Home / Articles / Volume 18 (2020) / Identity and Empowerment: Vernacular English Features used by Bilingual Mexicans online
Document Actions



This study examines why young Mexican bilinguals use vernacular language traits in online social media. Using an ethnolinguistic approach, I conducted interviews and employed discourse analysis of the Facebook feeds of members of a bilingual network, in which they make use of vernacular language varieties typically attributed to African American speakers. Findings show that at least one young woman uses these vernacular English features to support feminism and present herself as equal to men. This rhetorically motivated use forges a sense of belonging outside a particularly entrenched category of woman (e.g., old fashioned) and challenges the hierarchy and language ideologies imposed by men and older women. I argue that Mexican bilinguals (male or female) who use such vernacular features are not identifying as African American or as part of any specific ethnic group. Rather, they are constructing pan-ethnic identities in opposition to whites.


Increased mobility and contact with people from different backgrounds has made it nonviable to determine exactly which linguistic features belong to which language variety and cultural group. People draw from multiple linguistic resources to convey meaning, and many such resources are not from the same language or even from a single variety of language. A common language practice is the strategic selection of linguistic features from a particular repertoire for specific rhetorical purposes and effective communication. Terms that account for how the use of distinct linguistic features reflect the cultural practices and identities of their speakers include multicompetency (Cook, 1991), transidiomatic practices (Jacquemet, 2005), polylanguaging (Jørgensen et al., 2011) and, more recently, translanguaging (García, 2011). Each term rejects the notion that languages remain separate in speakers’ heads and instills the idea that people employ all their available linguistic resources for identity formation and meaning making.

It is well documented that people often borrow linguistic features from other ethnic groups (Fought, 2003). Rampton (1995) refers to such language use as crossing and claims that as “a form of everyday cultural politics” (p. 14), crossing situationally activates various facets of a person’s multiple identities either to align with or differentiate themselves from others (see also Bailey, 2000). Choice of language is not merely based on proximity, age, or generation, but also on ethnic affiliation and network (Fought, 2003; Milroy & Wei, 1995; Zentella, 1997). Studies in the field of sociolinguistics have long documented the use of ethnic speech markers by people who are not considered members of the group tied to that specific language variety. For instance, multiple studies have focused on various features of American English attributed to people from a particular US region and how these linguistic features have gone from being regional markers, such as the Southern drawl, to indexing different positive or negative social meanings and attributes that might include friendliness, familiarity, lack of education, gayness, and stupidity (Campbell-Kibler, 2007; Johnstone, 1999; Podesva, 2007). These studies, which focus on indexicality and styling, teach us that linguistic features are not fixed; rather, they are part of an indexical field, a “constellation of ideologically related meanings, any one of which can be activated in the situated use of the variable” (Eckert, 2008, p. 454). That is, a linguistic feature that typically belongs to a particular variety does not simply reflect the social categories and social changes it is usually attributed to, but it also serves as a “resource for constructing those categories and participates in social change” (Zhang, 2005, p. 431). This is the case for many features of African American English (AAE).

It is important to research the stylistic choices that people make when using different varieties of language in order to discover what social categories and identities such choices index. Studies of indexicality, language crossing, and styling have been fruitful in showing that the rhetorical use of AAE, for example, is common among people who do not self-identify as African American for reasons such as indexing hyper-masculinity (Chun, 2001), creating social boundaries among Asian-American populations (Reyes, 2005), and affiliating with an urban youth (Cutler & Røyneland, 2018) and hip-hop culture among white-American populations (Cutler, 1999). AAE use as a stylistic choice to index social categories is also found among some Latinos in the US. For example, some Latinos use AAE to construct a gang identity or index gang membership (Carter, 2013; Fought, 2004). Nowadays, features that before would have been attributed to AAE can index social categories of coolness, (hyper)masculinity, gang culture, hip hop culture, subversiveness, urbanism, and toughness.

The globalization of communication in online social media offers new possibilities for the study of stylistic language features by populations not typically associated with the language variety, providing a platform for people to write themselves into being (boyd, 2008). Online social spaces offer people another platform to construct their different identities, as well as access to all their linguistic resources to perform those identities and communicate with their networked audiences (Androutsopoulos, 2013; Christiansen, 2015a).

This article examines the use of AAE on Facebook by a group of Mexican bilinguals. I asked how young members (14-31 years old) of a social network of Mexican bilinguals living in Chicago use vernacular English traits on Facebook typically attributed to African-American speakers. Specifically, the guiding question in this article is: For what purpose do they use vernacular features in their posts? To answer this question, in the next section I give a brief overview of AAE, summarize the literature on AAE use by Latino populations, and discuss AAE use by non-African American individuals on social media. I next present the theoretical framework and methodology of the study. In the findings, I summarize the participants’ use of AAE features and discuss the possible reasons for their use based on interview data. I conclude with a discussion on racializing languages and identity formation, as well as the implications of these for the study of language varieties.

AAE Characteristics

African American English (AAE) is an umbrella term that describes and accounts for a continuum of linguistic traits attributed to African American speech. It ranges from the most standard speech forms to the most vernacular varieties (Benor, 2010; Rickford, 1998). Despite variations, certain well-documented AAE features are common across a variety of regional contexts, for example, the absence of third-person singular -s and variation copula leveling resulting in the single form was (e.g., “We was there”) (Weldon, 1994; Wolfram & Thomas, 2008). Other prominent features of AAE are invariant be commonly used to mark habitual aspect and existential emphasis (Alim, 2001), such as, “They usually be acting silly,” and habitual be for intermittent activity, as in “Sometimes my ears be itching.” These contrast with copula deletion/absence of be (e.g., “You ugly; she gonna do it”). Other syntactic features include the construction ain’t for negative forms and constructions such as “I ain’t lyin’” [ = am not] or “He ain’t never [ = hasn’t ever] had a pet in his life” (except ain’t for ‘didn’t’) such as in “She ain’t been there for a while.” Among phonetic features one can find r and l deletion after vowels, as in he’p for ‘help,’ afta for ‘after’; yo for ‘you’; simplification of consonant clusters, as in des’ for ‘desk’; a glide diphthong pronounced as a monophthong, as in “Ah think ah’ve got somethin’ in mah ah”; deletion of voiced stops /b/ /d/ /g/ when occurring as the first consonant in tense-aspect markers or auxiliary verbs, as in “Ah ma do it – I’m gonna do it”; and replacement of orthographic th. The h sound is replaced depending on voicing; voiced th or ð is generally replaced with d or v, and voiceless th or θ with t or f, as in tin for ‘thin’ or dem for ‘them.’

