What can journals do to increase the publication of research on the acquisition of understudied languages? A commentary on Kidd and Garcia (2022)

This commentary focuses on what editors and reviewers could do to increase the publication of research on understudied languages. Specifically, I discuss three areas in which editors and reviewers could shift their perspectives and, in so doing, support the goal of increasing diversity in our field: (1) Rethinking the criteria for novelty of contribution, (2) Contextualizing sample sizes and methods and (3) Embracing multilingualism as typical development. Finally, I discuss issues around achieving equity in an English-dominant publishing world.

The article by Kidd and Garcia (2022) expertly brings into sharp relief the diversity problem in the field of child language acquisition. In this commentary, I will focus on what I believe journal editors could do to increase the publication of research on understudied languages. Specifically, I discuss three areas in which editors and reviewers could shift their perspectives and, in so doing, support the goal of increasing diversity in our field: (1) Rethinking the criteria for novelty of contribution, (2) Contextualizing sample sizes and methods and (3) Embracing multilingualism as typical development. I also discuss issues around achieving equity in an English-dominant publishing world. My perspectives have been primarily shaped by my 6 years of experience as editor-inchief of the Journal of Child Language.

Rethinking the criteria for novelty of contribution
All too often, reviewers and editors look at manuscripts based on understudied languages from the perspective of what they add that is 'new' to our knowledge of acquisition, which is based mostly on English. I suggest that, as a research community, we rethink our reactions to a manuscript that includes a duplication of methods and/or findings from prior research on English or other European languages if that manuscript is based on an understudied or endangered language. Kidd and Garcia quite rightly point out that research on understudied languages has led to overturning some claims about universality in acquisition, for example, that passive voice in morphosyntax is always acquired late or that a motherese register is ubiquitous across societies. But, we must also consider that replication of key findings on European languages with non-European languages, especially those from non-WEIRD (i.e. Western, Educated, Industrialized, Rich and Democratic; Henrich et al., 2010) societies, is equally important. In short, both similar and dissimilar findings across languages and cultural contexts will yield a stronger empirical base for understanding dominant trends and potential universals. In fact, it will likely bring us away from a dichotomous universal versus language-specific mind-set about acquisition and closer to one where most linguistic phenomena are on a continuum between these two end points. In short, replication in science is necessary and important for theory building. Kidd and Garcia's quote from Pye (2021) is related to this point: 'The data needed to construct a representative sample of languages to test acquisition theories in a meaningful way do not exist' (p. 454).

Contextualizing sample sizes and methods
We need robust, in-depth descriptive work on endangered languages urgently -before the last generation acquiring a certain language grows up. During my tenure as editorin-chief of the Journal of Child Language, I have allowed case study or small sample size studies go forward for review when they are based on endangered languages, while I might do a desk-reject if this same study design were used for English or another major European language which has been well documented. Needless to say, such studies are often based on naturalistic language sampling rather than experimental approaches. Language sampling approaches are needed not only for the descriptive work that is lacking on understudied and endangered languages but also because experimental work requires larger samples and conditions for testing that might not be feasible outside of the mainstream urban communities where most of the child language research is conducted. Furthermore, relaxation of submission length rules, encouragement of links to Open Science sites for supplementary materials and so on can be used to augment the space allowed for articles requiring a lot of description of the language itself before getting on to the actual acquisition study. In sum, editors and reviewers need to consider quality and appropriateness of methods and study design within the context of the language being studied.
A related point here is that endangered and understudied do not mean the same thing. A highly descriptive case study on an understudied language like Korean might quite justifiably be judged as not meeting criteria of novelty and standards of methodological rigour because there are millions of children acquiring this language and will be for generations to come. By contrast, in Canada and other colonized countries, for some indigenous languages, the current generation of children will be the last to acquire the language as their first language. In the latter case, a highly descriptive case study should be viewed differently by editors in terms of appropriateness for publication.

Embracing multilingualism as typical development
Bi-and multilingualism among children are widespread across the globe and are likely to be more typical than monolingual development (Byers-Heinlein & Lew-Williams, 2013;Wei, 2007). The assumption of monolingualism as the norm or 'gold standard' likely stems from our overreliance on acquisition data from mainstream middle-class children in English-speaking countries. If children who are acquiring an understudied language are growing up in a bi-or multilingual setting, this should not be seen to undermine the quality of the acquisition data on the part of editors and reviewers. If studies include samples of children with varying degrees of bilingualism, this lack of 'homogeneity' should not be viewed as a methodological flaw in the review process because, in some social-cultural contexts, multilingualism is part of all children's language experiences, and thus is integral to their typical language acquisition patterns. In multilingual regions across the globe, it could be a challenge to recruit monolingual children for research and so it should not be an expectation or goal. For example, the current generation of Welsh and Catalan speakers is growing up bilingual, and monolingual child speakers of these languages are rare past early ages. For studies of children learning an understudied language in a multilingual context, editors need to be wary of requests from reviewers for a 'monolingual control group', or insistence that data be gathered on both/ all languages a child is exposed to when the study is focused on just one of the languages. However, where appropriate, researchers who are working with understudied or endangered languages in a multilingual context could be encouraged to consult the literature on child bilingual development to inform their data interpretation. Editors and reviewers can play a role in pointing authors in the right direction on this issue in a constructive and supportive manner.

Quality of writing and research: how to achieve equity in an English-dominant publishing world?
Kidd and Garcia bring forward the problem of 'the means of research production' being concentrated in a handful of wealthy, mostly English-speaking countries. This problem is not only reflected in a small sample of languages being studied, but it is also reflected in barriers to scholarly and English-language training in other societies that, in turn, limit opportunities for research dissemination on understudied languages. Level of English proficiency and level of scholarship are thorny issues in the review of manuscripts and are often at odds with the goal of increasing the diversity of languages studied. (By 'level of English proficiency' I am not referring to some infelicitous word choices, misused prepositions or misplaced adverbs; instead, I am referring to manuscripts where the level of English proficiency is too low to make them easily readable.) All too many times, I have reluctantly desk-rejected a manuscript based on an understudied language for these reasons. Discussing how to redress these inequities is worthy of a full-length article or more on this topic. However, I believe it is unlikely that equity actions addressing the issues of scholarship and English proficiency can begin at the editor's desk; they should be targeted at all the stages before manuscripts are submitted for review. The field needs active and purposeful discussions about how to make this happen. For example, discussions could include the following: building more partnerships between researchers from WEIRD and non-WEIRD societies, creating more opportunities for international graduate students at universities in WEIRD countries and fostering more collaboration between linguists specializing in language documentation and researchers of child language acquisition, along with more interdisciplinary opportunities for graduate students to receive training in cross-cultural, cross-linguistic research on understudied languages. Initiatives like LangVIEW (https://alecristia.github.io/LangVIEW/) are clearly a step in the right direction. Finally, journals could be encouraged to provide free/affordable English-language editing services, pre-submission, for authors whose manuscripts are based on understudied languages.
In conclusion, there are several concrete actions that could be taken in the review and editorial process to increase the diversity of languages in publications on child language acquisition. I heartily support this endeavour for all the reasons eloquently put forward in Kidd and Garcia's article.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.