In recent years the use of the term incomplete acquisition to refer to the grammatical development and outcome of heritage language acquisition (Benmamoun, Montrul, & Polinsky, Reference Benmamoun, Montrul and Polinsky2013; Montrul, Reference Montrul, Lefebvre, White and Jourdens2006, Reference Montrul2008; O’Grady et al., Reference O’Grady, Kwak, Lee and Lee2011; Polinsky, Reference Polinsky2006; Silva-Corvalán, Reference Silva-Corvalán2014) has been contested (Kupisch & Rothman, Reference Kupisch and Rothman2018; Otheguy, Reference Otheguy, Tortora, den Dikken, Montoya and O’Neill2016; Pascual y Cabo & Rothman, Reference Pascual y Cabo and Rothman2012). In their reflective piece, Domínguez, Hicks, and Slabakova (Reference Domínguez, Hicks and Slabakova2019) defend the use of the term as valid and accurate arguing that the notion of “incompleteness” has not raised brows when used to describe similar aspects of first and second language acquisition. As expressed in our most recent individual work (Montrul, Reference Montrul2018; Silva-Corvalán, Reference Silva-Corvalán, Shin and Erker2018) and in this commentary, we agree with Domínguez et al. that incomplete acquisition is equally valid and accurate to describe and justify the language acquisition patterns and outcomes we have observed in our own empirical work with many child and adult heritage speakers as well as adult immigrant input providers (the reference group) over the years. However, there are three points of disagreement with Domínguez et al. that we wish to comment on:

1. 1. Incomplete acquisition is not a process but a stage that can eventually turn into an outcome.

2. 2. The grammars of heritage speakers are not incomplete: certain aspects of their grammar do not fully develop under reduced input conditions when compared to the same domains in the input providers.

3. 3. The sociopolitical context is critical to understand the linguistic development of heritage languages.

Domínguez et al. remind us that “the term incomplete acquisition has come to describe a process of heritage language acquisition (distinct from attrition) which is often characterized by grammatical reductions, simplifications and reanalyses” (p. 242, our emphasis). In their view, incomplete acquisition is not a process but rather “one of the many possible outcomes of language acquisition” (p. 249). We agree that incompleteness is not a mechanism, but suggest that it may be considered to be an acquisitional outcome only if the grammatical features in question are shown to be stable in adults. Otherwise, we view incomplete acquisition as a descriptive term for a temporary stage in child language development, and in no way a causal factor (cf. Meisel, Reference Meisel, Enrique-Arias, Gutiérrez, Landa and Ocampo2014 for a critique of the idea of temporary stage). For example, Silva-Corvalán (Reference Silva-Corvalán2014, Reference Silva-Corvalán, Shin and Erker2018) has considered the issue of incompleteness by comparing bilingual children’s English and Spanish developing grammars; the more or less changed bilingual systems of adult second-generation immigrants; and the grammars of the input providers. She has shown that parallels between bilingual children’s linguistic behavior and that of adult heritage speakers imply that some aspects of the heritage language of adults, Spanish, are the outcome of an interrupted process of acquisition of this language between the ages of 3 and 5, when more intensive exposure to English reduces the amount and quality of exposure to the heritage language and diminishes the opportunities to use it. That is, during an age period when grammatical development is still ongoing, reduction of exposure to and use of the heritage language has led to a reduction or incomplete acquisition of grammatical domains compared to the same domains in the grammars of the adult models in the preceding generation. Features that are categorical in the input are lacking in the speech of adult second-generation bilinguals, such as the imperfect tense with stative verbs when the preterite is required; absence of the imperfect subjunctive and the compound tenses; and pragmatic constraints on the expression of coreferential subjects. For example, by ages 5 to 6, monolingual and bilingual children have not yet acquired the adult system of tense, mood, and aspect (TMA) in Spanish (Cuza & Miller, Reference Cuza, Miller, Klassen, Liceras and Valenzuela2015; Silva-Corvalán, Reference Silva-Corvalán2014). Notably, some features of bilingual children’s TMA system(s) are also characteristic of those of some adult second-generation speakers studied by Silva Corvalán (Reference Silva-Corvalán1994) and Montrul (Reference Montrul2002). These studies strongly support Domínguez et al.’s proposal that the term incomplete acquisition is justified when incompleteness is used to refer to grammar and not to individuals. Incompleteness may be an outcome in the case of adults and children when the amount of input is not sufficient to support further, and age-appropriate, grammatical growth.

