Introduction

The sociolinguistic landscape of Mexican-American (MexAm) communities in Los Angeles (L.A.) reflects a dynamic interplay of heritage language maintenance, linguistic innovation, and language shift. With over 37 million people of Mexican origin in the United States (Pew Research Center, 2023 accessed 29 January, 2025¹), MexAm communities offer critical insights into bilingual development, particularly the simultaneous acquisition of heritage Spanish (HerS) and Chicano English (ChE). While HerS has been the focus of extensive research (Montrul 2016; Polinsky 2018; Polinsky and Scontras 2019; Potowski 2018; Guardado 2020; Silva-Corvalán 2014; Lynch 2003 etc.), the development of ChE, a stable, contact-born dialect, remains understudied, especially in young children (Bayley and Santa Ana 2004). This paper examines the bilingual trajectory of a third-generation (3G) child acquiring ChE and HerS in L.A., highlighting how language mixing functions both as a communicative strategy and as a cultural marker of identity within her multigenerational family.

Heritage Spanish and Language Shift

HerS in MexAm families serves as a vital linguistic and cultural link across generations. However, in the U.S. context, particularly in English-dominant cities like L.A., intergenerational transmission of HerS is increasingly fragile. As Alvarez notes (2024), inconsistent use of Spanish at home, paired with discontinued institutional support at school, contributes to this erosion. These conditions are compounded by ambiguous language policies that have historically subordinated languages other than English (Silva-Corvalán 2014).

This dynamic is especially visible among 3G speakers. Language shift is well documented in the 3G, who often exhibit reduced exposure to and proficiency in HerS due to diminished input across home, school, and community contexts (Bustamante-López 2008; Pauwels 2016; Alvarez 2023). Still, the so-called “3G rule” must be reviewed with nuance: “a continuum with blurry boundaries, not necessarily as an inevitable fate in every case” (Field 2011: 62). This perspective aligns with a plurilingual view of bilingual development, where linguistic abilities are partial, fluid, and context-dependent (Piccardo 2016).

Indeed, many 3G MexAm children fall along a bilingual continuum (Valdés and Fugueroa 1994), with varying degrees of competence in HerS. Sevinç and Backus (2017) show that reduced use of a heritage language can diminish confidence, which in turn limits further use, an example of the recursive nature of language shift. In this view, language proficiency defined as the “apparent ease of performance that is manifested by the proficient bilingual” (Palij and Homel 1987) reflects both developmental exposure and sociocultural conditions, and HerS decline co-occurs with the rise of English dominance and emergent varieties like ChE.

Chicano English: A Product of Contact and Identity

To situate ChE, it is useful to recall that “Chicano” refers to people of Mexican heritage in the U.S. who maintain strong cultural ties and a salient ethnic identity (Arce 1981). Within these communities, ChE is spoken by those who acquired English as a first language, simultaneously with Spanish, when they enrolled in elementary school, usually around age 5 (Arce 1981: 177). In our study, the target child acquired both English and Spanish from birth within a cultural context where ties to Mexican heritage remain strong.

ChE is a distinct, autonomous dialect of English, reflecting the cultural and linguistic hybridity of MexAm communities. It is often misperceived as accented English spoken by second-generation (2G) Latinos, yet research shows it is a fully developed variety that persists independent of bilingualism or Spanish proficiency (Mendoza-Denton 1999; Santa Ana and Bayley 2008). The children of these 2G speakers thus belong to the 3G, such as the target-child in our study.

Spanish Influence and Language Mixing in Chicano English: Linguistic and Identity Dimensions

The influence of Spanish on ChE is visible across phonological, lexical, and syntactic levels. Notable features include its syllable-timed rhythm and apico-dental articulation of alveolar stops, diverging from mainstream American English (Santa Ana and Bayley 2008). Lexical borrowings, for example barely used for just recently, reflect transfers from Spanish apenas (Fought 2003). Syntactic structures like double negation (i.e., “I don’t want nothing”) parallel both Spanish syntax and patterns common in African-American English (AAVE) (Fought 1999; Santa Ana 1993). These innovations point to ChE as a multifaceted contact variety shaped by regional and social influences.

Generational variation also affects the extent of Spanish influence. Older speakers often exhibit more direct transfers from Spanish, while younger 3G speakers, many of whom are English-dominant, show reduced influence yet maintain distinct ChE features (Fought 2003). Even monolingual ChE speakers as Mendoza-Denton (1999) argues, may produce features like consonant cluster reduction and the use of habitual be, underscoring the dialect’s autonomy from active Spanish proficiency. ChE thus functions as a stable, dynamic variety, emerging from but no longer reliant on bilingualism.

A key bilingual practice intertwined with ChE is language mixing. For the purposes of this study, we adopt a broad definition of language mixing defined as “the use of two language varieties in the same conversation. It can occur between speakers, or between sentences in the same speaker’s turn, or within a sentence” (Myers-Scotton, 2006: 161). While earlier work distinguishes code-switching (a term traced back to Vogt 1954), and code-mixing (Alvarez-Cáccamo 1998; Auer and Eastman 2010), we treat mixing holistically. These practices, frequent in heritage bilingual settings, are socially patterned and interactionally meaningful, rather than random alternations. In the present study, language mixing surfaced as a communicative strategy and as a marker of identity, shaped by shifting proficiencies and intergenerational dynamics.

