Silva-Corvalan 2004; Zentella 1997). However, they also perceive a disconnection between verbal declarations and actions of intergenerational transmission. Also, among US-born Hispanics, the language seems to lack affective value; the orientation toward LM is not the preservation of traditions or integration, but the general, instrumental notion of knowing another language (Porcel forthcoming; Silva-Corvalan 2004: 222-224).
LMLS are acts of identity: as language practices, LMLS are behaviors used by speakers to resemble the groups "with which from time to time [one] wishes to be
i identified" (La Page and Tabouret-Keller 1985: 181). From this standpoint, the
practice of code selection, which fosters LMLS, constitutes an act of identity. Porcel (forthcoming) observes that whereas language pride emerges as intellectualized rationalizations, language shame evokes profound feelings. In general, the immigrant languages - even world languages "such as Spanish (as is the case in the United States)" - are despised as Romaine (2006: 396-397) observes. Ferguson and Heath (1981: xxvii-xxviii) claim that "[mjany Americans regard the use of another language in the USA as a sign of inferiority and disadvantage - to be kept hidden in one's case and educated away in the case of others." Therefore, a combination
i of cultural shame and assimilation pressures is also an important factor in US
Spanish maintenance (Porcel forthcoming). In the lobby of an upscale hotel a middle age Cuban American couple had this experience:
si, por ejemplo, de:: eh:: yo he ehtado por ejemplo en / en la puerta de un hotel en west ' palm beach, con mi ehposa, que ehtaba tomando un cuhso y eh:: ehtabamoh en el loby
del / del / del hotel y ehtabamoh conversando en ehpanol; paso u::/ una pareja de / de nohteamericanoh, obviamente sureno eh:: y noh dijeron asi, literalmente a la cara "fuck you." 2
i
Billy, born in the United States to Cuban American parents, was 20 at the time of the interview:
ehtabamoh en un rehtauran ahi, tu sa'e, y:: mi padre n::/ sa'e ingleh pero no muy t muy bien, con mi abuelo tambien, 'tonce ohdenaron ahi comi'a y:: no se la pehsona
ahi al la'o de losotro [nosotros-JP] loh [nos-JP] vieron un poco ehtrano asi, y eso eh lo que yo vi, tu sa'e.3
640  Spanish in the USA, Heritage Language, L2 Spanish
Language Maintenance 641
Carl, 19 years old at the time of the interview, arrived in Miami when he was 12 and attended a public school in Hialeah where the overwhelming majority of students, staff, and faculties were Hispanics.
lo que paso fue que se ehtaban buhlando de / de / yo ehcribi aggo en ingleh, un par-rafo y lo tu/ el profesor me dijo que lo tenia que leel en frente de la clase, entonce fui, y lo lei, y yo ehtaba leyendolo, pero muy mal entonce la gente buhlandose, entonce / pero no impohta, yo queria cumplir con mi eso:: y coger mi grado, mi tarea y hacer mi / coger mi grado ya, que era lo que impohtaba; pero bueno, cuando tehmine, me sente y to' el mundo siguio con aggunah palabrah y buhlandose entonce, no se eh:: tuve el problema con / con con el muchacho 4
The three narratives just presented open a window to language attitudes, conflict, and identity shame in the United States. Being silenced in public or bullied in the classroom are bad enough, but the perception of someone laughing at your father's and grandfather's expense is even far more painful. The paralanguage accompanying these narratives exude anger and shame. It is this deep connection with strong emotions that makes language attitudes so powerful and the future of LMLS almost unpredictable.
3   Concluding remarks
There is no question that the dynamics of LM within the US Latino community are complex and blurry, as the previous description should have made clear. Underlying such a situation is the fact that the term Hispanic - or Latino - represents an almost impossible abstraction. If history, place of settlement, time of arrival, phenotypes, citizenship and socioeconomic status conspire against the idea of a single, coherent Mexican American community (Valdes 1988), the same notion of a single, coherent community applied to US Latinos only could result in an almost fictional abstraction. Thus, the valid attempt of achieving an overall picture for the whole community has to pay that price. As a result, the image we obtained is one of different tendencies pulling in different directions. There is also another reason. The mere fact that LMLS are ongoing processes, whose directionality can change at any given moment, implies that we are trying to capture an unstable equilibrium of forces. In effect, the balance of power is crucial to the dynamics of LMLS. As we argued, subordinate groups are at a major disadvantage with respect to dominant groups. However, the weight of the variables is not set; it can change a given status quo. This is the effect that group size, coupled with concentration, seems to have in the current dynamics of LMLS. Stemming from this, Latino economic power has exponentially grown in the last two decades. To gain a share of this market, the private sector has implemented a series of strategies, which indirectly support Spanish LM. If the objective factors are important, attitudinal variables do not play a lesser role in the dynamics of LMLS. As claimed above, these processes take root at individual practices, since it is the individual who makes the conscious or subconscious decision to keep, change, or abandon altogether the use of a language in certain or every situation(s).