AAE is the most studied and controversial variety of American English, and studies devoted to explaining its syntax and usage exceed by a factor of five the number of studies about other varieties of English (Wolfram & Schilling-Estes, 2006). AAE is a system with specific features and linguistic rules whose speakers come from different regions in the US and belong to diverse social groups. AAE speakers are those who apply AAE features and rules systematically in their language use; however, not all speakers who apply and use some features are necessarily AAE speakers (Rickford & Rickford, 2000; Wolfram & Schilling-Estes, 2006). Most importantly, individuals do not have to use and apply all features and rules in their language use to be considered AAE speakers. Additionally, some people who have widely appropriated vernacular features believe they exclusively use AAE when, in fact, this is not the case (Rickford & Rickford, 2000). AAE language features are not always discrete. For instance, features of a variety often attributed to a particular group can be taken up by another group to the extent that those features become part of their particular repertoire, too.

Another issue is that some AAE features are used as fashionable slang terms. Acquired features of a variety differ by region and social class and are used and adapted according to the age, gender, status, and lifestyle of a person (Rickford & Rickford, 2000). Some words used by AAE speakers have crossed over into general US usage and are seen as fashionable. The global popularity of rap and hip-hop music and culture has also caused AAE features and slang to be taken up and used by diverse communities of speakers.

AAE Use by Latino Populations

Research shows that in areas where Latinos are in contact with African American communities, they appropriate some AAE traits (e.g., Wolford & Evanini, 2006). Examples include in communities sharing the same geographical location (e.g., Wolfram, 1969; Zentella, 1997) and youths sharing a multiethnic school (e.g., Carter, 2013; Dunstan, 2010). Fought’s (2003) study of Chicanos in Southern California provides a detailed analysis of how Chicanos use AAE features in their speech. However, Fought (2003) points out that it is not possible to determine the extent to which AAE influences Chicano English because both varieties, as well as other vernacular ones, include similar features. Recent studies of Mexicans and other Latinos in North Carolina establish that contact with African Americans has influenced the English varieties used by people from Latin America (e.g., Carter 2013). However, they also show that the influence is not uniform and that the appropriated features have different sociolinguistic functions within and outside communities. For example, Latinos who have less contact with African American classmates exhibit similar patterns to those who have more contact, suggesting other social networks outside of school also influence the speech of Latino populations (Carter, 2013; Dunstan, 2010). The purpose of the present study is to determine why young (14-31-year-old) members of a social network of Mexican bilinguals living in Chicago use vernacular English traits typically attributed to AAE and whether they have explicit consciousness of their use.

AAE Online

In online settings, spoken and written language are not dichotomous but are the product of specific literacy activities, which give way to new forms of written communication (Cunningham, 2014; Haas et al., 2011). In variationist studies, there is no methodology that captures whether speakers deliberately choose to write differently from others; however, several studies show that when the choice is conscious, avoidance (or increased use of standard English) is the preferred pattern (e.g., Christiansen, 2015a, 2017a; Cunningham, 2014; Lam, 2009), e.g., to prevent criticism or appearing uneducated, especially if the conversations are not private. Tagliamonte (2016) claims there is no degeneration of grammar in internet usage, but rather a fluid and complex new set of written registers. Although there is no systematic way to write non-standard varieties of language, variationist studies have found that abbreviated spellings and shorter forms differ by geographical location (see Christiansen, 2018a for examples). Thus, it is possible to investigate linguistic differences online in a similar way to physical environments.

Among internet studies covering vernacular features traditionally associated with AAE, Cunningham (2014) describes the language used by African Americans online as a hybrid of digital language and textese and claims that digital African American language (DAAL) is a “robust form of written communication” (p. 404) that is contextually, consciously, and pragmatically used among AAE speakers. While Cunningham’s study examined DAAL as a language category for African American speakers, other research has explored the use of AAE by non-African-Americans in online settings. For example, a study of a Taiwanese-American youth found that her various languages and language varieties, including AAE and “hip-hop language,” helped her navigate her transnational affiliations in instant messages (Lam, 2009). Similarly, a study of a Chinese-American YouTube star showed that adopting the stereotype of African-American black hypermasculinity helped him construct Asian masculinity and reproduce ideas of linguistic authenticity in the US (Chun, 2013). There have also been studies of white Americans in the hip-hop context crossing language and using AAE to portray a different persona (Bucholtz, 1999, 2011), mock AAE, and position themselves as better than African Americans both in and outside of social media (cf. Cutler, 2015). Although the research on AAE use by Asian American and white American populations has been fruitful, there are fewer accounts of online AAE use by Latinx. The goal here is to contribute to filling this gap by exploring why some Mexican bilinguals use certain features of AAE in their social media posts.

Theoretical Framework

An ethnolinguistic repertoire is “a fluid set of linguistic resources that members of an ethnic group may use variably as they index their ethnic identities” (Benor, 2010). The study of ethnolinguistic repertoires facilitates analysis of variation in a realistic way and accounts for different social dimensions within a single community, like social organization, profession, social class, sexual orientation, and region (Benor, 2010). Such an approach does not assume that linguistic variation correlates with predetermined social categories but rather focuses on individual agency and examines the role language plays in the construction of social categories, whether done consciously or not. It also considers that group members will use different linguistic resources depending on who they talk to and what identity they want to construct.

Among the ways to study identity, approaches that examine the use of language as a means of signaling ethnic identity are robust. Language is a “powerful tool in the display of the ethnic self” (Hall & Bucholtz, 1995, p. 357). However, language does not simply signal a particular ethnic identity, because it also includes linguistic features borrowed from other minority ethnic groups with whom speakers interact (Fought, 2004). For instance, studies of male Puerto Rican speakers in New York City show them adopting African American Vernacular English (AAVE) features in their speech as a result of contact with friends and acquaintances (Zentella, 1997). Apart from accounting for intra-speaker variation, the ethnolinguistic repertoire approach accounts for stylistic shifts (sometimes regarded as dialect- or code-switching or crossing) and examines how individuals change the way they speak their native language(s) over time (Benor, 2010). We can then observe a feature of language typically associated with one group being used by an out-group person, which aligns well with the idea that ethnic groups are fluid and so are their languages. Thus, an ethnolinguistic repertoire approach more accurately portrays language use within groups and facilitates understanding of how individuals construct membership within different social networks through the use of language. While this approach cannot account for whether their choices are conscious or not (Benor, 2010), it can account for an individual’s intentionality (Gumperz, 1982) or selective use of a distinctive repertoire to effect action from the interlocutor. What remains to be done is to investigate how aware speakers are of the distinctive features and to what extent they manipulate them and for what purposes.