Although some researchers (Domínguez et al., Reference Domínguez, Hicks and Slabakova2019; Otheguy, Reference Otheguy, Tortora, den Dikken, Montoya and O’Neill2016; Pascual y Cabo & Rothman, Reference Pascual y Cabo and Rothman2012; Polinsky, Reference Polinsky2006) have used the term incomplete to refer to heritage language grammars, we stress that specific aspects of a grammar can manifest incomplete development and can eventually fossilize, but not an entire grammar. Of course, this depends on how we describe “grammar”: As an entire cognitive system with a phonology, a lexicon, formal features, and a computational system (Chomsky, Reference Chomsky1981), or as a specific rule, such as the “grammar” that generates the past tense in English (Yang, Reference Yang2016)? We found in our own work that when several grammatical properties of the language of heritage speakers are analyzed in the same individuals, there is no incomplete acquisition across the board of their entire grammar. For example, Silva-Corvalán (Reference Silva-Corvalán2014) showed that while child and adult heritage speakers do not command the full range of tenses in Spanish, early on they do acquire the two copulas ser and estar. Montrul (2004) found that the object clitic system of heritage speakers of Spanish is strong, compared to their command of differential object marking, null/overt subject expression, or the preterite and imperfect contrast. And in a study of referring expressions in first mention contexts, Jia and Paradis (Reference Jia and Paradis2015) found that child heritage speakers of Mandarin (6- to 10-year-olds) showed incomplete acquisition of noun classifiers but their acquisition of relative clauses and postverbal noun phrases was on track. Therefore, we cannot claim that incomplete acquisition leads to an incomplete grammar, if grammar is understood as a full-fledged cognitive system. What we have found is incomplete acquisition of aspects of grammar. In this respect, incomplete acquisition in heritage grammars is no different from fossilization, the phenomenon of arrested development used to describe interlanguage systems in L2 acquisition. Although Selinker (Reference Selinker1972) claimed that entire interlanguages could fossilize, we know today that fossilization is local and specific (Lardiere, Reference Lardiere2007; Sharma, Reference Sharma2005) and might be better conceptualized as stabilization.

Finally, we take issue with the role of the social contexts in language acquisition and heritage language acquisition in particular. Domínguez et al. state that “[m]ore scrutiny on the sources of problems” (p. 243) is needed, but they argue for focusing “on I-language properties, just like research in formal linguistic theory” (p. 244), a goal that weakens the possibility to make “results impactful to society” (p. 243). They add that “[g]enerative research into L2 grammars does not, then, concern itself either with the socio-cultural context surrounding the act of learning or using a language, nor with the personal characteristics of its users (such as status, personality, education, motivation and so on)” (p. 244). Furthermore, they conclude that incomplete acquisition explains “why we find gaps in the grammatical knowledge of heritage languages” (p. 247). It is clear to us, however, that incomplete acquisition describes—it does not explain. While it is true that generative grammar does not take into account the social context of learning, we argue that the social context is critical to understand the language acquisition situation and ultimate attainment of heritage speakers. Aspects of the grammars of heritage speakers are incompletely acquired when input is insufficient, and reduced exposure to and use of the heritage language are very related to the minority status of the language: a language with lower availability, recognition, and prestige in the broader society. Thus, to search for an explanation it is imperative as well to examine such external factors as amount and quality of input, age of acquisition of the languages involved, and the personal characteristics of the speakers Domínguez et al. refer to and disregard. Domínguez et al. are quite correctly interested in addressing the terminology problem because language acquisition research (first, second, bilingual) is of value to society and should, therefore, not include terms that could have negative connotations. But both to have value for society and for a proper explanation of linguistic phenomena we believe that sociocultural variables must be taken into account.^{Footnote 1} In fact, social factors and input variation have begun to be explored in the acquisition of plural marking in Spanish, precisely because input variation leads to different acquisition schedules of particular linguistic variables (Miller & Schmitt, Reference Miller and Schmitt2012). Integrating insights and methods from psycholinguistics and sociolinguistics will lead to better understanding of the factors that shape the route and ultimate attainment of different grammatical properties during language acquisition under different sociopolitical circumstances.