By doing this, the individual performs acts of identity. The impact of language attitudes to LMLS can only increase the uncertainty of LMLS assessments, since attitudes are susceptible to change in any direction under the pressure of ideological discourses and positive or negative behaviors from other competing communities.
NOTES
1 I want to thank the following colleagues who read the original manuscript for their time and suggestions: the editor, Professor Manuel Diaz-Campos, and Professors Ed Brent, Jerry Cohen, Tanya Flores, Joe Salmons and Don Tuten. I also appreciate the help of the following students: Phillip McGee, Kyle Smith, and Travis Tackett, who helped me with the manuscript. As it is customary to say, all final decisions were my choices, as well as the mistakes.
2 yes, for example, I have been at a hotel door in West Palm Beach with my wife - I was taking a course - and we were in the hotel's lobby and we were speaking in Spanish. A couple passed by, an American couple, obviously from the south, and they said to to us, literally in our face "fuck you" (Otto 4.36,188).
3 we were at a restaurant, you know, and my father *don't know English very well, my grandfather was there too; then they ordered the food .and the people next to us looked at them in a strange manner, you know, and that's what I saw (Billy 1.24,114).
4 they were laughing at me because I wrote something in English, a paragraph, and the professor told me I had to read it in front of the class; so I went and read it but very badly and they were laughing, but it didn't matter to me, I just wanted to be done with it, get my grade; that was what mattered to me; but when I finished and got back to my seat, they kept laughing, and saying things to me and then I ended up having a fight with the guy (César 3.45,148).
REFERENCES
Bernstein, Robert. 2008. US Hispanic population surpasses 45 million. Now 15 percent of total. (US Census Press Release). Available at: http://www. census. gov / newsroom / releases / archives/population/cb08-67.html (last accessed March 31, 2010).
Bills, Garland D., Eduardo Hernandez Chavez, and Allan Hudson. 1990. The geography of language shift: distance from the Mexican border and Spanish language claiming in the Southwestern US. International Journal of the Sociology of Language 114. 9-27.
Boswell, Thomas D. 2000. Demographic changes in Florida and their importance
for effective educational policies and practices In Ana Roca (ed.), Research on Spanish in the United States, 406-431. Somerville, MA: Cascadilla Press.
Boyd, Sally and Sirkku Latomaa. 1999. Fishman's theory of diglossia and bilingualism in the light of language maintenance and shift in the Nordic Region. In Guus Extra and Ludo Verhoeven (eds), Bilingualism and migration, 303-324. Berlin: Mouton de Gruyter.
Brezinger, Matthias. 1997. Language contact and language displacement. In Florian Coulmas (ed.), The handbook of sociolinguistics, 273-284. Oxford: Blackwell.
642   Spanish in the USA, Heritage Language, L2 Spanish
Carreira, Maria M. 2002. The media, marketing and critical mass: portents of linguistic maintenance. Southwest Journal of Linguistics 21. 37-54.
Castellanos, Isabel. 1990. The use of English and Spanish among Cubans in Miami. Cuban Studies/Estudios Cubanos 20. 49-63.
Crawford, James. 1992. Spanish language rights in California: constitutional debates. In James Crawford (ed.), Language loyalties. A source book on the Official English controversy, 51-58. Chicago: The University of Chicago Press.
Dorian, Nancy. 2006. Minority and endangered languages. In Tej Bhatia and William C. Ritchie (eds), The handbook of bilingualism, 437-459. Oxford: Wiley-Blackwell.
Eckert, Penelope. 1980. Diglossia: separate and unequal. Lingusitics 18. 1053-1064.
Edwards, John. 1992. Sociopolitical aspects of language maintenance and loss: towards a typology of minority language situations. In Willem Fase, Koen Jaspaer, and Sjaak Kroon (eds), Maintenance and loss of minority languages, 37-54. Amsterdam: John Benjamins. El Nasser, Haya. 2008. US Hispanic population to triple by 2050. USA Today. Available at: www.usatoday.com/ news/nation/2008-02-ll-population-study_N.htm (last accessed March 31, 2010).
