Book announced at https://linguistlist.org/issues/34.460

AUTHOR: Silvina Montrul
TITLE: Native Speakers, Interrupted
SUBTITLE: Differential Object Marking and Language Change in Heritage Languages
PUBLISHER: Cambridge University Press
YEAR: 2022

REVIEWER: Chloe Castle

SUMMARY
‘Native Speakers, Interrupted’ by Silvina Montrul is a monograph which explores language change at the individual and community levels through the lens of heritage language speakers and their use of Differential Object Marking. It draws parallels with McWhorter’s (2007) concept of ‘interruption’, applying it to the heritage speaker experience with incomplete acquisition, exposure, and input. The book brings together literature on psycholinguistics, language acquisition, sociolinguistics, and historical linguistics in its thorough consideration of the processes behind language change. It is unique in its positioning of the heritage speaker as a driver of language change, and its suggestion that bidirectional influence occurs between the heritage speaker and the L1 attriter. Its comprehensive approach makes it a must-read for those with a background in multilingualism, heritage languages, language acquisition, and language change.
In the introduction section, Montrul takes care to define key terms important to understanding the content and theoretical frameworks drawn upon in the book. In alignment with current works on heritage languages, she states her aim to move away from a loss-based perspective and towards an investigation of the structural properties of heritage languages in different domains as dynamic language change. Heritage speakers are defined here as second-generation migrants: immigrant children or children of immigrants in the US. Montrul introduces the idea of heritage speakers as the potential agents of diachronic change in the heritage variety of their languages here. A brief description of each book chapter is then provided.
In Chapter 1, ‘On Heritage Speakers as Native Speakers’, Montrul provides a brief overview of academic and personal definitions and categorisations of native speakers, considering factors such as fluency, pronunciation, vocabulary, and place of birth. She then creates a much-needed typology of native speakers in terms of monolingual and bilingual contexts, and follows up with a discussion on acquisition, ultimate attainment and linguistic stability, non-native speakers, and measuring native-likeness. She argues that there are at least two types of native speaker: monolingually raised speakers and bi/multilingual speakers. L2 learners, who have traditionally been posited as agents of language change in the literature, are of the former type. Montrul suggests that the terms ‘native speaker’ and ‘monolingual speaker’ thus cannot be used interchangeably: bilinguals are regular native speakers and should be the central focus of linguistic research. Indeed, she points out that there is also much variability in monolingually raised native speakers, dependent on level of education, socioeconomic status (SES), ethnicity, gender, style and identity and location. She focuses on education level and knowledge of another language in this chapter, as these are most pertinent to the studies presented in this book. This is followed by a short section on L1 attrition and a longer section on bilingual and multilingual native speakers. She shows that, for both L1 attriters and heritage speakers, there exists a considerable amount of variation (visualised in a bilingual continuum). The interaction of both languages activated in the bilingual brain allow for cross-linguistic influence in bi/multilingual speakers, which results in differing outcomes for a diverse range of bilingual types e.g., simultaneous child bilinguals, sequential child bilinguals, adult L2 learners, child L2 learners, heritage speakers, and L1 attriters. Montrul links this and other factors (i.e., incomplete acquisition, processing difficulty) to language change in the diaspora varieties of the language.
In Chapter 2, ‘Structural Changes in Heritage Language Grammars’, Montrul presents common structural patterns characteristic of heritage language grammars from the literature, including a helpful table summarising the domains of change with examples and sources. She places a particular focus on intergenerational studies, as these are most pertinent in highlighting the relationship between first-generation immigrants and heritage speakers in terms of attrition. She first outlines the groups that heritage speakers have historically been compared to, recognising the debate regarding comparison with a baseline of monolingually raised native speakers in the homeland. She ultimately concludes that the choice of baseline group should depend on the research questions of the study. Common linguistic processes in heritage language development include simplification or reduction of forms, overregularization, reanalysis, and a preference for transparent mappings. Montrul then considers sources of variability in heritage language speech, including quantity of input, quality of input, the effect of linguistic modularity (i.e., not all properties develop at the same time or to the same level) in interaction with sociolinguistic variables (e.g., school starting age), age, the socio-affective dimension, dominant language transfer, interface vulnerability, and language processing limitations.
Chapter 3, ‘Differential Object Marking’ (DOM) gives an overview of the phenomenon. DOM is determined by animacy, referentiality, topicality, and affectedness of the object, as helpfully indicated in her Figure 3.1. Montrul discusses the phenomenon in light of the three languages of focus of the book (Spanish, Hindi, and Romanian), noting that they all have two-dimensional systems, wherein DOM is triggered by animacy and referentiality (definiteness and specificity), though the weighting of these factors differs. She gives a comprehensive outline of DOM in each language, exploring variability and changes occurring in different language varieties (particularly in Spanish), and discussing syntactic analyses from within generative grammar. The fact that the Differential Subject Markers (DSM) for dative experiencer subjects have the same surface form as DOM in Spanish and Hindi (and Romanian, in some cases) is also noted. This becomes relevant later in the book, where dative subjects with psych predicates are used to demonstrate the parallels between heritage language development and natural diachronic processes. However, she also notes that developments in the heritage language do not always go in the same direction as monolingual varieties. Montrul finishes the chapter with a summary of the distribution of DOM in the three languages of interest (Table 3.7).
In Chapter 4, ‘Language Change and the Acquisition of Differential Object Marking’, Montrul provides a model of language change which incorporates individuals, societies, and nations, and which links different subfields of linguistics with different time periods. She asserts the importance of subfields such as historical linguistics, psycholinguistics, language acquisition, and sociolinguistics collaborating, as every change begins with individuals and then spreads to groups and societies. Understanding the sociolinguistic factors (environmental, situational) surrounding the change is critical for the interpretation of its extent and spread. Montrul brings us back to basics with a summary of generative and variationist approaches to language acquisition and generational change. She follows up by critically engaging with the contact and acquisition literature regarding DOM and bringing to light key questions yet to be answered or agreed upon in these fields. She indicates that it may not be the monolingual child that is the agent of language change (as in theories of diachronic change), but rather L2 learners and bilinguals.
