Periodic Reporting for period 1 - MAG (Multilingual acquisition in Germany: the case of Upper and Lower Sorbian)
Période du rapport: 2021-09-01 au 2023-08-31
Many modern theories of second language acquisition approach the issue acquisition through the lens of perceptual acquisition. While this is an important aspect of the acquisition process, there is a missing link between the perception and production of novel second language segments and how learners acquire the representation required to articulate novel segments. As such, in this project, I am examining the second language acquisition of novel segments with respect to the acquisition of speech production. This project makes an important contribution to understanding how second languages are acquired, which is becoming increasingly important in an increasingly globalized and multicultural society. It is also becoming increasingly important for the economic success of individuals to master more than one language. Thus, understanding how people acquire a second language and specifically how individuals learn to speak is becoming important for the well-being and health of society as a whole. To this end, my project has three objectives:
(1) examine how second language learners acquire acoustic and articulatory speech patterns
(2) examine the connection between speech perception and speech production in language acquisition
(3) model patterns of acquisitional patterns to improve models of second language acquisition
To achieve these objectives, I am examining second language acquisition of multiple languages (Lower Sorbian, German, and Polish) using acoustic, ultrasound, and perceptual measurements. I am examining learners at different stages of acquisition (A [beginner], B [intermediate], and C [advanced]) to better understand to what strategies learners use when the are acquiring new speech segments. Acoustic and ultrasound data will be leveraged to understand how learners acquire novel speech patterns and the perceptual data is being leveraged to better understand what the connection between speech perception and speech production are.
The results implicate that language learning mechanisms active in L1 acquistion are also involved in L2 acquisition. However, many factors influence ultimate attainment of L2 acquisition. I additionally found that while typical acquisition in the absence of instruction involves naturalistic mechanisms retuning sensitivity to different phonetic correlates to language specific phonological contrasts, followed by adjusting articulatory configurations until contrasts between novel segments are learned. I also found that in learners can acquire novel segments for which they are not particularily discriminating between the target L2 segment and an acoustic-phonetically similar L1 segment (e.g. German uvular /R/ and Polish /r/). Some metalinguistic knowledge, however, likely contributes to acquisition in this case.
This research stands to benefit the way L2 acquisition is approached by language educators and also documents difficulties faced by communities who speak endangered languages.
(1) examine how second language learners acquire acoustic and articulatory speech patterns
(2) examine the connection between speech perception and speech production in language acquisition
(3) model patterns of acquisitional patterns to improve models of second language acquisition
To achieve these objectives, I am examining second language acquisition of multiple languages (Lower Sorbian, German, and Polish) using acoustic, ultrasound, and perceptual measurements. I am examining learners at different stages of acquisition (A [beginner], B [intermediate], and C [advanced]) to better understand to what strategies learners use when the are acquiring new speech segments. Acoustic and ultrasound data will be leveraged to understand how learners acquire novel speech patterns and the perceptual data is being leveraged to better understand what the connection between speech perception and speech production are.
The results implicate that language learning mechanisms active in L1 acquistion are also involved in L2 acquisition. However, many factors influence ultimate attainment of L2 acquisition. I additionally found that while typical acquisition in the absence of instruction involves naturalistic mechanisms retuning sensitivity to different phonetic correlates to language specific phonological contrasts, followed by adjusting articulatory configurations until contrasts between novel segments are learned. I also found that in learners can acquire novel segments for which they are not particularily discriminating between the target L2 segment and an acoustic-phonetically similar L1 segment (e.g. German uvular /R/ and Polish /r/). Some metalinguistic knowledge, however, likely contributes to acquisition in this case.
This research stands to benefit the way L2 acquisition is approached by language educators and also documents difficulties faced by communities who speak endangered languages.
My project has expanded in scope dramatically, but still maintains the original aims and goals of the proposed project. Data were collected for second language (L2) Lower Sorbian, German, and Polish learners. Additionally, language documentation in the form of acoustics, ultrasound, and magnetic resonance imaging (MRI) data were collected for Upper and Lower Sorbian. Data analysis and publication of L2 of Lower Sorbian acquisition of the three-way sibilant is completed. Currently, a submission to Second Language Learning for L2 acquisition of German front rounded vowels is being revised for resubmission, and two additional publications for the L2 acquisition of the sibilant system in Polish and the uvular rhotic (r-sound) in German are in preparation.
Data is currently being processed for MRI documentation of Upper and Lower Sorbian. I am comparing the vocalic system, the liquids (l- and r-sounds), and sibilant fricatives using 3D images of the vocal tract and acoustics. This part of the project is not only documenting Sorbian, but also advancing our understanding of the connection between vocal tract shapes and acoustics. An additional two to three publications are planned with this data.