Fasold, Ralph. 1987. The sociolinguistics of
society. Oxford: Blackwell. Ferguson, Charles A. and Shirley Brice
Heath. 1981. Introduction. In Charles A.
Ferguson and Shirley Brice Heath
(eds). Language in the USA, xxv-xxxviii.
New York: Cambridge University Press. Fishman, Joshua A. 1972a. Tlie sociology of
language. Rowley, MA: Newbury House
Publishers. Fishman, Joshua A. 1972b. Societal
bilingualism: stable and transitional.
Language in sociocultural change, 135-152.
Stanford, CA: Stanford University Press. Fishman, Joshua A. 1985. Bilingualism and
biculturalism as individual and as
societal phenomena. In The rise and the fall of the ethnic revival, 39-56. Berlin: Mouton de Gruyter. Fishman, Joshua A. 1991. Reversing language
shift. Clevendon: Multilingual Matters. Fradd, Sandra H. 1996. The economic impact of Spanish-language proficiency in metropolitan Miami. Miami: The Greater Miami Chamber of Commerce and the Policy Center of the Cuban American National Council. Fradd, Sandra H. and Thomas D. Boswell. 1996. Spanish as an economic resource in Metropolitan Miami. The Bilingual Research Journal 20 (2). 283-337. Fradd, Sandra H. and Thomas D. Boswell. 1999. Income patterns of bilinguals and English only Hispanics in selected metropolitan areas. In Sandra H. Fradd (ed.), Creating Florida's multilingual global work force: educational policies and practices for students learning new languages, 1-12. Tallahassee, FL: Florida Department of Education. Gal, Susan. 2007. Multilingualism. In Louise Mullany, Carmen Llamas, and Peter Stockwell (eds), The Routledge companion to sociolinguistics, 149-156. London; New York: Routledge. Garcia, Ofelia. 1995. Spanish language loss as a determinant of income among Latinos in the United States: implications for language policy in schools. In James W. Tollefson (ed.), Language policy and language education, 142-160. New York: Cambridge: Cambridge University Press. Garcia, Ofelia, Joshua A. Fishman, Silvia Burinat, and Michael H. Gertner. 1985. The Hispanic press in the United States: content and prospects. In Joshua A. Fishman (ed.), The rise and fall of the ethnic revival, 343-362. Berlin: Mouton de Gruyter.
Garcia, Ofelia, Jose L. Morin, and Klaudia M. Rivera. 2001. How threatened is the Spanish of New York Puerto Ricans? In Joshua A. Fishman (ed.), Can threatened languages be saved?, i4~73. Clevedon, UK: Multilingual Matters.
Language Maintenance 643
Garcia, Ofelia and Ricardo Otheguy. 1988. The language situation of Cuban Americans. In Sandra Lee Mckay and Sau-ling Cynthia Wong (eds), Language diversity, problem or resource? 166-192. Boston: Heinle & Heinle Publishers.
Giles, Howard, R.Y Bourhis, and D.M. Taylor. 1977. Towards a theory of language in ethnic group relations. In Howard Giles (ed.), Language, ethnicity, and intergroup relations, 307-348. New York: Academic Press.
Greenfield, Lawrence and Joshua A. Fishman. 1968. Situational measures of language use in relation to person, place, and topic among Puerto Ricans bilinguals. In Joshua A. Fishman, Robert L. Cooper, Roxana Ma, et al. (eds), Bilingualism in the barrio, 430-458. New York: US Department of Health, Education and Welfare.
Grin, Francois. 1999. Market forces, language spread and linguistic diversity. In Miklos Kontra, Robert Phillipson, Tove Skutnabb-Kangas and Tibor Varady (eds), Language: a right and a resource, 169-186. Budapest: Central European University Press.
Grosjean, F. 1982. Life with two languages: an introduction to bilingualism. Cambridge, MA: Harvard University Press.
Hidalgo, Margarita. 1993. The dialectics of Spanish loyalty and maintenance in the US-Mexico border: a two generation study. In Ana Roca and John Lipsky (eds), Spanish in the United States, i7-7i. Berlin; New York: Mouton de Gruyter.
Hill, Jane H. 1995. Mock Spanish. Language and culture: Symposium 2. Available at: http://web.archive.org/ web/20010711193457rn_l/www. language-culture.org/colloquia/ symposia/ (last accessed January 4, 2010).
Hudson, Allan, Eduardo Hernandez Chavez, and Garland D. Bills. 1995. The many faces of language maintenance: Spanish language claiming in five Southwestern states. In Carmen Silva-
Corvalan (ed.), Spanish in four continents, studies in language contact and bilingualism, 148-164. Washington, DC: Georgetown University Press.