Language Mixing in Identity and its Functional and Symbolic Dimensions

Beyond structural features, language mixing plays both functional and symbolic roles in bilingual communities. Functionally, it enables communication across generations with differing language proficiencies. In our study, for example, the target-child primarily used ChE, while older relatives favored Spanish. Language mixing thus served as a bridge across linguistic and generational divides (Kiaer 2023; Alvarez 2023).

Symbolically, language mixing indexes cultural hybridity and ongoing identity negotiation. As Alvarez (2023) notes, 2G and 3G speakers often mix languages to maintain heritage ties while navigating English-dominant environments. These practices reflect the lived realities of MexAm communities, where cultural affiliation and linguistic behavior intersect. Rather than indicating confusion or loss, they reflect bilingual agency, the ability to draw strategically on one’s full linguistic repertoire to express alignment, resistance, or belonging.

This perspective aligns with a plurilingual approach, which accounts for partial, context-dependent competencies observed in heritage speakers rather than treating languages as discreet systems (Piccardo 2016). Plurilingualism emphasizes how individuals use multiple linguistic resources fluidly across contexts, often without full mastery in each language contrary to the notion of multilingualism, which tends to treat languages as fully separate systems (Steil and Carrasco 2023). In this view, the child’s mixed-language utterances are not transitional but constitute a dynamic, integrated communicative repertoire.

Language mixing, then, supports both bilingual communication, and cultural expression. As Santa Ana (1993) argued, it allows speakers to assert their dual identity as members of both the MexAm community and broader American society. Following Bustamante-López (2008), we see language mixing not as a byproduct of incomplete bilingualism, but a core feature of Chicano identity. ChE emerges as more than a contact dialect; it is a resilient, expressive mode of speech, reflecting sociolinguistic adaptation across generations in contemporary L.A.

Community Dynamics, Resistance, and the Role of Chicano English

Chicano communities are defined by shared ethnic identity, cultural continuity, and resistance to linguistic assimilation. As Arce (1981) notes, individuals of Mexican heritage who maintain strong cultural ties often develop distinct language practices such as ChE and language mixing, practices that serve as expressions of identity and resistance.

The Florence-Firestone neighborhood in South L.A. exemplifies this dynamic. With over 90% of the population identifying as Hispanic and high rates of Spanish use at home (L.A. County Planning 2019), the area operates as a bilingual contact zone. Yet, within this setting, English continues to gain ground, while Spanish transmission is increasingly disrupted. As such, the tension between language maintenance and shift is experienced in everyday interactions.

In this context, ChE and language mixing become more than communicative tools, they act as symbolic markers. While “standard” English may be associated with assimilation (Eckert 2008), speaking ChE or Spanish, especially with non-standard features, can index resistance or authenticity, depending on audience and context (Anzaldúa 1987; Alvarez 2024).

These broader community-level tensions were mirrored in the multigenerational family observed in this study. Caregivers often used HerS or mixed speech to reinforce cultural ties, while the target-child consistently preferred English. Her language choices were not only shaped by input, but also by the sociolinguistic pressures surrounding her, choices that reflect ongoing negotiations of identity, belonging, and autonomy.

ChE and language mixing, then, function dually: as tools of adaptation to English-dominant realities and as resilient expressions of MexAm identity. Their persistence highlights both the pressure and the possibility of linguistic agency within a shifting sociolinguistic ecology.

Language Acquisition and Socialization in Bilingual Language Development

Language acquisition in bilingual contexts is deeply shaped by social interaction and cultural context. From a usage-based perspective (Tomasello 2003), language development is not innate, but emerges through exposure, frequency and interaction. Linguistic forms are built from input patterns, both direct caregiver speech and overheard interactions (Alvarez and Morgenstern 2024), which are culturally situated and socially mediated.

This theoretical lens considers cultural and linguistic variation in context, here, the interplay of ChE and HerS in one child’s communicative environment. These varieties are not isolated systems but components of a plurilingual repertoire, mobilized fluidly in a multilingual, multicultural ecology like L.A.

The language socialization paradigm (Ochs and Schieffelin 2006, 2011) complements this view, emphasizing how linguistic practices are embedded in cultural routines and shaped by ideologies, roles, and social positioning. Ochs and Schieffelin (2006) further argue that language mixing, and the use of each language separately, should be studied alongside the development of communicative competence. Indeed, language acquisition is “a fluid, unstable, and changeable process” (Guardado 2020: 39), especially in diverse family settings. Children’s linguistic and social competence thus evolves through community change, exposure, and their own agentive participation (Ochs and Schieffelin 2011).

In MexAm households, asymmetrical input patterns, where English dominates, often leads to receptive bilingualism (Romaine 1989; Montrul 2016): children understand HerS but may resist producing it (Alvarez and Morgenstern 2024). These patterns shape both language development and identity.

Taken together, usage-based and language socialization approaches reveal how bilingualism is co-constructed as the intersection of input, agency, and ecology. While ChE is recognized as a stable dialect, this study traces how it emerges precisely as HerS recedes in early childhood.

Research Questions

This study is guided by the following questions:

1. How does HerS influence the development of ChE in a 3G MexAm child, as evidenced by measures of mean length of utterance and type-token ratio?

2. What role does English-Spanish language mixing play in shaping bilingual language practices in the child’s family environment?