Additionally, results have been presented at 6 peer-reviewed international conferences (and 1 to be presented), two of which include published conference proceedings. I am also preparing 4 publications.
Data is currently being processed for MRI documentation of Upper and Lower Sorbian. I am comparing the vocalic system, the liquids (l- and r-sounds), and sibilant fricatives using 3D images of the vocal tract and acoustics. This part of the project is not only documenting Sorbian, but also advancing our understanding of the connection between vocal tract shapes and acoustics. An additional two to three publications are planned with this data.
Additionally, results have been presented at 6 peer-reviewed international conferences (and 1 to be presented), two of which include published conference proceedings. I am also preparing 4 publications.
Within the scope of the objectives of my project, two main findings have become apparent:
(1) language learning mechanisms are still available to learners beyond the so-called "critical period."
(2) L2 articulation of perceptually similar to L1 segments progresses by utilizing L1 segments and modifying them into novel representation.
The first major finding was achieved through examination of aspects of speech not taught in the classes and by examining learners that did not receive specific pronunciation training. This finding impacts aspects of second language acquisition and especially models of second language acquisition that assume learning mechanisms are fundamentally different from L1 learning mechanisms or theories that suggest that L1 mechanisms are no longer available to older learners.
The second major finding of this project was uncovered through examination of the development from A-level to C-level when acquiring novel sound segments. The ultrasound data revealed that in the early stages of acquisition, learners use constriction locations and tongue contours similar to that of similar L1 segments and that over time divergence into distinct tongue shapes occurred. Additionally, in situations where learners had to acquire 2 segments that were similar to 1 L1 segment, the trajectory was that initial divergence into a single category took place and that over time, separation of the two novel segments into two categories takes place. Overall, this implicates that novel category formation is a complex process but that the onset is quick, even though the process is a long-drawn out one.
The implications of these findings have potential impact for both language revitalization and second language learning. In the language revitalization context, if the aim of the community is to maintain a segmental inventory similar to that of older speakers, sufficient exposure to authentic speech is necessary. In situations like in Lower Sorbian where very few L1 speakers remain and exposure to authentic speech is difficult, additional measures would need to be taken. These measures include (but are not limited to) ultrasound or MRI videos that show tongue movements during the articulation of difficult to acquire segments, and perceptual training that highlights phonetic differences between L1 segments and the target L2. In the context of second language acquisition of non-moribund languages where significantly higher access to authentic speech is possible, it seems important that students be given the opportunity to be exposed to that speech as much as possible. The fact that language learning mechanisms appear to be available means that significant exposure to L1 pronunciation can assist learners in acquiring new articulatory representations. Additional exposure to pronunciations guides with ultrasound or MRI would also expediate the process and assist learners with difficulty in acquiring segments.
(1) language learning mechanisms are still available to learners beyond the so-called "critical period."
(2) L2 articulation of perceptually similar to L1 segments progresses by utilizing L1 segments and modifying them into novel representation.
The first major finding was achieved through examination of aspects of speech not taught in the classes and by examining learners that did not receive specific pronunciation training. This finding impacts aspects of second language acquisition and especially models of second language acquisition that assume learning mechanisms are fundamentally different from L1 learning mechanisms or theories that suggest that L1 mechanisms are no longer available to older learners.
The second major finding of this project was uncovered through examination of the development from A-level to C-level when acquiring novel sound segments. The ultrasound data revealed that in the early stages of acquisition, learners use constriction locations and tongue contours similar to that of similar L1 segments and that over time divergence into distinct tongue shapes occurred. Additionally, in situations where learners had to acquire 2 segments that were similar to 1 L1 segment, the trajectory was that initial divergence into a single category took place and that over time, separation of the two novel segments into two categories takes place. Overall, this implicates that novel category formation is a complex process but that the onset is quick, even though the process is a long-drawn out one.
The implications of these findings have potential impact for both language revitalization and second language learning. In the language revitalization context, if the aim of the community is to maintain a segmental inventory similar to that of older speakers, sufficient exposure to authentic speech is necessary. In situations like in Lower Sorbian where very few L1 speakers remain and exposure to authentic speech is difficult, additional measures would need to be taken. These measures include (but are not limited to) ultrasound or MRI videos that show tongue movements during the articulation of difficult to acquire segments, and perceptual training that highlights phonetic differences between L1 segments and the target L2. In the context of second language acquisition of non-moribund languages where significantly higher access to authentic speech is possible, it seems important that students be given the opportunity to be exposed to that speech as much as possible. The fact that language learning mechanisms appear to be available means that significant exposure to L1 pronunciation can assist learners in acquiring new articulatory representations. Additional exposure to pronunciations guides with ultrasound or MRI would also expediate the process and assist learners with difficulty in acquiring segments.