Chapter 5, ‘The Vulnerability of Differential Object Marking in Three Heritage Languages’, explains the research questions and hypotheses relating to each of the possibly relevant factors in heritage language speaker performance regarding DOM (situational, linguistic, patterns of language change in the homeland varieties, direct generational language transmission). Montrul explains her predictions based on the different hypotheses. She continues with an explanation of the methodology and procedure. Participants engaged in a language background questionnaire, written proficiency tests, production measures (oral narrative task, elicited production task, elicited written production task), an auditory/written comprehension task, and a bimodal acceptability judgment task (AJT).
In Chapter 6, ‘Differential Object Marking in Spanish as a Heritage Language’, Montrul presents the results for the Spanish language experiments. Starting with an overview of the Spanish-speaking population in the US, Montrul notes that there are more than 60 million Spanish speakers in the country, making Spanish the second most spoken language (Krogstad and Neo-Bustamente, 2020). She discusses the role of Spanish as an important minority language in the US, where there are great tensions between its expansion from continuous immigration and positive value as the most studied foreign language in the country on the one hand, and its contraction from intergenerational language loss and negative value associated with high numbers of low SES undocumented immigrants on the other. Information on participant biographical variables, self-rated proficiencies, preferred languages, reasons for wanting to improve Spanish, and patterns of language use in different time periods in childhood are given here. The Spanish heritage speakers are quite fluent in Spanish, using it within the family and in their professional lives. Input and use of Spanish diminish with age and across generations. Montrul then presents the results of the linguistic tasks, finding that DOM omission with animate, specific direct objects occurs more in heritage speakers in comparison with homeland speakers in Mexico. This erosion of DOM was also attested in first-generation immigrants, particularly for those who had lived in the US for a long time, were older, and who acquired English later in life. Age of onset of bilingualism and amount and quality of language use did not play a role in the results of the heritage speakers.
Chapter 7, ‘Differential Object Marking in Hindi as a Heritage Language’, describes the Hindi/Urdu speaking population in the US. This group is different from the others in several ways, including a higher SES, knowledge of English, and level of education. Hindi does not enjoy the same level of vitality as Spanish in the US, with a smaller population of speakers, though one with strong social networks. Information on the same background variables and linguistic task results are given in this chapter. DOM omission with animate, specific direct objects was found in the heritage speaker group. However, unlike for Spanish, the first-generation group showed no attrition of DOM in comparison with the Hindi speakers in India. The simultaneous bilinguals showed the greatest DOM omission of all the groups, as predicted due to lesser relative input in early childhood. Montrul suggests that the DOM omission observed in these heritage speakers cannot be attributed to the parental generation.
In Chapter 8, ‘Differential Object Marking in Romanian as a Heritage Language’, Montrul outlines the results for the Romanian language experiments. This population of speakers moved to the US in several waves in the 20th and 21st centuries, resulting in a community including a range of SES and educational backgrounds. Participants in the study had an overall good command of Romanian. Information regarding the same background variables and linguistic task results are also given in this chapter. As with the Hindi group, there were no differences in DOM between the first-generation immigrant group and the homeland groups, but the heritage speakers did show variability. Age of onset and bilingualism effects are present for this group. Montrul states that for the Romanian speakers the variability in the heritage speaker group cannot come from first-generation immigrants. There was also less DOM omission in the Romanian group than the others, possibly due to the more salient accusative clitic doubling co-occurring with DOM in this language.
In Chapter 9, ‘Comparing the Three Heritage Languages’, findings from the three studies are brought together and compared. Montrul considers a range of factors in theorising and explaining the differences within and between the groups in terms of DOM use, including: biographical variables; linguistic factors (grammatical complexity, majority language transfer, language processing and lexical access, acoustic salience of DOM markers, other language-particular properties, e.g., definiteness and specificity); and situational factors (size and vitality of the language, education, and SES). To detect whether the erosion of DOM present in first-generation Spanish speaking immigrants is unique to the variety of Spanish spoken by the participants (heritage Mexican Spanish in the Chicago area), Montrul carried out a follow-up study on speakers of other varieties of Spanish in the US and in the respective homelands (Colombia, Ecuador, El Salvador, Guatemala, Nicaragua, Peru, Puerto Rico, and Venezuela). She found that the same pattern holds for these speakers, indicating the development of a widespread ongoing language change in US Spanish. She ultimately concludes that a combination of sociolinguistic and structural factors explains why the erosion and variability of DOM is different in Spanish compared to Hindi and Romanian in the US. She indicates that qualitatively different input from the first-generation immigrants can be related to variable use of DOM only for the US Spanish speakers.
In Chapter 10, ‘Intergenerational Transmission’, Montrul uses the findings from the three studies to posit a new possibility for language change: the bidirectionality of influence between first-generation immigrants and heritage speakers of Spanish in the US. She suggests that the variability observed in both populations is arrived at independently, constrained by individual psycholinguistic processes of structural activation and competition. The first-generation immigrants take some time in the US for their use of DOM to ‘attrite’, at which point this attrition may be reinforced by variable input from the heritage speaker generation. Montrul adapts Yang’s (2000: 232) figure of the dynamics of language acquisition and language change for DOM in Spanish in the US (Figure 10.3), linking children’s E-language (externalised language as the input) back to the parents’ I-language (internalised abstract grammar). She concludes that the situation for US Spanish indicates an emerging dialectal feature: optionally non-DOM marked animate specific direct objects. For Hindi and Romanian, however, she suggests that there is broken transmission of DOM between heritage speakers and first-generation immigrants.
In the final chapter, ‘Implications’, Montrul summarises the findings of the book. She discusses the importance of moving from a deficit view of bilingualism to one where the normality of linguistic diversity is recognised. With this recognition, educational policies can be shaped toward supporting bilingual speakers’ heritage languages. Montrul demonstrates the contributions of this book to the linguistic literature, including: her assessment of the connection between heritage language acquisition and diachronic language change with a focus on individuals in real time; the proposal that heritage speakers are active agents in the diachronic evolution of Spanish as a distinct language variety of the US; and the demonstration of how the main source of input to the heritage speakers may have contributed to the patterns of variability present in their speech.