3. How do intergenerational family dynamics impact the child’s linguistic development and emerging language identity?

Methods & Material

The Florence-Firestone Community and Participant Profile

This study is based on longitudinal ethnographic fieldwork in Florence-Firestone, a bilingual contact zone in South L.A. The neighborhood is home to nearly 66,000 residents, over 90% of whom are of Mexican origin (L.A. Country Department of Regional Planning 2019). While Spanish is still widely spoken intergenerational transmission varies: 65% of residents are bilingual, but 20% speak only Spanish. In this context, English-Spanish bilingualism reflects both maintenance and shift, shaped by intergenerational transmission, family practices and broader societal pressures.

The focal participant was a 3G bilingual child, referred to as LIN (a pseudonym used in Alvarez 2023), observed from age 3;10 to 4;9. She belongs to a multigenerational, matrilineally organized MexAm family, with up to a ten regularly interacting members across three generations. Family communication occurred in English, Spanish, and mixed-language speech.

Data collection followed Ochs and Schieffelin’s (2011) linguistic anthropological paradigm, which views children’s language development as embedded in culturally specific communicative practices. This framework situates LIN’s bilingualism within her everyday socialization experiences, emphasizing the co-construction of language and identity in interaction.

LIN was selected based on prior research (Alvarez 2023) documenting receptive comprehension of HerS, but limited active production. Her profile, marked by asymmetric bilingualism and frequent family-based language mixing, offered a compelling opportunity to examine how heritage language input and interactional experience shape the emergence of ChE in early childhood.

As the daughter of ROX and MAR, LIN was a simultaneous bilingual exposed to English, Spanish, and mixed-language input from birth but had not yet begun formal schooling during the study. Prior findings (Alvarez 2020) showed her productive vocabulary was predominantly English by age 1;10. By the time of the present recordings, English remained her dominant language, especially in adult-directed speech.

Ethnographic Data Collection & Ethical Concerns

Ethnographic recordings were collected during naturalistic family activities such as meals, play, and informal conversations. These interactions offered rich insights into the family’s bilingual practices, with members alternating between English, Spanish, and mixed-language speech. Generational patterns emerged clearly: older members tended to favor Spanish; younger ones primarily used English, and often incorporating Spanish elements.

These dynamics align with Tang’s (2009: 67) observation that “the second or later generations of Mexican-Americans, who speak English since they were born and were educated in the USA, might also use Spanish at home or within their community”. In the present study, language mixing was frequent and socially meaningful, particularly among LIN and her cousins.

Following established traditions in bilingualism research (Wei 2000; De Houwer 2009; Gardner-Chloros 2009), this study adopts an insider ethnographic perspective. The researcher’s familial embeddedness provided access to nuanced sociolinguistic behaviors. To mitigate potential bias, analytic rigor was maintained through systematic transcription protocols, video triangulation, and peer debriefing. The study thus follows a long-standing tradition of documenting heritage language use within researchers’ own families (Kabuto 2010; Kiaer 2023; Kouritzin 2000; Ronjat 1913).

This work also responds to calls for greater linguistic equity by centering the lived experiences of heritage bilingual families, groups often marginalized in U.S. research and education policy (Ortega 2019a; Santa Ana 2004). Ethical protocols included bilingual parent consent and withdrawal rights. The study received full approval from the Université Sorbonne Nouvelle Ethics Committee and the Los Angeles Unified School District’s Committee for External Review (LAUSD, Project 384, CERR).

Data Analysis

Video data were transcribed using CLAN2 (Computerized Language Analysis), following CHAT conventions, the international standard for human speech transcription (De Houwer, 2009). The software enabled detailed annotation of language choice and calculation of mean length of utterance (MLU), and type-token ration (TTR) in English and Spanish. MLU serves as a developmental measure of syntactic complexity, while TTR reflects lexical diversity. As MacWhinney (2008: 8) explains, CLAN “emphasizes the automatic computation of indices such as MLU” and performs analyses “based on the ratio of different words (Types) to the total number of words (Tokens)” (MacWhinney 2000: 110). While useful, these metrics must be interpreted with caution (Crystal 1974; Alvarez, 2023), even if “MLU is generally interpreted as a developmental index of language proficiency” (Silva-Corvalán 2014: 45).

Transcription yielded a plurilingual corpus of approximately 30,000 utterances, comprising English, Spanish, and language mixing. A plurilingual corpus, as defined by Léglise and Alby (2013), features multiple languages within the same transcript, including code-switching or other kinds of language mixing. This approach is essential in bilingual acquisition research, where speakers fluidly alternate languages depending on communicative needs (Grosjean 2015). Relying on monolingual transcripts would thus obscure such dynamics and fail to capture language mixing as a key developmental feature.

Field notes, taken both digitally and manually, provided contextual insights into sociolinguistic routines and attitudes. Totaling over 17,000 words, these notes documented family discourse not always evident on video. For example, they revealed subtle yet consistent negative stances toward Spanish, echoing broader patterns of language shift (Alvarez 2024). As Fought (2003) observes, such attitudes have contributed to Spanish loss and the rise of English monolingualism among Chicanos, in part due to the prestige asymmetry between English and Spanish in the U.S.

Analytical Techniques

This study used a mixed methods design (Stivers 2015) combining descriptive statistics with qualitative ethnographic analysis to explore how individual, familial, and environmental practices shape bilingual development. In bilingual acquisition research, such integration is effective. For example, Beaupoil-Hourdel and Morgenstern (2021) demonstrate its value in a cross-linguistic study of French and British children’s shrugs. Following Creswell and Plano Clark (2018), this study pairs statistical measures with ethnographic interpretation to offer a multidimensional account of the 3G bilingual experience.

