Causal Relations and Cohesive Strategies in the Narratives of Heritage Speakers of Russian in Their Two Languages

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Comments for author File: Comments.pdf

Author Response

Dear reviewer,

I'm grateful for your meticulous revision of the document and for the important comments you have made. 

In the letter attached we have provided detailed answers. 

Sincerely,

Judy

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

I applaud the author(s) for a thourough and inetresting study and encourage to take a closer look at the Intro. A significant restructuring of the opening arguments will make this a better paper the study definitely deserves. 

Comments for author File: Comments.pdf

Author Response

Dear reviewer,

Thank you for your extraordinary comments. It has been a learning experience to follow them. I hope the quality of the manuscript has improved after the major changes we have made on it. 

In the attached letter we provided more detailed responses to each of your suggestions. 

Sincerely,

Judy

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This study is a welcome addition to the line of inquiry into development of narrative abilities in bilingual children. It sets to examine the way causal relations are expressed in elicited narratives of Russian-Hebrew bilingual children aged 5-7 in both their heritage (HL) and societal (SL) languages. More specifically, the authors examined frequency and type of episodic components, the number of enabling and motivational relations expressed as well as linguistic expressions of causal relations in the narratives, which are retellings of the picture book Frog Where Are You? In addition, data on narrative complexity (which the authors call “narrative productivity” are presented for both languages, i.e. total number of words (TNW), Number of Different Words (NDW), Number of C-units and mean length if C-unit. The authors report that some features of the narrative were similar in both languages, while others differed.  For example, participants used both syntactic and referential means of cohesion in their narratives in both languages, had similar “narrative productivity” scores, and lexical chains were the most common way of inferring relations. It is not surprising that young children produced more episodic components in both languages at the start of the narratives, considering the length of the picture book (29 pages). The authors also report that the age of onset of bilingualism did not affect the number of episodic components and causal relations while proficiency did. As far as differences were concerned, more episodic components were found in SL than in HL. It was interesting to find that there was a different distribution of framing of the causal elements, with more enabling relations in Russian (HL) and more motivational relations in Hebrew (SL). They report more correlations between “narrative productivity”, episodic components, and causal relations for SL than HL. 

The paper is clearly written and overall presents a compelling argument. However, I would recommend addressing the following concerns before it is published:

The literature review provides a good overview of the relevant constructs and research findings. However, while earlier work using LITMUS-MAIN instruments is discussed and said to have influenced the current study, there are more recent publications by Gagarina and Rodina with colleagues, which may help make the discussion in this paper more nuanced, including discussions of macro and micro-structure and the measures of narrative complexity/”narrative productivity.”  Also, in this study, is TNW calculated with or without hesitations, repetitions, false starts, code-switches to the other language?

There needs to be more information provided about the construct of proficiency as used in the paper and in the distinct instruments that were used to test it in SL and HL. I was not sure what the proficiency scores mean, considering that the tests for Russian and Hebrew were developed separately and have different components. The way proficiency is presented in the methods section and the analysis leads us to think that the authors assume that the total scores in two very different batteries of tests, which were designed separately for Russian and Hebrew, are comparable in overall assessment of proficiency, but are they? Since little information about proficiency measures is provided, it makes me wonder what the findings about the effect of proficiency are really telling us: is it really showing us differential effect of proficiency in HL and SL on narrative ability or is it showing differential suitability/sensitivity of the two different proficiency measurement instruments for finding correlations with narrative features/skills in bilingual children.

Also, it seems to be assumed that all the children were dominant in the societal language, but we do not know if that was measured in any way by the proficiency measures or background questionnaire. I wonder if dominance rather than AOB has been considered in the analysis, especially since cross-linguistic comparisons are made? Do all Russian narratives have fewer episodic components than their Hebrew counterparts, regardless of SL vs HL dominance of the participants?  

Related to this, I would appreciate a bit more explicit synthesis of the reported findings from the point of view what might these findings tell us about cross-linguistic differences vs effects of proficiency/language dominance and other individual factors. When certain features are more present in Russian vs Hebrew narrative, is this evidence that one language is stronger or that the participants are sensitive to what is the dominant pattern of the language? 

Discussion of the qualitative data seems asymmetrically focused on Russian data with some examples only given from Hebrew. Is there are reason for that?

Some specific comments to the manuscript:

P18 – there is an implicit argument that Frog Story is better than MAIN, as it resulted in more episodic components than in Gagarina and colleagues’ (2016) study. The difference could be for a host of reasons, including per authors own findings, based on different proficiency of participants Gagarina’s study was able to recruit. Also, although the care is taken to counterbalance the learning effect, the fact that the children were retelling the same story in too languages is a disadvantage, while the instruments in MAIN-LITMUS were designed to trace parallel elements in narratives in two languages. So, I would not make very strong claims about superiority of this picture book. Also, as the authors themselves report, most of these young kids did not have the attention span or other capacity to report all the episodic components of such a long story. So, the length is hardly an advantage. 

P19 – The answer to the second research question needs to be elaborated a bit more in the discussion! What do these findings tell us?

Also, “Hebrew advantage” in the number of episodic components sounds too strong to me.

 

Author Response

Dear reviewer,

We're grateful for your insightful comments. These have been addressed to in the document attached here. 

Sincerely,

Judy

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for the improved manuscript.

Author Response

Dear Reviewer,

We're grateful to you for having revised the second - improved - version of our manuscript. We're addressing here the minor corrections you suggested to perform to the manuscript. 

Our answers are marked in bold, following the reviewer's comments:

1) Grammatical inaccuracies: 

-We have thoroughly revised the whole document again and corrected all the grammatical inaccuracies. In addition, we have reformulated some of the sentences and paragraphs for the purpose of clarity. These are marked in green. 

LINE 245

"Thus, unlike more strictly linguistic measures of microstructure, the present study adopts a more exhaustive quantitative and qualitative methodology that considers narratives as a discourse platform that integrate (should be 'integrates') forms and functions as two interacting levels of knowledge."

-Corrected

LINE 416:

"In verbs, Hebrew – like Russian – mark (should be 'marks') person, number, gender, and tense – past, present, and future (Berman, 1978; 2014) but only Russian marks perfective versus imperfective aspect (Gagarina, 2011).

-Corrected

Please, perform a thorough language check to identify and correct these and other similar issues.

-Thank you for calling our attention to these kind of errors. We truly hope that we haven't missed any of them. 

2) References missing from the Bibliography:

Gagarina et al. 2012, 2019

Bohnacker et al. 2022

Lindgren et al. 2023

Tribushinina et al. 2022, etc.

These are just some of the missing titles, so please, make sure that all references in the main text are also in the bibliography.

-We have checked meticulously the references in the main text and in the list to make sure that all of them appear in both sites. We have also checked that all the references are relevant to the contents of the manuscript.

Sincerely,

Judy