Context of Study

This study emanates from a long-term ethnographic project focused on the language ideologies and language practices of approximately 58 members of different generations of a transnational, close-knit, social network of Mexican bilinguals with a traditional patriarchal structure (cf. Christiansen 2015ab, 2017ab, 2018b). For the current study, I draw on two sets of data. The first includes online conversations between members who were Facebook users (14 out of 58), interview data that focuses on language use and ideologies, and field notes of participant observations both online and offline between October 2011 and May 2012. In the second set of data, I narrow the focus to interviews of two focal members, Dalila and Armando, from September 2017 to May 2018. I focus on Facebook because it remains one of the most widely used social network sites to date (Smith & Anderson, 2018). While the participants also use private platforms such as WhatsApp and Messenger or public ones such as Instagram and Snapchat, they link their Instagram to their Facebook and share screenshots of their Snaps on Facebook. It is then on Facebook that they usually construct conversations or participate in private Facebook Groups, such as a family group.


Table 1 lists all Facebook users by gender and age as of 2011. All were born in the US and grew up bilingual, except Lola and Deya, who were born in Mexico and learned English when they moved to the US in their 20s. However, the participants’ degree of bilingualism varies. Spanish is their first language, and they learned English at school. All except Natalia, Casiano, and Flavio had at least one year of schooling in Mexico. For additional demographic information, refer to Christiansen (2015b). These participants form part of a tight-knit, multiplex, network. That is, they are all relatives, go to school or work together, and socialize online and offline together. They often have family parties at each other’s houses and share the same friends.

Table 1. Participants’ information

Data Collection

I followed a discourse-centered methodology that calls for simultaneous data collection and analysis (Androutsopoulos, 2008). To capture the AAE use and changes over time, for this article, I collected two sets of online data from the Facebook users and conducted interviews at two time periods:

First Dataset – 2011

I collected online data from all 14 Facebook users during one year of their participation in the study. I also conducted one ethnographic interview with each Facebook user. The online data resulted in 8,729 Facebook posts. A Facebook post was considered to be anything (text, image, video, hyperlink, or a combination of any modality) published by a user on Facebook when they hit ‘enter.’ I saved the Facebook pages as PDFs and coded them using the program Dedoose, a cross-platform app for analyzing data that include text, photos, audio, and videos.

Second Dataset – 2017-2018

For the second set of data, I focused only on Dalila and Armando, who were frequent AAE users on Facebook in the first set of data, to understand how AAE is used and, more importantly, why. This dataset includes one interview with Dalila and one with Armando, which took place in August 2017 and January 2018, respectively, when I noticed that their use of Facebook had decreased overall since 2011. I carried out these ethnographic interviews via video-call on Facebook Messenger, the medium chosen by the participants. I began the interviews by asking each participant about changes in their participation on Facebook; I showed them posts I had coded as a vernacular variety and asked them why they previously used the vernacular and why they no longer did so. I concluded by asking what they thought about using specific language features to represent themselves and what they thought of others who used by them. I was careful not to use terms such as AAE, Black English, or anything that would put a label on any variety of English. They provided me with the labels they used. Both interviews were transcribed for analysis.

Data Analysis

First Dataset – 2011

I first analyzed the language use of all 14 Facebook users to contextualize how they use different language varieties.[1] The analysis focuses on Facebook posts, including statuses, images, videos, hyperlinks, and memes in their threads, as applicable. A thread is when a main post receives one or more replies in any mode (text, emoticon, hyperlink, video, etc.) and not just a “like.” Some posts and conversation threads happened in more than one language and more than one variety of a particular language. For the present study, it was key to include a category on language, which I then subdivided into linguistic categories. Because I am using an ethnolinguistic repertoire framework, I categorized language varieties according to distinctions alluded to by the participants themselves.[2] For instance, for English I had a category for AAE (Rickford & Rickford, 2000; Wolfram & Schilling-Estes, 2006), White-European American English EAE (Wolfram & Schilling-Estes, 2006), and Latino English LE (Fought, 2003). For Spanish, I used categories for Ranchero Spanish (Farr, 2006; Santa Ana & Parodi, 1998) and Mexican-origin Spanish in the US (Lipski, 2008). Many posts contain English and Spanish. I focus on vernacular use of Spanish elsewhere (Christiansen, 2015b); here, I focus on the use of AAE. For the purposes of this study, I analyzed the first and last month of data collection for each of the 14 participants, resulting in a total of 1,364 posts.

To determine whether a feature was AAE or not, I followed Cunningham’s (2014) approach. The social media posts were considered AAE if:

• A single post included two or more AAE features;

• A single post included only one AAE feature, but it was followed by a post in the thread containing one or more AAE features.

However, if a single post only contained one AAE feature, but it was not followed by another post in the thread containing AAE features, then the first post was not considered AAE. Note that acronyms (such as lol), emoticons, and emojis are their own category; they were not considered for this analysis, since they could appear in any post containing any variety or combination of languages.

After coding all features found in the Facebook posts of the 14 participants into language categories, I employed discourse analysis (Saville-Troike, 2003) to examine the posts for language use patterns. The notion of audience in social network sites (SNS) (Christiansen, 2017ab) was helpful in coding how participants use language and whether they changed their language patterns depending on whom they were addressing. I coded for whether the participant was male or female and whether they were responding to a male/female in their threads. I knew this demographic information from my interviews and participant observation. I also paid attention to the type of post (e.g., main post or response post, as part of a thread), what features participants used in each post, and how they were used. At the end of these rounds of coding, I tallied the use of AAE by participant and identified Dalila and Armando as the female and male who made the most use of AAE features in terms of number of posts in which AAE appeared (see Table 2 in results).