Quantitative Analysis

Descriptive statistics focused on LIN’s MLU and TTR, tracking her ChE development from 3;10 to 4;9. MLU served as a baseline indicator of syntactic growth (Brown 1973; Miller 1981), while TTR reflected lexical diversity. These metrics helped contextualize LIN’s developmental path within a bilingual, bicultural ecology.

Data were also processed using Excel, using 14 sociolinguistic categories. The coding was kept simple in both quantity and quality following Ochs (1979). Key codes included language choice (English, Spanish, mixed), input type (child-directed vs. overheard), and speaker generation (1G, 1.5G, 2G, 3G). Findings showed a decline in Spanish input from 50% to 19-22%, while English rose from 43% to about 69-72%, highlighting an ongoing shift toward English dominance. These patterns reflect broader 3G HerS shift and the increasing prominence of ChE as LIN’s expressive variety.

Qualitative Analysis

Ethnographic analyses used “thick” description (Geertz 1973) to interpret language use within her sociocultural environment. This approach allowed for layered readings of recurring interactional data, emphasizing how bilingual and bicultural practices shape meaning.

Focus was placed on key episodes of language mixing, language choice, and explicit language socialization. These interactions were analyzed for their communicative functions, i.e., how mixing sustained family cohesion, indexed identity, or marked a shift from HerS to ChE. Such practices were viewed not just as linguistic strategies but as acts of cultural agency within a community navigating continuity and change.

Results: Descriptive Statistics

The table below presents LIN’s MLU across three sampling periods. At age 3;10 (February 1, 2018), her MLU was 2.8 based on a sample of 80 utterances. While slightly below the expected 2.5-3.0 range for children approaching age 4 (De Houwer 2009), other samples showed higher scores, up to 4.4, indicating active sentence production in English.

These results suggest that LIN’s overall MLU falls just below the developmental expectations, possibly reflecting asymmetric bilingual input and frequent language mixing in her environment. The variability across samples may point to early effects of bilingual dominance and reduced expressive use of Spanish in shaping her linguistic trajectory.

Table : MLU Values English & Spanish

Age	Media file	number of utterances	MLU English	number of utterances	MLU Spanish
3;10	FEB_1_2018_ERI_LIN_JUL_ROX.cha	80	2.8	-	-
	FEB_14_2018_ROX_LIN_ERI_RIC.cha	64	4.3	-	-
	FEB_15_2018_GLO_JUL_LIN_ERI.cha	58	3.5	-	-
	FEB_21_2018_GRC_ROX_ERI_LIN.cha	62	3.0	-	-
	FEB_22_2018_ERI_LIN_JUL_GRC.cha	82	3.5	-	-
	FEB_22_2018_LIN_ROX_JUL_RIC_GRC.cha	59	4.0	-	-
	FEB_22_2018_ROX_GRT_GRC_ERI_LIN_ALE.cha	53	3.3	-	-
	FEB_23_2018_ROX_LIN_JUL_GRC.cha	66	3.4	-	-
	FEB_24_2018_LIN_ERI_GLO.cha	91	2.9	-	-
4;4	AUG_26_2018_GLO_ROX_GUI_ALE_ERI_LIN_RIC.cha	70	3.7	-	-
	AUG_26_2018_LIN_GLO_JUL_ALE_GOY_ERI_GUI.cha	57	3.9	-	-
	AUG2018_LIN_JUL_ERI.cha	92	4.1	-	-
	AUG2018_LIN_ERI.cha	56	3.6	-	-
	AUG_30_2018_ERI_LIN_GLO_JUL_ROX_JUL.cha	75	3.8	-	-
	AUG_30_2018_LIN_ROX_ALE_ERI_GLO_JUL_RIC_MAR.cha	63	3.4	-	-
4;9	JAN_01_2019_LIN_ALE_ROX_GLO_GUI_MAR.cha	116	4.0	-	-
	JAN_02_2019_LIN_JUL_ERI_ROX.cha	86	3.6	-	-
	JAN_03_2019_ERI_LIN_GOY_RIC_ALE_ROX.cha	101	3.7	-	-
	JAN_04_2019_JUL_GLO_ERI_LIN_ROX.cha	85	3.5	-	-
	JAN_05_2019_GLO_ROX_JUL_ERI_LIN.cha	90	4.2	-	-
	JAN_06_2019_GLO_ROX_ERI_GUI_LIN_JUL_RIC.cha	62	4.3	-	-
	JAN_06_2019_GUI_ERI_GLO_RIC_JUL_ALE_LIN.cha	83	3.9	-	-
	JAN_09_2019_JUL_ERI_ROX_GLO_LIN_GRC.cha	62	4.2	-	-
	JAN_09_2019_LIN_GLO_ERI_ROX_MAR.cha	102	3.4	-	-
	JAN_11_2019_LIN_ROX_GRC_ERI_JUL.cha	284	4.3	-	-
	JAN_11_2019_ROX_LIN_JUL_GRC_RIC_ERI.cha	221	4.3	-	-
	JAN_12_2019_GLO_ARA_ALE_LIN_JUL_GOY_ROX.cha	76	3.8	-	-

Following MacWhinney³ (2000) and Silva-Corvalán (2014), MLU was calculated only for video segments where LIN produced at least 50 utterances, ensuring statistical reliability. Since CLAN does not yet support automatic MLU computation for language mixing, mixed-language utterances were excluded from calculations. They were instead coded separately and analyzed qualitatively. Their exclusion reflects a current technical limitation, not an analytical oversight.

