12 Replies to “Connecting Some Dots in Heritage Language Acquisition of Mood Selection (Melisa Dracos, Pablo E. Requena, Raquel Nuñez)”

  1. Dear all, I would like to congratulate you on a wonderful presentation and a great study. I just wanted to comment how fascinating it was to me that your retrospective measure was so robust it soaked up all the variability in your model leaving all other predictors non-significant; this is truly strong evidence in support of the predictive power of your measure, congratulations!

    1. Hi Priscila,
      Thank you so much for your comment. We are always trying to find ways to bridge the gap in data that throughout development that is very hard to get, so we tried this way of asking adult HSs to prvide data on their exposure/use ass they parents would have reported it when they were kids, and we used the exact same measure we used for the actual kids in our study. We were also surprised by this finding.

    2. Thank you, Priscila! Yes, as Pablo mentioned, we were also surprised that this retrospective measure was a stronger predictor than not only current exposure/use but also current Spanish proficiency. It suggests to us that at least for the acquisition of particularly vulnerable, later-acquired structures like subjunctive, that it might be critical to have a measure that sheds as much light as possible into their language environment and exposure/use patterns during the time these structures are acquired. Thanks again!

  2. Excellent presentation, as always!

    How interesting that primary school use/exposure patterns are more predictive of subjunctive use in adult HSs than current use/exposure patterns!

    #1: If the adults were college students, perhaps their current language use/exposure was not predictive because it may be more dependent on their specific setting at the time of completing the survey. (Lots of heritage speakers at my university report using their Spanish much less on campus than at home.) If this is true and respondents were describing their language use/exposure patterns on campus, then maybe this serves to (unintentionally) flatten out a predictor variable that would have had more variability and predictive power elsewhere (e.g., if respondents had been reporting their language use/exposure patterns at home…when they’re not at college).

    #2: Assuming that my speculation in #1 is off-track, can you speculate about other possible reasons why primary school use/exposure would be a greater predictor than current exposure? Do you think heritage speakers with relatively lesser exposure during primary school can become more “categorical” with subjunctive if they have intensive Spanish exposure as adults?

  3. Hello dear presenters.
    I enjoyed so much your presentation. The topic is interesting and covers a gap in the HLLs research. Is it possible that could send me your references? I would like to see the composite questionnaire you mentioned in your presentation. Thank you.

    1. Dear Adrian,
      We designed the composite questionnaire, but if you send me an email (pablo.requena@utsa.edu), we would be happy to share it with you. Just to add something, I don’t think there will be anything special about the questionnaire per se. I think it is more about tracking exposure and use of the HL at the time when these structures are acquired/cemented. But we keep exploring this in more detail.

  4. Great work guys, love it! 🙂

    I would have loved to chat with you in person and have a coffee/beer con los demás subjuntivistas, but this will have to do (for now, at least!). I have a couple of comments:

    1. You present the use of subjunctive with ‘cuando’ as being categorical (which it is in the context that you’re testing), however, it would not be fair to say that it is as categorical as in the volitional or the other adverbial contexts presented, where indicative is not expected or does not carry any changes in interpretation. This does not really matter for the purpose of this study, but I think we ought to be more careful when talking about categorical vs. variable uses of subjunctive, especially given the role of certain triggers (or lack thereof) when integrating morphosyntactic information. In this structure, for example, I bet there’s still some competition between indicative/subjunctive forms (even for speakers whose level of proficiency/experience is high and would in principle facilitate subjunctive use) than in volitional contexts, for example (and not only because of AoA of the structure, but because of the co-existence of both forms in this particular adverbial clause).

    2. You know me and my love for production data. What were the most usual divergences found across groups? Was indicative the only form used, or did you find other alternatives? If so, were they associated (or more likely to appear) in specific groups?

    Thanks as always for your inspiring work!

    1. Hi Silvia,
      Thanks so much for your comment and question! And yes, I’m wishing we could again chat about all this over a drink like we did last year…sure hoping before long the subjuntivistas can gather again!

      Regarding your first comment, we agree that ‘cuando’ should not be classified as ‘categorical’, and in fact I just had to return to the presentation to realize that we did in fact put that on the first slide. We did debate this classification initially because there is some precedent for referring to the specific context ‘cuando + future reference’ as categorical (e.g., as Viner 2016 did). However, we did ultimately decide for example in our recent paper not to go with this classification, precisely for some of the reasons you mentioned. Now, we should also note that consistent with some of your research regarding modality and variability, we found no difference between SUBJ use with cuando vs antes de que (which of course lexically selects subjunctive). Unlike variability with other Subjunctive contexts (e.g., relative clauses, presupposition, non-assertion etc), there is such a strong semantic distinction between cuando + subjunctive and cuando + Indicative that we think this may mitigate effects of variability in this context. Or it could be that variability doesn’t matter as much when modality is controlled, as I know your work has shown for deontic modality, and Lustres and others have shown within epistemic. Would love to chat more about this!

      As for your second question, almost all alternative responses were indicative present, although there were a couple tokens of indicative in imperfect, and a couple instances of an English verb form. These forms were produced almost exclusively by the lower exposure/use group. While proficiency (based on DELE) wasn’t a significant predictor, we did also see a trend to use more of these non-subjunctive alternative responses with decreased levels of proficiency.

      Thanks again for your great comments, and looking forward to talking more!

  5. Thank you all for this informative and interesting presentation! Can you provide us with the reference about the composite questionnaire used?

Comments are closed.