Second Dataset – 2017-2018

For the interview data, I employed discourse analysis by coding inductively and carrying out recursive comparison of the data (Glaser & Strauss, 1967). Because my focus is to identify the reasons why participants used a vernacular feature and whether their reasons were implicit or explicit, I created categories that emerged from the data by using a language ideology framework (Woolard & Schieffelin, 1994). To code the explicit categories, for instance, if a participant said, “my English is never good enough,” that segment was coded as ‘explicit language ideology’ and more specifically, ‘ideology of native speakerism.’ To identify the tacit language ideologies and categorize those data into patterns, I used theories of indexicality (Silverstein, 1976), a property of speech through which language points to particular stances, acts, and identities (Bucholtz & Hall, 2004; Duranti & Goodwin, 1992). For example, if a participant expressed that people spoke “gringo English” (English spoken by a white person from the US), I coded the segment as ‘implicit language ideology’ and more specifically ‘language purism’ and ‘racialization.’ I added the code ‘racialization,’ as proposed by Shuck (2006), to describe a language feature typically associated with a particular racial/ethnic group. I use indexicality, then, to discover how participants view the use of vernacular features not typically associated with themselves (Latino culture) and their own reasons for making such stylistic choices.

Findings and Discussion

The findings are divided into three sections. In the first section, Language Use, I explain how the 14 Facebook users employed language in 1,364 posts and exemplify some AAE features they used. I also report on men’s and women’s use of vernacular features on Facebook. In the second and third sections, I contrast Armando and Dalila’s early interviews with those in 2017 and 2018. I analyze the two main reasons why they used AAE early on and argue that one is to construct a non-white identity, and another is to challenge entrenched gender hierarchies. I show that men negatively judge women for vernacular use but that women feel empowered by using it.

Language Use

Types of Use

Although the Facebook data show some lexical items and attempts to use syntactic patterns and features from AAE, in general, the 14 Facebook users do not show syntactic competence in AAE. Instead, features of AAE and those shared with other vernacular varieties can be viewed as crossing (Rampton, 1995) into these linguistic varieties. Many times, a vernacular feature was used but did not follow established conventions. For example, the copula be for habitual or intermittent activity was written in Facebook posts as in examples 1 and 2:

1) “I be loving this free food!

2) “truth iss:D… yur funny – we be having fun in mexico:))!!!

In some other cases, the use of AAE did follow conventions, such as in examples 3 and 4:

Absence of copula for contracted forms of is and are:

3) “hes on his period.... :) he hungry he need be feed

r and l deletion after vowels:

4) “he’s no playa

These correct usage cases, however, are single uses, and other AAE features do not occur in the same post. Thus, the data demonstrate that Facebook users who use AAE try to incorporate some features in their Facebook Timelines but lack full syntactic competence, which supports findings by Carter (2013) and Dunstan (2010). I suggest that their use is a borrowing of AAE features that serves stylistic purposes rather than being a part of the users’ overall linguistic repertoire.

Frequency of Use

During the two months of data collection in 2011, the 14 Facebook users made 1,364 posts. While Standard English was widely used in the majority of posts (79.44%), AAE features appeared in 8.81% of them (n=120). AAE features in isolation (not occurring with other AAE features) accounted for 11.73% of the latter usage. Table 2 lists the 14 participants and the percentage of AAE found on their Timelines in those two months in the first column.

Table 2. AAE use on Facebook by participants in 2011

Among the 120 posts where AAE features appeared, I focused on Armando and Dalila, as they were among the most active Facebook posters (122 and 169 posts, respectively) in 2011, and they used a considerable amount of AAE (in 23.77% and 10.65% of their posts, respectively).

In November and December 2017, in contrast, Dalila and Armando’s Facebook posts only occasionally contained AAE (1 out of 67 posts for Dalila, and 3 out of 41 – two of which were in memes – for Armando). I attribute this to a change in context for both speakers. Dalila now works as a college advisor, while Armando now lives in Mexico and communicates primarily in Spanish. Both have also changed their social networks online, and most of their posts express either political awareness or social criticism.

AAE Use and Gender

A pattern of use among men and women can be seen in Table 2 above. All the men use AAE features, even if sporadically; however, no women 25 or older used AAE features even once. This is unlike the younger male and female participants, all of whom use AAE features with varying degrees of frequency. Further, the use of AAE features varies depending on the gender of the addressee. Table 3 shows the use of AAE according to the gender of the sender and the addressee (the person who was either tagged in the message or who responded directly to the post as if it was for him/her).

Table 3. Use of vernacular features according to gender of sender and addressee

When a man addresses a woman, he uses more Standard English. Notice, for instance, that while in male-to-male interactions, AAE features occur in 38 posts or 31.67% of this data set (n=120), but when men address women, men’s use decreases to 7 posts or 5.83%. When women direct their comments to males, they increase their AAE use frequency. Female-to-female interactions use AAE in only 1.67% of posts, but when women engage in conversation with men, their use of English vernacular features increases to 11.66%. A degree of accommodation from both genders to their interlocutors can be observed here.

Men avoid addressing women using AAE even when women address them in AAE. An example from 2011can be seen in this Facebook exchange initiated by Lola:

Excerpt 1. "Answer your phone"

In this Facebook thread, Lola, who is angry, used both cussing language and AAE features. Although Casiano uses AAE extensively, he does not do so (or even point it out) in this short interaction.

Another example of female AAE use from 2011 is the following conversation between Natalia (Dalila’s sister) and a non-black male friend (as evidenced by his name and profile picture and confirmed by Natalia). In this exchange thread, Natalia makes a statement that she does not want to go back to university (located outside of Chicago). Then the talk shifts to sports, and she lets him know she saw him play.

Excerpt 2. “Later bruh”

Notice that Natalia’s status update contains no AAE features. It is not until line 4, after a male friend replies to Natalia, that she answers using AAE lexical features to her friend, who informally requested information (“feed me”). She promises to catch up with him in line 7, where she uses a colloquial version of “though” and uses the term “black” to describe him. After his response in line 8, Natalia again uses AAE features in line 9 to respond. She shortens “you” and “though,” and displays copula absence. Her friend responds with a double negative and use of “aint” and a shortened form of “more.” AAE use has been shown to be a masculine practice (Chun, 2013), and the gender patterns in Tables 2 and 3 largely support this. But the two examples above show female participants using AAE features to engage in conversations with men, something rarely evident when talking to other women. In the first example, AAE is used to be assertive, show anger, and call attention; in the second exchange it is used to banter and create an equal status between the male and female speakers (note the praise in line 7, “tugging,” and congratulations in line 9 while admitting she was rooting for the other team).