Notably, LIN never produced 50 Spanish-only utterances in any sampling period. The absence of MLU data for Spanish (noted “-” in the table) points to her limited expressive use of HerS and reinforces patterns of receptive bilingualism.

The table below synthesis LIN’s average English MLU across three points, offering a longitudinal snapshot of her language development. These values, when read alongside input patterns and language choice trends, help contextualize LIN’s evolving expressive repertoire within a bilingual, bicultural setting.

Table : LIN’s longitudinal development of English MLU versus Expected MLUw

LIN’s age in: Month, Year, Age	Average MLU in English longitudinally per sampling period	Expected MLUw according to Brown’s (1973) Stage based on MLU
February 2018= 3;10	3.4	3.75-4.5
August 2018= 4;4	3.7	3.75-4.5
January 2019= 4;9	3.9	3.75-4.5

MLU is a widely used measure in child language development research, indicating the average number of morphemes per utterance and serving as an index for grammatical complexity. It typically increases with age in monolingual children and helps identify developmental stages.

Debate continues over the best unit of measurement, morphemes (MLUm) or words (MLUw). De Houwer (2009), Johnson (2001), and Silva-Corvalán and Sánchez-Walker (2007) argue that word-level comparisons (MLUw) remain suitable for English and Spanish, despite the complexity of cross-linguistic comparisons. This study adopts MLUw (hereafter MLU) for consistency.

According to Brown (1973), children aged 3;5 to 3;10 typically reach Stage V, producing multi-clause utterances with MLUm values between 3.75 and 4.5. In bilinguals, however, MLU values in each language are often slightly lower due to factors such as input richness, language dominance, and exposure. Still, overall development tends to follow a similar trajectory as monolinguals.

LIN’s English MLU rose modestly across the three sampling periods, from 3.4 at 3;10 to 3.9 at 4;9, indicating steady development, albeit slightly below monolingual norms. This likely reflects both English dominance and bilingual dynamics. Her Spanish output, by contrast, was insufficient for MLU analyses. Yet as A. Morgenstern (personal communication, April 20, 2021) notes, MLU alone is insufficient to assess bilingual development. To deepen our understanding, we turn to an analysis of her lexical diversity.

Vocabulary diversity measures are necessary and widely used in child language research (MacWhinney 2000; Malvern and Richards 2002). Accordingly, we examine LIN’s TTR as a complementary measure.

As with MLU the low quantity of Spanish output made her HerS TTRs scores unreliable. Therefore, the following analysis focuses on LIN’s English TTRs, calculated from the same video segments. On February 1, 2018, for instance her TTR was 0.449, offering a baseline for tracking lexical development over time.

Table : TTR Values: English for LIN & Spanish & English

Age	Media file	Speaker
		LIN
		TTR Eng.	TTR Spa.
3;10	FEB_1_2018	0.449	-
	FEB_14_2018	0.331	-
	FEB_15_2018	0.371	-
	FEB_21_2018	0.351	-
	FEB_22_2018	0.398	-
	FEB_22_2018	0.423	-
	FEB_22_2018	0.447	-
	FEB_23_2018	0.317	-
	FEB_24_2018	0.388	-
4;4	AUG_26_2018	0.402	-
	AUG_26_2018	0.476	-
	AUG2018	0.318	-
	AUG2018	0.455	-
	AUG_30_2018	0.391	-
	AUG_30_2018	0.442	-
4;9	JAN_01_2019	0.332	-
	JAN_02_2019	0.344	-
	JAN_03_2019	0.315	-
	JAN_04_2019	0.420	-
	JAN_05_2019	0.300	-
	JAN_06_2019	0.463	-
	JAN_06_2019	0.354	-
	JAN_09_2019	0.370	-
	JAN_09_2019	0.373	-
	JAN_11_2019	0.205	-
	JAN_11_2019	0.220	-
	JAN_12_2019	0.389	-

LIN’s English TTR values ranged from 0.32 to 0.47, with occasional peaks near 0.48. While indicating modest lexical diversity, these scores should be interpreted with caution due to sample size variability. As MacWhinney (2000: 72) notes, TTR is sensitive to sample size and “can only be used to compare samples of equivalent size”. This limitation affected the February 2018 and August 2018 samples.

Repetition also impacts TTR. A speech sample with 20 unique words used once would yield an “ideal” TTR of 1.000 (Richards 1987), but reflects a monolingual standard, often experimental norm. In contrast, LIN’s spontaneous speech within a bilingual environment shows that lexical development cannot be judged according to monolingual benchmarks. Her speech, mostly in English with some mixing, reflects a dynamic and functionally diverse vocabulary.

As noted earlier, LIN’s Spanish output was too sparse for valid MLU or TTR calculation. Her English MLU rose gradually, though it remained slightly below expected developmental norms. Her English TTR, by contrast, remained relatively stable, consistent with vocabulary growth.

These findings align with established trajectories for 3G heritage speakers, who often shift toward English dominance. LIN’s lexical and syntactic development highlights how bilingual children’s practices challenge monolingual benchmarks and expand our understanding of typical development in language-minoritized contexts.

