The Acquisition of Quotatives and Quotative Be Like among Chinese L2 Speakers of English in Australia
Round 1
Reviewer 1 Report
PLEASE SEE ATTACHED REPORT
GENERAL
This is a potentially interesting article that looks at the acquisition of the quotative system and the innovative BE LIKE (hereafter, BL) quotative by 14 Chinese L2 speakers (stratified by proficiency and length of residence) residing in Melbourne, Australia. Systematic comparisons are effected with an L1 baseline, drawing on Rodriguez Louro’s (2013) study of quotative variation and change in Perth, Australia. The analytical framework is embedded in variationist sociolinguistics.
The findings of the investigation are pretty straightforward: BL is used considerably less frequently by Chinese L2 speakers contrasted with its robust presence in the L1 baseline, and the only L2 speaker who avails himself of this vernacular option to any appreciable degree is 032M, a speaker with a high proficiency [IELTS] score, and one (p.16/532-538) who has attended high school in Australia and had over 6 years of education at Australian educational institutions. An important take-home message of the research is that regular and sustained exposure to the vernacular norms of AusE is a prerequisite for the acquisition of BL. But a key question that remains beyond the confines of the article, at least with respect to BL, is what exactly has been acquired? I fully concur with the authors of the paper (p.3/83-84) that acquisition cannot be adequately gauged from “usage rates alone” and that a more refined diagnostic of acquisition resides in the fine-grained constraints that condition variant choice. Unfortunately, with only 16 tokens of BE LIKE in the L2 data out of a modest total of some 217 quotatives, the general paucity of BL means that the authors have very little quantitative latitude to explore this question. This difficulty is further compounded by the revelation (p.14/Table 6) that only two of the four L2 speakers who used BL produced more than one token of that variant. In sum, BL is hardly productive in the targeted L2 population. The paucity and limited dispersion of BL tokens in the L2 dataset should oblige the authors to dial back their rather speculative claims (p.15/504-515) about the constraints that condition its use in L2 speech, and should likewise prompt them to re-consider how their findings speak to the putative advancement of BL along the pathway of grammaticalization. There just aren’t sufficient tokens of BL in their L2 data to make any compelling claims. There is no “quick fix” that I can offer to rectify these data limitations, although it may be worth investing a little more effort in investigating whether the L2 speakers use SAY and ZERO, the quantitatively preponderant options in their speech, in ways that are commensurate with variant patterning in the L1 benchmark.
OTHER ISSUES
P.1-2 The authors dedicate some space to providing a relevant synopsis of quotative variation and change in L1 varieties, but the rationale for the focus on quotative variation and change in L2 populations (p2./51-67) feels truncated and under-developed to me. Why should this be intrinsically interesting? What kind of learnability issues does discourse-pragmatic variation, and quotatives in particular, pose for L2 learners? At least one of the major issues, adumbrated in Meyerhoff & Schleef (2013)—not cited in the article under review—is that second-language learners are confronted with acquiring what is essentially “a moving target.” I also note that p.2/51-67 makes no reference to other recent and important publications that deal with the acquisition of quotatives by L2 speakers. No mention, for example, is made of Corrigan (2020) or Davydova (2020). Their results, especially those reported in Davydova (2020), do not always harmonize with some of the findings reported in the article under review.
P2. The use of Rodriguez Louro (2013) as an L1 baseline is not without its problems. I think that the authors are probably cognizant of what these problems are, but they should be more upfront about them. The fact that Rodriguez Louro’s (2013) data were collected some years earlier than the authors’ data is, in my view, a relatively minor issue. A more serious issue is that her data, collected as they were in Perth, represent a speech community that is at some geographical remove from Melbourne, where the L2 speakers were recorded. I should also point out that according to information on p9./333-334, the L1 baseline appears to be based on speakers aged between 11-63. Yet none of the Chinese L2 speakers, according to p.6/Table 3, is older than 35. Given that quotative variation and change is highly sensitive to aged-based differences, how accurate is the comparison of the L2 data with the L1 AusE baseline? Furthermore, can it be safely assumed that the local L1 quotative system in Melbourne AusE is necessarily isomorphic with its counterpart system in Perth? This brings me to a more general issue, namely, the nature of L2 speakers’ exposure to the L1 target variety. Where had these L2 speakers acquired their knowledge of English? Via the formal education system in their home country, at least initially? Or partially, or entirely, in Australia? If so, where? In what contexts? Knowing a little more about the speakers’ L2 acquisitional histories might shine a light on their variable usage patterns.
P4./93-99 There’s some (unintentional?) ambiguity here. Mimetic discourse extends beyond the use of non-lexicalised sounds—they are not one and the same thing. I also note (Table 7/p.15) that this constraint is not explored in the present article.