Non-White Identity

In the ethnolinguistic repertoire approach, determining the linguistic variety of speakers is not necessary. Speakers can use an array of linguistic resources, some of which are associated with other speaker groups without necessarily enacting that other ethnic identity. However, although speakers may not be intentionally enacting a specific ethnic identity when using language, interlocutors assign social identities to them, nonetheless.

For example, Dalila views language as an intricate part of culture, and using a non-Latino vernacular variety can result in disapproval from her social network. Moreover, Dalila is perceived as having darker skin than her other family members. Her complexion can prompt people to think she is half African American, and after hearing her use a non-standard variety of English, people, especially middle class white European Americans, classify her as a speaker of AAE. In a 2011 interview, Dalila further explained where this confusion occurs:

“at work... they thought oh that I was ghetto. And they would associate that with me being from the south side, cause most of them were from the north side and I guess they are really preppy in the north side… the north side is like the yuppies so... I mean when you go over there it’s like really quiet… like they are more hostile. So that's why I guess cause I've seen the other more hostile or like more... I mean there's bad things do happen in our side but I guess you hear more about it from the south side…”

In 2011, Dalila worked on the north side of Chicago, where there is a higher concentration of white people, whom she refers to as “yuppies.” She explains that they view her speech as being from the predominantly black south side of Chicago. She also refers to the south side as “our side.” Wolfram (1969) found that the use of AAE fostered racial isolation (blacks would use it more often with other blacks), and Dalila here shows that this is still the case, as whites grouped her with black Americans. Additionally, while she did not directly affiliate herself with black speech, she affiliated herself with “this side” (the south side), thus indexing a non-white, non-yuppie identity.

While Dalila is assigned a non-white identity, Armando self-identifies as non-white based on his own language ideologies, and he assigns a non-Mexican identity to other Mexican friends who try to speak standard English. He said in 2017, “I don’t speak the ‘correct’ English… [in the US] my English was never good enough.” In this quote, Armando exhibits an ideology of native-speakerism that places the middle-class white European English variety as the standard or ‘correct’ language and the way he speaks as not “good enough.”

He also thinks that younger generations do not want to sound Latino/Mexican and instead want to sound “white,” but they fail to pass as white.

“[Younger] generations want to speak like native speakers and they want to speak a white coated English. Even though we are indeed native speakers from [the US], we have a weird accent, I don’t know why but we have a little accent not like a white person. I even have some [Mexican] female friends that speak gringo (American) English like people here [in the US] and they think of themselves as gringuitas (little blondes/white girls)… some of them speak more ‘white’ and that makes them feel superior to us.”

In this quote, Armando uses the term native speaker again, but this time it is to say that even though they (Mexicans) are native speakers of English, their English has a funny accent different from the accent of whites (and not good enough, see previous quote). He positions non-Mexican accented speech (i.e., that of whites) as desirable to some in his community and criticizes younger generations who “want” to speak white English because it makes them feel superior. It is worth noting that in Armando’s family network, most people regard themselves as “US born Mexicans” and reject labels such as Mexican-American, Latino, or Hispanic.

In the case of Dalila, she is othered by her use of vernacular English, whether AAE or Latino English, and she is aware of this othering. In the case of Armando, as he others US born Mexican girls who speak “white coated” English, he distances himself from an American (white) identity. He speaks disparagingly of Mexican friends who speak gringo and do not display an ethnic accent. In both cases, language use is racialized by indexing or assigning a non-white American identity based on the use of vernacular features and characterized as adopting a white identity in the absence of such features (e.g., accent, AAE, Latino English). The use of language to index ethnic or racial identities is similar to earlier findings (Christiansen, 2015b) where members of this network used Spanish to index Mexicanness. That is, members equate the use of standard English with an American identity and Spanish with a Mexican identity.

Using vernacular features in English is tied to their sense of non-whiteness (Bailey, 2000). Thus, these data show that language variation can be seen as racialized (Christiansen, 2018; Shuck, 2006). While in the present case, the Mexican participants racialize their linguistic practices, Shuck (2006) showed that white, middle-class, native English speakers born in the US invoked an ideology of monolingualism to construct their nativeness as unmarked and normal, reproducing a hierarchical social order that privileges native speakers and Caucasians and positions such traits as ‘natural’ characteristics intrinsic to Americanness. This explains, in part, why Armando kept describing Mexican Americans as speaking like they were from ‘here’, even though most are US-born citizens and native speakers of English (Latino English). Therefore, the use of AAE and other vernacular features for these participants does not index a Mexican or an African American identity, but rather such use constructs a more generally non-white identity.

Changing Hierarchies through Language

A previous analysis of language among members of this social network showed that participants care deeply about their Mexican-ness and construct it using four emic criteria: language, color (skin and hair), practices of culture, and transnationality (Christiansen, 2015a). While the oldest male traditionally held most power, now the most bilingual and transnational does. Based on this, I argued that members are contesting the hierarchical order of their social network and that changing hierarchies entails challenging the status quo which traditionally places males above females.

One way to see specific hierarchical change is in language. For example, non-standard language use by females is policed by both males and females, yet despite this, women increase their use of vernacular language when talking to men (Table 3 above). In Dalila’s view, women are empowered through language. At one time, they had to avoid the use of vernaculars or apologize for using them. Now, as seen in Table 3, females use more vernacular features when talking to men than to each other. Dalila, in her recent interview in 2017, explained that there have been many changes in the broad family network. People have returned to or now live in Mexico, some female members have divorced, other young women have decided to move out of the family household and live separately. These changes have affected the traditional family dynamic, and women are now more independent.