EVALUATION
This book represents the first treatment of the relationship between heritage language development and diachronic change. It provides an extensive comparative analysis of three heritage languages in the US, addressing links between language development, variation, and change from both a psycholinguistic and sociolinguistic perspective. Montrul takes care to explain her proposal that the heritage speaker is a driver of language change, skilfully weaving her convincing argument throughout the book and culminating with her interpretations of the data in conjunction with relevant linguistic theory.
A minor point of criticism is the lack of comparison with first-generation immigrants without children. This might have helped to further confirm her claim regarding the bidirectionality of language change. However, as Montrul mentions in her book, this is a possible future direction for her research. I am looking forward to seeing the results of this continuation.
Overall, this incredibly comprehensive book is a valuable resource for anyone interested in language contact, language acquisition, heritage languages, psycholinguistics, and sociolinguistics. It is especially nice to see work linking traditionally separated fields of linguistics in striving to answer important questions.

REFERENCES
Krogstad, Jens Manuel & Luis Noe-Bustamante. 2020. Key facts about US Latinos for national Hispanic heritage month. Pew Research Center.
McWhorter, John. 2007. Language interrupted: Signs of non-native acquisition in standard language grammars. Oxford: Oxford University Press.

ABOUT THE REVIEWER

Chloe Castle is a postdoctoral researcher in the AcqVA Aurora Lab at UiT The Arctic University of Norway. Her research interests include multilingualism, language acquisition, language attrition, heritage and community languages, and language change and variation. She is interested in taking a combined psycholinguistic and sociolinguistic approach to examining these research areas. Her previous research has focussed on grammatical borrowing and replication between Czech and English in two parallel communities, namely Czech South Australians and L1 English Ln Czech speakers in Prague. Her current research as part of the ADIM project focuses on morphosyntactic crosslinguistic influence in L3 and Ln acquisition.