P6./215-218 “…as the participants of the study will have been exposed to native AusE in their daily lives, the constraints governing BE LIKE for Chinese L2 learners are expected to be similar to the findings from Rodriguez Louro (2013)” This statement is at variance with what we know about L2 acquisition. If (adult) L2 speakers acquire innovative forms such as BE LIKE via diffusion rather than transmission, we would hypothesize, following Labov (2007:349), that there would be a weakening or re-ordering of constraints underlying variant use rather than faithful replication of those constraints. This brings me back to a primordial question that I raised earlier: what does it mean to have acquired BL, or any other L1 quotative variant, for that matter? How is this assessed? A clearer statement on this issue would enhance the analytical approach of the study.
P.6/P.7 The L2 participants are said (p.6) to range in proficiency from IELTS band scores of 6.0 to 9.0. On page 7, however, it says that participants had to have English proficiency to at least IELTS band 4. Why mention this if the IELTS band scores are clearly stated on the previous page?
P.9/319: “A total of 216 quotative tokens were identified among the L2 speakers…” but 217 are reported in Table 5. Figure 1 also replicates the same information shown in Table 5. This replication feels redundant (to me). Why not just keep Table 5?
P.10. In their discussion of DM like as a quotative, the authors draw comparisons with an earlier study of Irish English. This confused me. Why are no direct comparisons drawn with Rodriguez Louro (2013)? Is this because there were no instances of DM like used as a quotative in the AusE L1 baseline (see Table 5, p.9). Alternatively, might Rodriguez Louro (2013) have collapsed tokens of DM like with ‘Other’?
P15/510-516 Travis & Kim (2021) reportedly characterize the quotative behaviour of Chinese Australians as “conservative.” The authors of the article under review state that their “L2 learners do not appear to be mirroring this conservative behaviour.” I don’t understand this. I see no compelling evidence in Table 5/p.9 indicating that the L2 speakers are behaving innovatively. About 70% of their variable context is comprised of just two variants: SAY and ZERO, neither of which, to the best of my knowledge, has been implicated in innovation, and both of which are long-attested variants in the quotative system. SAY is likely the prototypical quotative transmitted to learners in classroom environments. The ZERO variant, apparently a universal option in the quotative system (see Güldemann 2008), may allow (lower-proficiency) L2 learners to circumvent some of the morpho-syntactic challenges of using an overt verb of quotation, as the authors suggest.
P.18. What proportion of OTHER was made up of graphic dialogue introducers or semantically richer quotative expressions like “believe” or “wonder”. How might the use of such options reflect the (possible formal language learning) contexts in which L2 speakers had acquired their (early) knowledge of English?
P.20 How might the L2 learners’ L1 influence their choice of quotative variants in their second language?
REFERENCES
Corrigan, Karen. (2020). Linguistic Communities and Migratory Processes. Berlin: Mouton de Gruyter.
Davydova, Julia. (2020). The role of sociocognitive salience in the acquisition of structured variation and linguistic diffusion: Evidence from quotative be like. Language in Society 50: 171 – 196
Güldemann,Manfred. (2008) Quotative Indexes in African Languages. Berlin: Mouton de Gruyter.
Labov, William. (2007). Transmission and diffusion. Language 83: 344-387.
Meyerhoff, Miriam and Erik Schleef. (2013). Hitting an Edinburgh target: Immigrant adolescents’ acquisition of variation in Edinburgh English. In Robert Lawson (ed.) Sociolinguistic Perspectives in Scotland. Basingstoke: Palgrave Macmillan, pp. 103-1
Comments for author File: 
Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
Adopting variationist paradigm, this study examined the acquisition of quotative system and the use of variant "be like" by Chinese L2 speakers of English in Australia. The results were compared with native speaker patterns reported in Rodriguez Louro's 2013 and Author 2's 2019 studies. The paper is well-written and the study is well organized and the design is solid. The results really showed that the L2 speakers' usage pattern of quotative variants, especially "be like", is very different from native speakers'. My two comments/suggestions are the following:
The first one regards the unbalanced data due to outliers. The study mentioned that the speaker 032M in the upper proficiency group contributed 10 tokens of "be like" out of a total of 16. He seems to be an outlier in the data. I am curious about how many other quotative variants he contributed. If a table with each speaker and his/her number of each variant produced can be provided, it would be helpful to interpret the results.
Another one is that I can see L1 influence on the Chinese L2 speakers' use of variants such as "feel" and "think", the equivalents of which can be found in Chinese -- "ganjue" ”juede". However, this needs to be confirmed by further studies and the authors did mention this as a future direction. There is literature that addresses the use of them as discourse or pragmatic markers and I think the authors can cite a couple of them and then frame L1 influence as a future research direction. The following study is one example. https://francis-press.com/uploads/papers/p7hB9z7qDdHwMerHJ40Zy12bcDqppA7d9GUQQ1SO.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Please see attached file.
Comments for author File: Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