Dalila also rationalizes the language women used on Facebook in 2011 and since then in speech. In fact, I did not need to show her examples or finish my questioning during our 2017 interview, as she quickly interjected that she noticed that some of her female relatives’ language and behavior was different than in 2011. She described these changes as “a little more ballsy… a little more like assertive.” Now that some of the women relatives are divorced or they are more independent, men do not police their speech as much. Dalila thinks females are more assertive, taking more risks toward being independent than in the past. When specifically asked about female use of vernacular features, Dalila explained:

“[women] may try to do it to kind of like a feminist way of being like they also have power and they speak however they want. It almost makes it ok for the man to go ahead and do it”

In this excerpt, Dalila sees the female use of vernacular varieties as a way for women to position themselves on an equal plane as men, who are the ones known to use such language varieties. She expresses that using a language variety they have been explicitly criticized for empowers women to speak in any way they want. Additionally, males usually try not to use vernacular varieties when they talk to females. When I showed Dalila the numbers in Table 3 and asked why females increased their vernacular use when talking to males, she commented that they were doing it to be feminist, empowering themselves and asserting their place as equals, which in turn gave men permission to speak that way in front of them.

Interestingly, the same was not the case for Armando, who i­­n his 2018 interview answered the following when asked of his views on female use of vernacular features:

“A woman when she speaks like that, it’s not empowering is degrading, like women wants to think she has power, but I hear it and it sounds wrong to me, but there are a lot of [females] that speak like that because their families speak like that, but that doesn’t matter, a woman is a woman and I don’t see it is empowering to want to be eye to eye, I think they are different level and we can manipulate language for advantage or disadvantage depends how we’d like to see it but I don’t know, I may sound bad putting it this way but that language doesn’t sound pretty or sexy and it sounds rough with cussing words and it’s not nice.”

From this quote, it is evident that Armando holds the same language ideologies as his older generation relatives. Women should not use foul or vernacular language because it makes them “sound rough,” not “sexy” and “not nice,” implying that the only way a female can be empowered is by sounding nice and being sexy, and further linking language use and gender. What is more, although Armando is aware that females who use vernacular and profanity have been socialized to speak that way, and he does find blame in males whose speech may contribute to females learning to speak that way, he still punishes the females for using vernacular language. He thinks that a woman must speak differently regardless of how her family speaks, as it is her duty to sound ladylike (free of vernaculars and cuss words). Armando also recognizes that “we” manipulate language for advantage and recognizes women may use language to empower themselves; however, he disregards this rhetorical use in favor of his belief that women should be pretty, sexy, and nice and not equal to men, because they “are a different level.”

When asked how he viewed men who use the same type of language as the young women, Armando explained that they were just “homies” and “compas” and that it is not necessarily good if they are not friends, but it’s still ok. Note that Armando recognizes that some of his relatives do speak that way (the parents, the male siblings), but not even in the family context does he think women should use this language. He also recognizes that women are the only ones being negatively judged for using vernaculars and foul language, but he does not see its use as empowering. On the contrary, Armando thinks that using vernacular language causes women to lose respect, as seen in the following excerpt from 2018:

“A female, she’s being judged, I feel like my mother, I don’t want my mom to speak that way, it’s not empowering… and [females] do not gain respect by speaking like that. A woman must gain respect, and in that way, she can empower herself, if not, all they get is being seen as if they are from the hood”

Armando expresses that the purposeful use of AAE language causes women to be judged as from the “hood,” something Armando does not consider desirable. Like many males in US and Mexican societies, Armando considers that women “must gain respect,” suggesting that women are born with a deficit or an inherent lack of respect. One way to gain respect and maintain it is to use non-vernacular language. So, when females use vernacular varieties, they lose rather than gain respect. From this deficit perspective, Armando exposes the common ideology that females are good and sexy if they use nice language. This exemplifies the ideology that the speech equals the speaker, and, in this case, nice speech equals a nice speaker. Thus, ideologies of language use can be tied to gender. In Dalila’s view, vernacular language use can empower women to be equal with men, and in Armando’s view, it disempowers women from being sexy and nice, from which a woman’s traditional sources of power derive.


These data show that some bilinguals make use of all their available repertories to create meaning and challenge hierarchies. Consistent with previous research (Carter, 2013; Dunstan, 2010), while the bilingual Latinos in my study use vernacular features typically attributed to AAE, they do not conform to AAE syntax and use. That is, they use it following their own patterns and for particular purposes. Such use does not construct an African American identity, but rather it indexes non-whiteness.

While the use of vernacular features of spoken language does occur, as evidenced by Dalila and Armando’s interviews, most Facebook posts do not include them. Participants and members of their network do not widely use AAE and avoid some vernacular varieties even in Spanish on Facebook for fear of criticism, as is evident from their evaluations of vernacular use as “ghetto” in face-to-face conversations. At the same time, the absence of vernacular features on Facebook and in their spoken English, in particular, would make them sound like “yuppies” (or whites), which would also be criticized, since they are proud to be Mexican. So, in this case, the occasional use of AAE features constructs a Southside of Chicago identity, an American identity. They are also not constructing a Mexican-American identity or their version of Latinidad or Hispanic identity, as in Carter (2013). They are constructing a non-white identity (Bailey, 2000; Chun, 2001), which is why they are also critical of speech and written prose void of AAE or Latino English features.

This research focused exclusively on the role that vernacular features play in the construction of identities in a transnational bilingual network, especially those features regarded as AAE by the participants. Findings show that vernacular features do not represent who speakers are. That is, while participants used AAE vernacular features to express their opinions, it was not to form an African American identity but to emphasize their non-white identities. In the case of some women, AAE was used to empower themselves, and as Dalila said (and Armando recognized but did not accept), enact feminism. This finding is particularly important to the online feminism literature, where women of color are underrepresented. Feminist research shows that use of profanity and non-standard language is part of strong discourse used “to establish and challenge power structures in interaction” (Morikawa, 2018, p. 120). AAE is used by both women and men to portray strength, even aggressiveness, but also coolness. This research has implications for the study of language and identity. Linguistic analysis of varieties of English that use frameworks such as ethnolinguistic repertoire eliminate the need to attach a specific variety to a single group or assume that a person using a variety is enacting that specific identity or are even in direct contact with the variety.

Finally, ethnic and national origin terms used for coding sociolinguistic data present a range of problems because diverse groups of people use various languages and language varieties (Bailey, 2000). People shift ethnic identifications over their lifetimes due to work, communities of practice, and belonging to multiple imagined communities (Benor, 2010). While ethnic groups are socially constructed (Benor, 2010), the labels and associations with them have real-life consequences for those who are considered members. One consequence is the erasure of “variation within the ethnic/racial category African American/Black in and of itself” (Blake, 2014, p. 548). Taking an ethnolinguistic repertoire approach departs from the simplistic categorization of ethnic groups based on language use and facilitates an understanding of speakers and how they decide to draw on language ideologies to enact their social and ethnic identities through the use of any given feature or variety of language.