Jenkins, Devin L. 2009. The costs of
language loyalty: socioeconomic factors in the face of shifting demographic trends among Spanish speakers in the Southwest. Spanish in Context 6. 7-25.
Kloss, Heinz. 1977. The American bilingual tradition. Rowley, MA: Newbury House.
Kymlicka, Will and Alan Patten. 2003. Language rights and political theory. Oxford: Oxford University Press.
Lambert, Wallace E. and Donald M. Taylor. 1996. Language in the lives of ethnic minorities: Cuban American families in Miami. Applied Linguistics 17. 477-500.
Laponce, Jean. 2003. Babel and the market: geostrategies for minority languages. In Michael A. Morris and Jacques Maurais (eds), Languages in a globalising world, 58-63. Cambridge: Cambridge University Press.
Le Page, R.B. and Tabouret-Keller, A. 1985. Acts of identity: creole-based approaches to ethnicity and language. Cambridge: Cambridge University Press.
Lopez Morales, Humberto. 2000. Sociolinguistica. Madrid: Gredos.
Macias, Reynaldo. 2000. The flowering of America: linguistic diversity in the United States. In Sandra Lee McKay and Sau-Ling Cynthia Wong (eds), Neiv immigrants in the United States, 11-57. New York: Cambridge University Press.
Mackey, William F. 2006. Bilingualism in North America. In Tej K. Bhatia and William C. Ritchie (eds), The handbook of bilingualism, 607-641. Oxford: Wiley-Blackwell.
Macnab, Jane and Chester C. Christian. 1966. Spanish language and culture in the southwest. In Joshua A. Fishman (ed.), Language loyalty in the United States. The maintenance and perpetuation of non-English mother tongues by American ethnic and religious groups, 280-317. London, The Hague and Paris: Mouton & Co.
i
644   Spanish in the USA, Heritage Language, L2 Spanish
Language Maintenance 645
Martinez, Glenn A. 2009. Hacia una sociolinguistica de la esperanza. Spanish in Context 6.127-137.
McCullough, Robert E. and Devin L. Jenkins. 2005. Out with the old, in with the new? Recent trends in Spanish language use in Colorado. Southwest Journal of Linguistics 24. 91-110.
Morris, Michael A. 2003. Effects of North American integration on language diversity. In Michael A. Morris and Jacques Maurais (eds), Languages in a globalising world, 143-156. Cambridge: Cambridge University Press.
Meyer-Scotton, Carol. 2006. Multiple voices. An introduction to bilingualism. Oxford: Wiley-Blackwell.
Paulston, Christina B. 1994. Linguistic minorities in midtilingual settings. Amsterdam: John Benjamins.
Pedraza, Pedro. 1980. Rethinking diglossia. New York: Language Policy Task Force, Centro de Estudios Puertorriquefios, CUNY.
Perea, Juan. 1998. Death by English. In Richard Delgado and Jean Stefancic (eds), The Latino condition. A Critical Reader, 583-595. New York: New York University Press.
Perez, Lisandro. 1992. Cuban Miami. In Guillermo Grenier and Alex Stepick (eds), Miami Now! Immigration, ethnicity, and social change, Gainesville, FL: University of Florida Press.
Porcel, Jorge. 2006. The paradox of Spanish among Miami Cubans, journal of Sociolinguistics 10. 93-110.
Porcel, Jorge. Forthcoming. Language discrimination among Miami Cubans.
Portes, Alejandro and Robert L. Bach. 1985. Latino journey: Cuban and Mexican immigrants in the United States. Berkeley, CA: University of California Press.
Ramirez, Roberto. 2005. Nosotros: hispanos en los Estados Unidos. Available at: www. census.gov/prod/2005pubs/censr-18sp. pdf (last accessed March 31, 2010).
Rathus, Spencer A. 2000. Psychology: the core. Fort Worth, TX: Harcourt College Publishers.
Roca, Ana. 1991. Language maintenance and language shift in the Cuban American community of Miami: the 1990s and beyond. In David F. Marshall (ed.), Language planning: Focusschrift in honor of Joshua A. Fishman, 245-257. Amsterdam: John Benjamins.
Romaine, Suzanne. 2006. The bilingual and multilingual community. In Tej K. Bhatia and William C. Ritchie (eds), The handbook of bilingualism, 385-105. Oxford: Wiley-Blackwell.
Rusinovich, Yolanda. 1990. Bilingualism: stable or transitional? The case of Spanish in the United States. International Journal of the Sociology of Language 84. 35-80.