Results: “Thick” Descriptions

The following excerpt offers an in-depth analysis of a key interaction within the multigenerational MexAm family under study. This “thick” description (Geertz 1973) illustrates how bilingual development unfolds in real time. It shows how LIN navigates language ideologies, enacts bilingual agency, and responds to shifting patterns of dominance and identity.

The excerpt highlights the interplay of ChE and HerS in LIN’s communicative environment and serves as a window into language shift, and emerging linguistic identity with a 3G bilingual ecology.

For reasons of length, this article focuses on a single illustrative episode: the “Sí spañol” interaction.

To guide the reader, transcription conventions are summarized below:

Table : Transcription conventions used in the excerpts below

Symbol	Meaning
@s	Word-final indicator of language mixing (e.g., “spañol@s”)
[- spa]	Spanish utterance (English translations follows)
[- mix]	Mixed language utterance
xxx	Unintelligible word(s)

“Sí spañol”

This example analyzes LIN’s emerging ChE through informal family interaction, highlighting how her bilingual practices take shape in naturally occurring English-Spanish contact settings.

While other examples (Alvarez 2023) underscore LIN resisting HerS, here she engages with Spanish on her own terms, supported by family scaffolding. Rather than reversing language shift, this moment reflects a plurilingual repertoire in development, one where partial HerS competence coexists with dominant ChE use.

LIN’s strategic language mixing illustrates broader MexAm dynamics: ChE is not merely the product of Spanish loss but emerges from the interplay of accommodation and identity negotiation across generations.

Figure
Ex. 4: JAN_11_2019_ROX_LIN_JUL_GRC_RIC_ERI_sí spañol_A

LIN is 4;9

*ERI: so, she wants to come up on your phone. @ROX *ROX: she swears. @ERI *ROX: but would she do that? @ERI *ROX: is the question. @ERI *ROX: no. @ERI *LIN: she actually not do that sorry Eric. *ERI: [- spa] LIN tú qué sabes tú no hablas español. LIN what do you know you don’t speak Spanish. *LIN: [- spa] sí. @ERI Yes. *ERI: [- spa] sí qué? @LIN Yes what? *LIN: [- spa] sí xxx [: hablo] spañol [: español]. @ERI Yes I speak Spanish. *ROX: like what sense would that make? @ERI *ERI: [- spa] qué? @LIN What? *LIN: umum. @ERI *ROX: Eric +… *ERI: [- spa] LIN tú no hablas español. LIN you don’t speak Spanish. *ROX: Eric. *LIN: [- spa] sí. @ERI Yes. *ERI: [- spa] sí qué? @LIN Yes what? *JUL: xxx. *LIN: [- spa] sí xxx [: hablo] spañol [: español]. @ERI Yes I speak Spanish. *ERI: sí hablas español? @LIN Yes, you do speak Spanish? *LIN: [- spa] uno dos tres uno +… @ERI One two three one +… *ROX: [- spa] cuatro cinco seis siete +… @LIN Four five six seven +… *ERI: [- spa] sabes contar? @LIN You know how to count? *LIN: [- spa] cuatro cinco seis +… @ERI Four five six +… *ERI: [- spa] y qué más? @LIN And what else? *ROX: [- mix] I like it when she says Julian ven@s. @ERI I like it when she says Julian come here. *ERI: [- spa] y qué más? @LIN And what else? *LIN: [- spa] dulce. @ERI Candy. *LIN: [- spa] dulces. Candies. *ROX: [- spa] dulces. @LIN Candies.

When GRC leaves the table, ROX, ERI, JUL, and LIN remain. ERI, absorbed on his phone, references GRC’s earlier desire for ROX’s phone (l.1). As he finishes, ERI glances at ROX, who immediately responds (l.2), continuing with a rhetorical question and emphatic “no” (l.3 – l.5). ERI remains visually and physically disengaged.

LIN (l.6), without direct eye contact, joins in: “she actually not do that sorry ERI”, shifting the interaction from dyad to triad. LIN’s timely interjection reflects emergent communicative competence: as an overhearer, she inserts herself in a socially appropriate way. This aligns with Duranti’s (1997: 27) view that communicative competence includes “certain patterns of thought, ways of understanding the world, making inferences and predictions”, evident in LIN’s contribution.

At this point, the exchange has been in English, until ERI (l.7) abrubtly switches to Spanish : “¿LIN tú qué sabes tú no hablas español ?” (“LIN what do you know you don’t speak Spanish ?”). Eye contact comes only at the end, subtly challenging her, and positioning Spanish as the threshold to full participation.

Instead of retreating, LIN (l.8) meets his gaze and responds in Spanish : “Sí, spañol”, reclaiming her bilingual agency and shifting the interaction to identity negotation. This moment marks a shift in LIN’s linguistic behavior : instead of rejecting Spanish, she uses it strategically to challenge ERI’s framing and to assert her place in the bilingual interaction. The figure below captures the moment LIN looks at ERI to affirm her Spanish competence, a turning point in the exchange that signals her active negotiation of bilingual identity.

Figure : LIN looks at ERI to affirm that she speaks Spanish

As LIN asserts her Spanish ability, ROX remains focused on preparing the goo4, avoiding eye contact and offering no response. ERI (l.9), still on his phone, pushes with a low effort but pointed : “¿Sí qué ?” (“Yes what ?”), prompting LIN elaborate.