Thanks are due to Marcia Farr for checking my initial coding for the vernacular varieties of AAE and Latino Spanish. I would also like to thank the anonymous reviewers for their constructive comments on earlier versions of this manuscript, and the editor for her detailed comments that strengthened the clarity of the argument. I appreciate all your time and energy.


  1. As part of the larger project, I followed the discourse-centered methodology to collect and analyze all 8,729 posts. For a specific account of the procedures, refer to Christiansen (2018b).

  2. For the larger project, I also supplemented data collection with ethnographic interviews, which were analyzed to identify the language varieties that participants alluded to (cf. Christiansen, 2015a, 2018b). The ethnographic interviews focused on how these Facebook users employed Facebook for the purpose of maintaining transnationalism (see Christiansen, 2013, for the full interview protocol). Each interview lasted one to two hours and was conducted at their homes. At the time of data analysis and coding, while quantifying their use of English and Spanish, I realized that participants were using several varieties of English and Spanish both in interviews and on Facebook (see Christiansen, 2015b, for an account of Spanish varieties).


Alim, H. S. (2001, October 11). I be the truth: Divergence, recreolization and the equative copula in Black Nation Language. Paper presented at New Ways of Analyzing Variation, 30. Raleigh, NC: North Carolina State University.

Androutsopoulos, J. (2008). Potentials and limitations of discourse-centred online ethnography. Language@Internet, 5, article 8. Retrieved from http://www.languageatinternet.org/articles/2008/1610

Androutsopoulos, J. (2013). Networked multilingualism: Some language practices on Facebook and their implications. International Journal of Bilingualism, 19(2) 185–205. doi:10.1177/1367006913489198

Bailey, B. (2000). The language of multiple identities among Dominican Americans. Journal of Linguistic Anthropology, 10(2), 190–223.

Benor, S. B. (2010). Ethnolinguistic repertoire: Shifting the analytic focus in language and ethnicity. Journal of Sociolinguistics, 14(2), 159–183.

Blake, R. (2014). African American and black as demographic codes. Language and Linguistics Compass, 8(11), 548–563. doi:10.1111/lnc3.12118

boyd, d. m. (2008). Taken out of context: American teen sociality in networked publics. Ph.D. dissertation, University of California, Berkeley.

Bucholtz, M. (1999). You da man: Narrating the racial other in the production of white masculinity. Journal of Sociolinguistics, 3(4), 443-460.

Bucholtz, M. (2011). White kids: Language, race, and styles of youth identity (1st ed). Cambridge, UK: Cambridge University Press.

Bucholtz, M., & Hall, K. (2004). Language and identity. In A. Duranti (Ed.), A companion to linguistic anthropology (pp. 369–394). Malden, MA: Blackwell Publishing Ltd.

Campbell-Kibler, K. (2007). Accent, (ING), and the social logic of listener perceptions. American Speech, 82(1), 32–64. doi:10.1215/00031283-2007-002

Carter, P. M. (2013). Shared spaces, shared structures: Latino social formation and African American English in the U.S. south. Journal of Sociolinguistics, 17(1), 66–92. doi:10.1111/josl.12015

Christiansen, M. S. (2013). Facebook as Transnational Space: Language and Identity among 1.5 and Second-Generation Mexicans in Chicago. [ProQuest Dissertations Publishing]. Retrieved from http://search.proquest.com/docview/1718184746

Christiansen, M. S. (2015a). ‘A ondi queras’: Ranchero identity construction by US born Mexicans on Facebook. Journal of Sociolinguistics, 19(5), 688-702. doi:10.1111/josl.12155

Christiansen, M. S. (2015b). Mexicanness and social order in digital spaces: Contention among members of a multigenerational transnational network. Hispanic Journal of Behavioral Sciences, 37(1), 3-22. doi:10.1177/0739986314565974

Christiansen, M. S. (2017a). Creating a unique transnational place: Deterritorialized discourse and the blending of time and space in online social media. Written Communication. doi:10.1177/0741088317693996

Christiansen, M. S. (2017b). Language use in social network sites: The influence of orality in the digital writing of Mexican bilinguals. Writing & Pedagogy, 9(2), 369–392. doi:10.1558/wap.30281

Christiansen, M. S. (2018a). Cultural influences in texting. In J. I. Liontas (Eds.), The TESOL encyclopedia of English language teaching. Wiley Blackwell and TESOL International. doi:10.1002/9781118784235.eelt0683

Christiansen, M. S. (2018b) ‘¡Hable bien m’ijo o gringo o mx!’: Language ideologies in the digital communication practices of transnational Mexican bilinguals. International Journal of Bilingual Education and Bilingualism, 4(21). doi:10.1080/13670050.2016.1181603

Chun, E. W. (2001). The construction of white, black, and Korean American identities through African American Vernacular English. Journal of Linguistic Anthropology, 11(1), 52–64.

Chun, E. W. (2013). Ironic blackness as masculine cool: Asian American language and authenticity on YouTube. Applied Linguistics, 34(5), 592–612. doi:10.1093/applin/amt023

Cook, V. J. (1991). The poverty-of-the-stimulus argument and multicompetence. Second Language Research, 7(2), 103–117. doi:10.1177/026765839100700203

Cunningham, J. M. (2014). Features of digital African American language in a social network site. Written Communication, 31(4), 404–433.

Cutler, C. A. (1999). Yorkville crossing: White teens, hip hop and African American English. Journal of Sociolinguistics, 3(4), p. 428-442.

Cutler, C. (2015). White hip-hoppers. Language and Linguistics Compass, 9(6), 229–242. doi:10.1111/lnc3.12139

Cutler, C., & Røyneland, U. (2018). Multilingual youth practices in computer mediated communication. Cambridge, UK: Cambridge University Press.

Dunstan, S. (2010). Identities in transition: The use of AAVE grammatical features by Hispanic adolescents in two North Carolina communities. American Speech, 85(2), 185–204. doi:10.1215/00031283-2010-010

Duranti, A., & Goodwin, C. (1992). Rethinking context: Language as an interactive phenomenon. Cambridge: Cambridge University Press.