Safran, William. 1999. Nationalism. In Joshua A. Fishman (ed.), Handbook of language and ethnic identity, 77-93. Oxford: Oxford University Press.
Schmidt, Ronald. 1990. Language policy and identity in the United States. Philadelphia: Temple University Press.
Siguan, Miquel. 2001. Bilinguismo y lenguas en contacto. Madrid: Alianza Editorial.
Silva-Corvalan, Carmen. 2004. Spanish in the Southwest. In Edward Finegan and John R. Rickford (eds), Language in the USA, 205-229. Cambridge: Cambridge University Press.
Sole, Carlos A. 1982. Language loyalty and language attitudes among Cuban Americans. In G.D. Keller and Joshua A. Fishman (eds), Bilingual education for Hispanic students in the United Statesr 254-268. New York: Teachers College Press, Columbia University.
Stein, Rob. 2007. US fertility rate hits 35-year high, stabilizing population. Washington Post. Available at: www. washingtonpost.com/wp-dyn/content/ article/2007/12/20/AR2007122002725. html (last accessed March 31, 2010).
Stepick, Alex, Guillermo Grenier, Max Castro, and Marvin Dunn. 2003. This land is our land. Berkeley, CA: University of California Press.
Suarez-Orozco, Marcelo M. and Mariela M. Paez. 2002. Introduction. In Marcelo
M. Suarez-Orozco and Mariela M. Paez (eds). Latinos remaking America, 1-37. Berkeley, CA: University of California Press.
Thomason, Sarah G. 2001. Language contact. An Introduction. Washington, DC: Georgetown University Press.
Tuominen, Anne. 1999. Who decides the home language? A look at multilingual families. International Journal of the Sociology of Language 140.59-76.
Urciuoli, Bonnie. 1996. Exposing prejudice. Puerto Rican experiences of language, race, and class. Boulder, CO: Westview Press.
Valdes, Guadalupe. 1988. The language situation of Mexican Americans. In Sandra Lee Mckay and Sau-ling Cynthia Wong (eds), Language diversity. Problem or resource? 111-139. Boston: Heinle & Heinle Publishers.
Veltman, Calvin. 1988. The future of the Spanish language in the United States. New York: Hispanic Policy Development Project.
Veltman, Calvin. 2000. The American linguistic mosaic: understanding language shift in the United States. In Sandra Lee McKay and Sau-Ling Cynthia Wong (eds), New immigrants in the United States, 58-93. New York: Cambridge Univesity Press.
FURTHER READING
Villa, Daniel. 2000. Languages have armies, and economies too: the presence of US Spanish in the Spanish speaking world. Southwest Journal of Linguistics 19.143-154.
Villa, Daniel and Susana V Rivera Mills. 2009. Spanish maintenance and loss in the US Southwest: history in the making. Spanish in Context 6. 7-25.
Zentella, Ana C. 1981. Language variety among Puerto Ricans. In Charles Ferguson and Shirley B. Heath (eds), Language in the USA, 218-238. New York: Cambridge University Press.
Zentella, Ana C. 1997. Growing up bilingual. Oxford: Blackwell.
Zentella, Ana C. 2002. Spanish in New York. In Garcia Ofelia and Joshua A. Fishman (eds), The multilingual apple, 167-202. Berlin: Mouton de Gruyter.
Zentella, Ana C. 2004. Spanish in the Northeast. In Edward Finegan and John R. Rickford (eds), Language in the USA, 182-204. Cambridge: Cambridge University Press.
Zurer, Barbara and Arlene McGee. 1993. Language choice in Hispanic-background junior high school students in Miami: a 1988 update. In John Lipsky and Ana Roca (eds), Spanish in the United States. Linguistic contact and diversity, 91-102. Berlin: Mouton de Gruyter.
Aleman, Steven R., Bruno Andorra, and Charles V. Dale. 1997. English as the official language of the United States: an overview. Washington, DC: Congressional Research Service, the Library of Congress.
Henslin, James M. 2001. Sociology. Boston: Allyn and Bacon.
Laosa, Luis. 1975. Bilingualism in three United States Hispanic groups: contextual use of language by children
and adults in their family. Journal of Educational Psychology 67. 617-627.
Mackey, William F. 2003. Forecasting the fate of languages. In Jaques Maurais and Michael A. Morris (eds), Language in a globalising world, 64-81. Cambridge: Cambridge University Press.
Portes, Alejandro and Ruben G. Rumbaut. 1996. Immigrant America. A portrait. Berkeley, CA: University of California Press.
i