LIN responds in Spanish but, distracted by picking up a toy avoids eye contact. Her effort to say español reveals both cognitive strain and linguistic effort.

As LIN hesitates, ROX (l.11) briefly returns to her topic, but ERI ignores this shift and maintains focus on LIN (l.12), now inviting rather than teasing. This transition signals a shift in tone, from playful provocation to genuine prompting, and opens a space for LIN to assert her bilingual abilities. The figure below captures the moment ERI lifts his gaze from his phone to address LIN directly, visually reinforcing his invitation to continue speaking Spanish and advancing the interaction toward a deeper engagement with HerS.

Figure : ERI gazes at LIN asking her to repeat herself

LIN (l.13) looks directly at ERI, challenging his claim that she does not speak Spanish. ERI (l.15) teases her again, but LIN (l.17), maintaining eye contact, responds firmly: “Sí” (“Yes”). When ERI (l.18) presses with: “¿Sí qué?” (“Yes what?”), LIN (l.20) responds: “Sí xxx [: hablo] spañol” (“Yes, I xxx [: speak] Spanish”), though she hesitates and repeats the phrase, showing both effort and determination.

ERI (l.21) prompts again. LIN nods and (l.22) holding his gaze begins to count in Spanish, offering proof of her skills. This shift from teasing to demonstration marks a turning point, LIN engages with HerS actively, on her own terms.

After counting to “three”, LIN pauses, her gaze drifting as she struggles to recall “four” in Spanish. Unable to proceed, she restarts from “one” while ducking under the table, a gesture likely signaling embarrassment or self-consciousness. This marks a clear shift: the confidence she showed moments earlier gives way to the limits of her productive competence.

Her gaze tracks this change, first searching upward, then lowering as she physically withdraws. The retreat suggests an emerging awareness of the gap between her claimed ability and actual recall.

Yet LIN persists. Rather than disengage, she recalibrates, showing both effort and vulnerability. This moment underscores her growing metalinguistic insight and social pressures shaping heritage language use. The following figures illustrate the trajectory of LIN’s gaze. The first captures her looking up into space, visibly searching for the next word. The second shows her ducking under the table, restarting her count from “one”, visually anchoring her moment of difficulty and recovery.

Figure : LIN’s gaze wanders as she tries to recall the next number in Spanish

Figure : LIN hides behind the table as she starts counting with number one again

ROX (l.23) without looking at LIN, begins modeling the next numbers in Spanish. Now fully immersed in a Spanish-language frame, LIN looks to ROX for support and repeats the modeled numbers while gazing into space. ERI (l.24) asks LIN if she knows how to count in Spanish, prompting further engagement.

After completing the count, ERI (l.26) asks what else she can say in Spanish. LIN (l.29) responds with: “dulce” (“candy”) then adds: “dulces” (“candies”), turning to ERI with growing confidence, signaling a shift from hesitation to active participation.

LIN’s responses (l.26 and l.28) to ERI’s prompts are contextually appropriate and increasing clear. With ROX’s encouragement (l.27), she shows both receptive bilingualism and a willingness to engage with HerS when scaffolded by trusted interlocutors. Rather than resistance versus acceptance, this moment highlights a dynamic process: LIN’s selective engagement with Spanish reflects how 3G speakers navigate bilingualism. While ChE remains her dominant code, voluntary HerS use supports her bilingual identity and deepens her integration into the family’s plurilingual ecology.

Discussion

The study explored the bilingual development of LIN, a 3G MexAm child in a multigenerational L.A. family, focusing on how HerS influences ChE acquisition, how language mixing shapes bilingual practices, and how familial dynamics shape linguistic identity. Through descriptive and ethnographic analysis, we offered a nuanced examination of LIN’s bilingual trajectory.

Rather than framing LIN’s development through a deficit lens, where Spanish loss signals failure, we emphasized LIN’s agentive negotiation of bilingual resources, via language mixing, partial competencies, and identity-making speech acts. LIN’s development underscores a plurilingual trajectory: she selectively draws on HerS and ChE, integrating both into a dynamic communicative system shaped by familial support. This challenges rigid ideals of “balanced bilingualism” and foregrounds the fluid, hybrid, and socio-affective realities of real-world bilingualism.

A key finding was the role of HerS in ChE development. While LIN demonstrated receptive bilingualism, her productive use of HerS was minimal, suggesting a trajectory of language shift common among 3G heritage speakers. Montrul (2022), for example, identifies the omission of the Spanish direct object marker ‘a’. Yet LIN did not simply replace Spanish with monolingual English; she developed a robust ChE repertoire. This aligns with Bayley and Santa Ana (2004) who argue that structural features of ChE may reflect underlying influence from Spanish, especially in its grammar.

Moreover, LIN’s MLU and TTR results provide insight into her expressive development in ChE. While her TTR scores fall within age-expected norms, her MLU remained slightly below developmental benchmarks. This could relate to the linguistic dynamics of her home environment, whereby her main playmates, her younger sibling and two cousins, were all diagnosed with speech impairments during the study. However, our focus on the speech patterns captured the data.