Eckert, P. (2008). Variation and the indexical field. Journal of Sociolinguistics, 12(4), 453–476. doi:10.1111/j.1467-9841.2008.00374.x

Farr, M. (2006). Rancheros in Chicagoacán: Language and identity in a transnational community. Austin, TX: University of Texas Press.

Fought, C. (2003). Chicano English in context. New York, NY: Palgrave Macmillan.

Fought, C. (2004). Ethnicity. In J. K. Chambers, P. Trudgill, & N. Schilling-Estes (Eds.), The handbook of language variation and change (pp. 444–472). Retrieved from http://onlinelibrary.wiley.com/doi/10.1002/9780470756591.ch18/summary

García, O. (2011). Bilingual education in the 21st century: A global perspective (1st ed.). Malden, MA: Wiley-Blackwell.

Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory; strategies for qualitative research. Chicago: Aldine Publishing Company.

Gumperz, J. J. (1982). Discourse strategies. Cambridge, UK: Cambridge University Press.

Haas, C., Takayoshi, P., Carr, B., Hudson, K., & Pollock, R. (2011). Young people’s everyday literacy: The language of Instant Messaging. Research in the Teaching of English, 45, 378–414.

Hall, K., & Bucholtz, M., Eds. (1995). Gender articulated: Language and the socially constructed self. New York: Routledge.

Jacquemet, M. (2005). Transidiomatic practices: Language and power in the age of globalization. Language & Communication, 25(3), 257–277. doi:10.1016/j.langcom.2005.05.001

Johnstone, B. (1999). Uses of Southern-sounding speech by contemporary Texas women. Journal of Sociolinguistics, 3(4), 505–522. doi:10.1111/1467-9481.00093

Jørgensen, J. N., Karrebæk, M. S., Madsen, L. M., & Møller, J. S. (2011). Polylanguaging in Superdiversity. Diversities, 13(2), 23–38.

Lam, W. S. E. (2009). Multiliteracies on Instant Messaging in negotiating local, translocal, and transnational affiliations: A case of an adolescent immigrant. Reading Research Quarterly, 44(4), 377–397.

Lipski, J. M. (2008). Varieties of Spanish in the United States. Washington, DC: Georgetown University Press.

Milroy, L., & Wei, L. (1995). A social network approach to code-switching: The example of a bilingual community in Britain. In One speaker, two languages (pp. 136–157). doi:10.1017/CBO9780511620867.007

Morikawa, N. (2018). #YesAllWomen’s language: Women’s style shifting in feminist discourse on Twitter. Discourse, Context & Media, 28, 112-120. doi:10.1016/j.dcm.2018.11.001

Podesva, R. J. (2007). Phonation type as a stylistic variable: The use of falsetto in constructing a persona1. Journal of Sociolinguistics, 11(4), 478–504. doi:10.1111/j.1467-9841.2007.00334.x

Rampton, B. (1995). Crossing: Language and ethnicity among adolescents. London; New York: Addison Wesley Publishing Company.

Reyes, A. (2005). Appropriation of African American slang by Asian American youth. Journal of Sociolinguistics, 9(4), 509–532. doi:10.1111/j.1360-6441.2005.00304.x

Rickford, J. R., (1998). African American Vernacular English: Features, evolution, educational implications. Malden, MA: Blackwell.

Rickford, J. R., & Rickford, R. J. (2000). Spoken soul: The story of Black English. New York, NY: John Wiley & Sons.

Santa Ana, O., & Parodi, C. (1998). Modeling the speech community: Configuration and variable types in the Mexican Spanish setting. Language in Society, 27(1), 23–51.

Saville-Troike, M. (2003). The ethnography of communication: An introduction (3rd ed.). Malden, MA: Blackwell.

Shuck, G. (2006). Racializing the nonnative English speaker. Journal of Language, Identity & Education, 5(4), 259–276. doi:10.1207/s15327701jlie0504_1

Silverstein, M. (1976). Shifters, linguistic categories, and cultural description. In K. Basso & H. Selby (Eds.), Meaning in anthropology (pp. 11–55). Albuquerque, NM: University of New Mexico Press.

Smith, A., & Anderson, M. (2018, March 1). Social media use in 2018. Retrieved July 10, 2018 from the Pew Research Center: Internet, Science & Tech website: http://www.pewinternet.org/2018/03/01/social-media-use-in-2018/

Tagliamonte, S. A. (2016). So sick or so cool? The language of youth on the internet. Language in Society, 45(01), 1–32. doi:10.1017/S0047404515000780

Weldon, T. L. (1994). Variability in negation in African American Vernacular English. Language Variation and Change, 6(3), 359–397.

Wolford, T., & Evanini, K. (2006). Features of AAVE as features of PRE: A study of adolescents in Philadelphia. University of Pennsylvania Working Papers in Linguistics, 12(2), 231-244.

Wolfram, W. (1969). A sociolinguistic description of Detroit negro speech. Urban Language Series, No. 5. Washington, D.C.: Center for Applied Linguistics.

Wolfram, W., & Schilling-Estes, N. (2006). American English: Dialects and variation. Malden, MA: Blackwell Publishing Ltd.

Wolfram, W., & Thomas, E. (2008). Development of African American English (1st ed.). Retrieved from https://trove.nla.gov.au/work/28486040

Woolard, K. A., & Schieffelin, B. B. (1994). Language ideology. Annual Review of Anthropology, 23, 55–82.

Zentella, A. C. (1997). Growing up bilingual: Puerto Rican children in New York. Malden, MA: Blackwell.

Zhang, Q. (2005). A Chinese yuppie in Beijing: Phonological variation and the construction of a new professional identity. Language in Society, 34(03). doi:10.1017/S0047404505050153

Biographical Note

M. Sidury Christiansen [marthasidury.christiansen@utsa.edu] is an associate professor of Applied Linguistics/TESL in the Department of Bicultural Bilingual Studies at the University of Texas at San Antonio. Her research focuses on the study of transnational bilinguals' engagement with each other in digital spaces, exploring the intersection between literacy, language ideologies, identities, and culture.


Any party may pass on this Work by electronic means and make it available for download under the terms and conditions of the Digital Peer Publishing License. The text of the license may be accessed and retrieved at http://www.dipp.nrw.de/lizenzen/dppl/dppl/DPPL_v2_en_06-2004.html.