LIN did not produce enough Spanish-only utterances for reliable MLU or TTR scores, underscoring her limited active use of HerS. Nonetheless, in specific interactions, she engaged with Spanish in ways suggesting it still forms part of her latent repertoire. Indeed, “Chicano communities (may) create a new variety (of language) through which they may express their identity” (Bayley and Santa Ana 2004: 181). In LIN’s case, ChE, HerS, and language mixing all contributed to her evolving bilingual identity, enabling to express stance, navigate meaning, and index belonging in her family’s semiotic ecology.

Language mixing emerged as a particularly salient strategy. It supported intergenerational communication, especially with elders who primarily used Spanish, and symbolically indexed LIN’s bicultural identity. Drawing on Duranti’s (1997: 24) view of culture as transmitted through everyday linguistic interactions, LIN’s patterns show how ChE, HerS, and language mixing co-exist dynamically. Though English dominates in her family and the wider U.S. context, Spanish persists in affective and symbolic ways, shaping LIN’s communicative world.

The “thick” interactional analyses, particularly “Sí spañol”, captured LIN’s bilingual agency in real time. In this episode, she momentarily embraced Spanish when prompted, reflecting her ability to navigate context-sensitive language expectations. Rather than viewing such moments as either resistance or acceptance, they reflect a dynamic bilingual repertoire shaped by both family input and broader sociocultural pressures.

Indeed, these micro-level practices are situated within macro-level ideologies, including family attitudes, institutional norms, and public discourses (Alvarez 2024). LIN’s responses show how children not only absorb but also reshape these ideologies in practice, sometimes complying, other times pushing back, and often doing both simultaneously.

Her caregivers primarily reinforced English, while older family members attempted to maintain HerS through modeling and prompts. Yet the sociolinguistic context of L.A., where English is socially and institutionally dominant, drives a broader shift away from Spanish. Indeed, “a number of contextual factors play a crucial role in the development of different types of bilingualism. Among them … their social status … and family and community attitudes toward each of the languages and toward bilingualism.” (Silva-Corvalán 2014: 22). Our data showed Spanish declining over time and stabilizing around 19-22%, while English rose to 69-72%, a trend consistent with prior studies on intergenerational language shift (Bustamante-López 2008; Pauwels 2016).

Nonetheless, LIN’s development shows that bilingualism persists in adapted forms. Her case supports Piccardo’s (2016) notion of plurilingualism, where speakers draw on overlapping, context-dependent repertoires. LIN mobilizes English, Spanish, and language mixing not as discrete codes but as part of a fluid communicative strategy, one shaped by stance, affect, and interactional positioning.

The present study contributes to language socialization research by documenting how ChE and HerS unfold interactively within a multigenerational MexAm household. Following Ochs and Schieffelin (2011), we sought “an ethnographic sensibility that accounts for the socializing forces of these semiotic resources in terms of enduring and shifting sociocultural meaningful practices, events, situations … relationships etc.” (2011: 10-11). Our analyses thus emphasized the contextual and negotiated nature of linguistic identity. LIN’s case shows that children are not passive recipients of input but active participants in shaping their bilingual repertoires.

From a plurilingual perspective, LIN’s development highlights how expressive resources are combined creatively to navigate identity. Her preference for English and emergent ChE use reflect the evolving practices of 3G MexAm speakers, with ChE serving as a vehicle for maintaining cultural ties within English-dominant contexts. The findings carry implications for language education and heritage language maintenance. First, they reinforce the need to recognize ChE as a legitimate dialect. Programs should avoid deficit models and adopt pluralistic approaches that validate students’ full linguistic repertoires (Piccardo et al. 2021). Second, given the challenges of HerS maintenance, identity-affirming and comprehension-based strategies, such as intercomprehension, may better support children like LIN, who display strong receptive but limited productive skills. Curricula integrating ChE and HerS can promote linguistic ownership and pride, though such programs may not always ensure sustained multilingual development (Alvarez 2024; Nava 2009; Wolfram, Hudley and Valdés 2023).

By combining descriptive statistics with ethnographic analyses, the study bridges disciplinary approaches, what Ochs and Schieffelin (2008: 7) describe as one face oriented toward acquisition, and the other toward linguistic anthropology. While limited in scope, it offers a detailed portrait of bilingual development in context. Future research should examine larger cohorts of 3G bilinguals, follow them longitudinally, and explore how peer and school interactions shape ChE and HerS use since community institutions are crucial in reinforcing or discouraging linguistic practices (Alvarez 2024). A more comprehensive view must account for the intersecting ecologies: home, school, community, that jointly shape the language trajectories of U.S. Latino children.

Conclusion

Mexican-American bilinguals in Southern California navigate a plurilingual repertoire that spans English, Spanish, and mixed language use, a dynamic triad that reflects not only communicative need, but also cultural identity and social positioning (Bustamante-López 2008). This study highlights the importance of acknowledging Chicano English as a culturally embedded and linguistically robust variety, not a transitional or deficient form of English. LIN’s case illustrates how bilingual children actively construct bilingual identities through strategic engagement, using language mixing and ChE to assert belonging and navigate intergenerational expectations.

These findings reinforce the need for educational, and community initiatives that foster positive attitudes toward bilingualism, validate students’ full linguistic repertoires, and support heritage Spanish maintenance without pathologizing variation. Embracing a plurilingual perspective allows for more inclusive language policies that reflect the real-world complexity of Mexican-American communities. Future research should continue to explore how social, cultural, institutional forces, including family language ideologies, school environments, and policy discourses, shape the evolving bilingual identities of third-generation speakers and beyond.